Idle Speed Motor ***SPLIT from Liner Seals***

[Giamanut]

Just Finished my head gaskets and they work fine! I did have anti go by my liner seals at some point however I don't think they provide much of a seal with nothing holding them down. Anyway before I go to far along I just torqued the heads to 45 LB ft. and with the oil plug out filled the engine with coolant If it was gonna leak it would have came out the oil plug. No coolant getting by the seals so I finished the install and went for a drive! Pretty cool but the 21 year old tires have got to go! LOL
Did have some idle issues but got that on the run. Trouble shot that to IAC motor not closing all the way. Close loop Idle goes to 1500 and FV goes to 136 % to compensate for the air going through the IAC tapping on the IAC reduced Idle but plugging the pipe of agony and setting the butterflies to 750 cured it so I have a new IAC motor on the way.
So far here is what I have working on the car
Mirrors
Clock
Antenna
windows
washer/ wipers
Clutch
Brakes
fuel gauge
speedometer
all lights
cigarette Lighter
Heater
AC
Roof box repair
Lambda System
Radiator
cooling fans
horns
various relays
and Circuit breakers
and of course
Head gaskets and water pump.
that took about 5 weeks with my work schedule
I work 4 tens so about 15 days.
Still need to do a proper AC charge,
change the IAC
1 blinker indicator
Odometer trip reset
Tires
At that point as far as I can tell 5051 should be very close to 100%
Glad to see your doing well with your project also!

[Bitsyncmaster]

Did have some idle issues but got that on the run. Trouble shot that to IAC motor not closing all the way.

FYI:

With the stock idle ECU the idle motor does not close all the way. The amount it can close is dependent on the thermistor resistance. Higher resistance (cold engine) the IAC can not close much under 1/3 open and even with the hot engine it still will not close all the way. That is how the ECU provides a "fast idle" when you first start a cold engine. And the amount of fast idle is very dependent on how you have the adjustments done on the engine. That is also why you get 2000 to 3000 idle RPM if your thermistor gets an open circuit.

[Giamanut]

FYI:

With the stock idle ECU the idle motor does not close all the way. The amount it can close is dependent on the thermistor resistance. Higher resistance (cold engine) the IAC can not close much under 1/3 open and even with the hot engine it still will not close all the way. That is how the ECU provides a "fast idle" when you first start a cold engine. And the amount of fast idle is very dependent on how you have the adjustments done on the engine. That is also why you get 2000 to 3000 idle RPM if your thermistor gets an open circuit.

Thanks. Right the IAC does not seal shut. But mine is sticking too far open in closed loop, I would think the difference in how far it closes at 1500 and 750 Rpm is not a lot. When I tap on it it does close a little further and drops the rpm to about a 1100.
Idle does not hunt or surge. FV duty cycle percentage is way high in closed Loop which tells me the Lambda system is working, just need less air to bring that down. I will post the fix. But so far Fuel Pressure is good Timing is right No external vac leaks.
It does rev good so mixture is close but not right yet WOT switch works. coming up on days off again so should get another look at it soon.

[Giamanut]

+1

Another problem is when your deceleration valves are bad. That extra vacuum leak makes the Idle motor run to it's minimum opening. The idle motor has very little torque at the extreme ends and any sticking will most likely happen then. When I designed my idle ECU I noted the best torque at the center value (FV duty cycle at 50%). What I did to get a little better was to reduce the frequency of the motor control from 100 Hz. to 66 Hz.

So going back to the thermister for a second. What i got from the manual is that under 59 degrees it closes and provides power or ground to a separate circuit in the Idle ECU and just opens above 59 So if it were unplugged it could not activate the High Idle circuit in the ECU. Am I reading that wrong? I know you are the guru on this ECU. Also when I come off the idle switch the IAC stops as it should so that is working If the thermister was stuck closed it would keep the IAC open more than it should be when warm but unplugged it should just kill the High Idle circuit. Correct? I don't know if that is right?

[Bitsyncmaster]

So going back to the thermister for a second. What i got from the manual is that under 59 degrees it closes and provides power or ground to a separate circuit in the Idle ECU and just opens above 59 So if it were unplugged it could not activate the High Idle circuit in the ECU. Am I reading that wrong? I know you are the guru on this ECU. Also when I come off the idle switch the IAC stops as it should so that is working If the thermister was stuck closed it would keep the IAC open more than it should be when warm but unplugged it should just kill the High Idle circuit. Correct? I don't know if that is right?

The thermistor feeding the idle ECU is a variable resistance. From memory it normal runs from about 50 Kohm cold to less than one Kohm hot. From about 50 Kohm (or higher) to 10 Kohms it varies how far the idle motor will close. The idle switch does nothing I can find with the operation of the stock idle ECU. The idle switch turns the ignition advance on if the switch is open on a warm engine. If the switch is closed it turns the advance off on a warm engine. If the engine is cold you do not get advance either way.

With a stock idle ECU the idle motor never shuts off and it does close the idle motor if your above 775 RPM. With the idle motor closed you probably would not feel vibration because of the very small duty cycle but it is still powered.

With my ECU I also close the idle motor (above 1500 RPM) but I power the idle motor down for most of the time. I do power it up for a very short time about every 10 seconds to make sure the idle motor valve stays closed. I do that to save electrical power, less heating of the idle motor and reduce wear on the idle motor bearings.

[Giamanut]

The thermistor feeding the idle ECU is a variable resistance. From memory it normal runs from about 50 Kohm cold to less than one Kohm hot. From about 50 Kohm (or higher) to 10 Kohms it varies how far the idle motor will close. The idle switch does nothing I can find with the operation of the stock idle ECU. The idle switch turns the ignition advance on if the switch is open on a warm engine. If the switch is closed it turns the advance off on a warm engine. If the engine is cold you do not get advance either way.

With a stock idle ECU the idle motor never shuts off and it does close the idle motor if your above 775 RPM. With the idle motor closed you probably would not feel vibration because of the very small duty cycle but it is still powered.

With my ECU I also close the idle motor (above 1500 RPM) but I power the idle motor down for most of the time. I do power it up for a very short time about every 10 seconds to make sure the idle motor valve stays closed. I do that to save electrical power, less heating of the idle motor and reduce wear on the idle motor bearings.

D:05:02 workshop manual. The idle speed control system is only in operation when a microswitch (Fig. 37) (1) is closed by the throttle plates being in a fully closed position. This switch is located on the engine throttle linkage and allows the ECU module to operate. The ECU module senses engine speed from the tachometer and activates the idle speed regulator. The regulator changes the position of the rotary air valve in order to maintain a specified idle speed.
So what am I missing? Does the IAC still get power off idle but the ECU don't?
Also "A thermistor (3), located in the coolant system distribution pipe, will activate another circuit within the ECU module when the coolant temperature is below 15°C (59°)F. This circuit will provide a fast idle condition."
This statement makes it appear that the thermistter only effects the fast idle circuit in the ECU.
I don't get it, I know you know what your talking about but this is out of the manual. It does look as though the Themister is variable as you state. So off idle the motor is powered but the ECU is not????????????????

[Bitsyncmaster]

D:05:02 workshop manual. The idle speed control system is only in operation when a microswitch (Fig. 37) (1) is closed by the throttle plates being in a fully closed position. This switch is located on the engine throttle linkage and allows the ECU module to operate. The ECU module senses engine speed from the tachometer and activates the idle speed regulator. The regulator changes the position of the rotary air valve in order to maintain a specified idle speed.
So what am I missing? Does the IAC still get power off idle but the ECU don't?
Also "A thermistor (3), located in the coolant system distribution pipe, will activate another circuit within the ECU module when the coolant temperature is below 15°C (59°)F. This circuit will provide a fast idle condition."
This statement makes it appear that the thermistter only effects the fast idle circuit in the ECU.
I don't get it, I know you know what your talking about but this is out of the manual. It does look as though the Themister is variable as you state. So off idle the motor is powered but the ECU is not????????????????

The engineers may have wanted it to work like that but it does not. The ECU runs all the time power is on (Key on). The idle motor gets power (12 volts to the center pin) from the ECU when the ECU gets power. The ECU then drives one side of the motor to ground until the duty cycle changes then the other side of the motor is driven to ground. The idle switch does not turn the the duty cycle drivers off. They drive all the time there is power to the ECU.

You can test it by looking at the voltage or the duty cycle at the idle motor. Press the idle switch and release it and see what you get. If you have the idle motor in your hand you will feel it vibrate all the time.

[Giamanut]

Right, keeping this coolant where it belongs is the name of the game.

Yep that is for sure. I do suspect those seals are pretty stout and maybe a little more forgiving than they appear to be. I mean seems like they would be hard to mess up, indeed trying to match paper or metal expansion to steel and aluminum has got to leave a wide margin for error. So when people say use the fattest shims I agree, they will flatten right in to position. Now if you over heat your engine all bets are off!

[Giamanut]

The engineers may have wanted it to work like that but it does not. The ECU runs all the time power is on (Key on). The idle motor gets power (12 volts to the center pin) from the ECU when the ECU gets power. The ECU then drives one side of the motor to ground until the duty cycle changes then the other side of the motor is driven to ground. The idle switch does not turn the the duty cycle drivers off. They drive all the time there is power to the ECU.

You can test it by looking at the voltage or the duty cycle at the idle motor. Press the idle switch and release it and see what you get. If you have the idle motor in your hand you will feel it vibrate all the time.

I am heading over to the car right now with a new small IAC motor will check it all out and post tonight when I get back. Thanks. like I said I don't don't you just getting my head around this is a bit of a task!
Will post back soon!

[Bitsyncmaster]

I am heading over to the car right now with a new small IAC motor will check it all out and post tonight when I get back. Thanks. like I said I don't don't you just getting my head around this is a bit of a task!
Will post back soon!

Yes with the motor in your hand, you will know it never shuts off.

FYI:

The smaller Volvo motor is much better than the OEM because it runs without dead spots and draws half the current. I run it without any bracket holding it and it stays in place just fine.

[Giamanut]

Yes with the motor in your hand, you will know it never shuts off.

FYI:

The smaller Volvo motor is much better than the OEM because it runs without dead spots and draws half the current. I run it without any bracket holding it and it stays in place just fine.

Well alright then the motor does indeed run off idle, Manual appears to be dead wrong.
So my system seem to be working correctly but trying to dial in the mixture is rough.
I have a Craftsman VOM that has a freq percent function just need to replace the battery in it and find the correct spot to hook up. I still have the diagnostic connector on the back of the engine driver side.
I am able to get 750 idle without surge but it seems to rough I think I am lean at this point curb idle screw is backed way out.
Basically I just fattened the mixture until I got black smoke on throttle blip then backed off a touch. It was smoother when rich.
Well new Tires tomorrow couple of screws here and there a little more fine tuning and she should be good to go!

[Ron]

Well alright then the motor does indeed run off idle, Manual appears to be dead wrong.
So my system seem to be working correctly but trying to dial in the mixture is rough.
I have a Craftsman VOM that has a freq percent function just need to replace the battery in it and find the correct spot to hook up. I still have the diagnostic connector on the back of the engine driver side.
I am able to get 750 idle without surge but it seems to rough I think I am lean at this point curb idle screw is backed way out.
Basically I just fattened the mixture until I got black smoke on throttle blip then backed off a touch. It was smoother when rich.
Well new Tires tomorrow couple of screws here and there a little more fine tuning and she should be good to go!
There are two places to hook up shown in the first post of THIS THREAD (http://dmctalk.org/showthread.php?12909-How-To-Set-CO-Dwell).
The last post has a chart showing the values to set/check/trouble shoot with, using Dwell or % DutyCycle.
(Don't use 6 cyl mode!)
Most people prefer using Dwell and an analog meter.

I'm not sure of which screw you are calling a curb idle screw...if you mean the brass screw common to both banks, it is not intended to be used on a DeLorean. (D:04:02, 19 is another place in the manual that can be misleading). All 3 brass screws should be closed.
Note ALL adjustments/settings should be done before setting the CO.

[Giamanut]

There are two places to hook up shown in the first post of THIS THREAD (http://dmctalk.org/showthread.php?12909-How-To-Set-CO-Dwell).
The last post has a chart showing the values to set/check/trouble shoot with, using Dwell or % DutyCycle.
(Don't use 6 cyl mode!)
Most people prefer using Dwell and an analog meter.

I'm not sure of which screw you are calling a curb idle screw...if you mean the brass screw common to both banks, it is not intended to be used on a DeLorean. (D:04:02, 19 is another place in the manual that can be misleading). All 3 brass screws should be closed.
Note ALL adjustments/settings should be done before setting the CO.

Brass screws are closed. The screw below the mircoswitch screw. just the Idle screw.

[Bitsyncmaster]

Well alright then the motor does indeed run off idle, Manual appears to be dead wrong.
So my system seem to be working correctly but trying to dial in the mixture is rough.
I have a Craftsman VOM that has a freq percent function just need to replace the battery in it and find the correct spot to hook up. I still have the diagnostic connector on the back of the engine driver side.
I am able to get 750 idle without surge but it seems to rough I think I am lean at this point curb idle screw is backed way out.
Basically I just fattened the mixture until I got black smoke on throttle blip then backed off a touch. It was smoother when rich.
Well new Tires tomorrow couple of screws here and there a little more fine tuning and she should be good to go!

Most test instruments read percent backwards from an automotive meter. You can test the meter by pressing the wide open throttle. WOT should read 60%, if you read 40% then your reading backwards. So instead of setting 30% you should be setting 70% when setting the mixture with that meter.

[Giamanut]

Most test instruments read percent backwards from an automotive meter. You can test the meter by pressing the wide open throttle. WOT should read 60%, if you read 40% then your reading backwards. So instead of setting 30% you should be setting 70% when setting the mixture with that meter.

No surge 775- 750 Idle Fires right up hot warm or in-between Drives smooth!
So Renault really has the answers it would be nice to have a manual from them.
So here's what I found that cured this
Assuming no Vac leaks and Fuel pressures are good and mixture is close Lamba will comp for mixture to some degree.
1 Take the IAC motor out of the system.
2 close the brass plugs But not the one with a slot.
3 back the slotted brass Idle idle screw out until you get a RPM rise.
4 back the butterfly screw out and the micro out until butterflies are closed! Note: best done with the throttle rod disconnected.
5 screw the butterfly adjustment screw in until it touches the stop then at most another half turn,
6 monitor the vacuum at the vacuum advance and screw the Mirco switch screw in until the vacuum stops plus a half turn.
7 screw the slotted brass screw in until you very close but not above 750 RPM on the nose is best.
8 reinstall IAC motor check mixture.
Note Make sure the throttle rod/ linkage is not stopping the throttle plates/butterfly from closing fully.
This needs to be done an a warm engine.
This is what I did and the only thing I need to adjust for Idle is the brass slotted screw. So adjusting the mixture lean raises Rpm, I drop it back down with that screw.
This is from a Renault Mechanic that specializes in these engines.

[Ron]

What about the unmeasured air this introduces into the system after the metering plate (RE: Warning at D:01:13)...off idle...?

[Giamanut]

What about the unmeasured air this introduces into the system after the metering plate (RE: Warning at D:01:13)...off idle...?

So the balance screws are closed. I am using the air adjusting screw. Don't see any warning at D:01:13. but It is the same as cracking the Butterfly but a lot easier to sweet spot because there no moving parts. Also The Volvo manual does it the same way.
Oh you mean additional air when the throttle is cracked open. It would only come into play at WOT and I am not sure you could even measure it at that point. Volvo and Renault both use this way to adjust. I have P J Grady books I don't see any ill effects at all.

[Ron]

So the balance screws are closed. I am using the air adjusting screw. Don't see any warning at D:01:13. but It is the same as cracking the Butterfly but a lot easier to sweet spot because there no moving parts. Also The Volvo manual does it the same way.
Oh you mean additional air when the throttle is cracked open. It would only come into play at WOT and I am not sure you could even measure it at that point. Volvo and Renault both use this way to adjust. I have P J Grady books I don't see any ill effects at all.No, ...misunderstanding(s).
I think I'm with you now- When you "take the IAC motor out of the system..." you block off its passage, making it ~Basic K-Jet instead of K-Jet with lambda. Then open the screw so it gets enough air to run.... Which means the air is still measured by the plate (but not necessarily the amount the system expects after you set it, similar to the procedure in D:04:01 where the O2 sensor is disconnected....). Yes?

[Giamanut]

No, ...misunderstanding(s).
I think I'm with you now- When you "take the IAC motor out of the system..." you block off its passage, making it ~Basic K-Jet instead of K-Jet with lambda. Then open the screw so it gets enough air to run.... Which means the air is still measured by the plate (but not necessarily the amount the system expects after you set it, similar to the procedure in D:04:01 where the O2 sensor is disconnected....). Yes?

Yes! Get it as close as possible and the IAC barely has to do any work so it can't search to much. I found this thread about some Renault prv guru that explained it and after checking the Volvo manual they where pretty close. I was surprised, it was relatively easy and just brings it down to 2 adjustments, Mixture and what Volvo calls an Idle or air bypass screw. Pretty cool.

[Giamanut]

Ya guy's know I am going to have to sell this car, I just don't have the jack to keep it. But that being said I knew that when I bought it. Some people said sell it to someone who would love it and do a better job bringing it back to life. All I can say is I would drive my D anywhere after the work I have done on it! I am not a hack and this car is a Prime example of the D. On the upside I may have put myself in a position to actually own one that I can keep! Yes as incredible as it may seem I have a friend who has a friend who knows of another D with my name on it! Wish me luck as I am about to put 5051 up for sale. This is not a pitch but 5051 has a gorgeous underside and her topside is very very nice I am thinking of starting mid 40's any thoughts, Also everything on this car works from AC to Clock. I hope this does not violate any forum rules. It is not for sale yet! Couple light bulbs left to sort out!

[DMCVegas]

No, ...misunderstanding(s).
I think I'm with you now- When you "take the IAC motor out of the system..." you block off its passage, making it ~Basic K-Jet instead of K-Jet with lambda. Then open the screw so it gets enough air to run.... Which means the air is still measured by the plate (but not necessarily the amount the system expects after you set it, similar to the procedure in D:04:01 where the O2 sensor is disconnected....). Yes?

All air that is delivered to the Idle Speed Circuit is pulled in after the Fuel Metering plate. The inlet is below the plate. All air is metered and accounted for to maintain AFR. The problem with removing the Idle Speed Motor is that there is no Auxiliary Air Valve to compensate for fast idle conditions.

With the original idle speed system, the K-Jetronic manifold here was designed to use 3 brass screws and a balancing tube. Take a look:

61763

You'll notice that aft of the butterflies on those 90° manifold tubes, there is an additional tube plumbed in which connects both sides. The 3 screws that are present in the manifold for adjusting Idle Speed are first a feeder screw, and the final two being for the left and right cylinder banks. The tube that was included above on this earlier system allows for ΔP to occur in order for the engine to naturally correct the volumes between both cylinder banks for proper emissions. It also connected to the oil filler cap for crankcase ventilation.

Here is the problem with removing the Idle Speed Motor and using those screws to set Idle:

Once you set idle speeds, you're done. You cannot increase idle speeds beyond that point on a warmed engine. Once the advent of A/C became prevalent on cars, the additional load on the engine during idle could stall them. Thus manufacturers resorted to kicker solenoids on the carburetor throttle linkage assemblies to activate whenever the compressor clutch was engaged. But rather than install a kicker solenoid, BOSCH came up with this:

61764

Whenever the engine needs a fast idle condition, that valve slides open and allows additional air to bypass the throttle plates.

K-Jetronic is a modular system that accepts add-ons, and later on BOSCH came out with the updated microprocessor-controlled Idle Speed Motor that we use now. Instead of just blindly allowing air to pass through, it controls the amount through feedback and adjusts to 775± RPMs automatically.

So, if you're opting for utilizing the Idle Speed Screws instead, you need to either never use A/C again, or splice in an AAV to compensate. Even if you retain the stock idle speed control system, you have no way of properly balancing the left and right cylinder banks without measuring CO off of each manifold! The result is one side will be too lean, and the other too rich. Aside from LAMBDA now perpetually fighting your engine, you'll be robbed of power and may do other long-term internal damage to the engine.

LAMBDA also has noting at all to do with the Idle Speed system. LAMBDA only wants to measure O2 content within the exhaust stream. It's adjustment points are within a bypass of the Primary Pressure Regulator on the Fuel Distributor by way of the Frequency Valve to control line pressures feeding the Fuel Injectors. The two systems are completely independent of on another.

Now there is another possibility for a malfunctioning Idle Speed Circuit: wiring. The test is simple though. Just test for continuity. Had this problem with my own car, grafted in a new wire to the ISM, and it worked just fine.

I also readjusted the Microswitch as well. The factory recommends setting the stop screw to only trigger the switch once the throttle plates are back at a fully closed position. Nope, that ain't happening. I lengthened the screw and bent the bracket up to bury that damn screw almost to the bottom without interfering with the throttles fully closing. The result was that the idle circuit triggered early and the RPMs would gently float down instead of rapidly falling. Same as a modern car, and far less voltage spikes from my alternator.

[David T]

Good post. More reasons why you should have an automatic way to maintain a stable idle speed include the fact that there is no damper on the throttle so on the 5-speed, if you let off the throttle quickly, the engine could stall as the butterfly plates snap shut. The idle also must be increased for a cold engine so it doesn't stall. While it may be tempting to remove a complicated sub system, it isn't that hard to get it to work.

[Ron]

All air that is delivered to the Idle Speed Circuit is pulled in after the Fuel Metering plate. The inlet is below the plate. All air is metered and accounted for to maintain AFR. The problem with removing the Idle Speed Motor is that there is no Auxiliary Air Valve to compensate for fast idle conditions.
Yes. That's the way I've understood it for years. FWIW, I was asking because "taking the ISM out of the system" left me wondering if he was replacing it with a pipe or blocking it off or ???... Seemed a pipe the size of the ISM inlet/outlet would let way to much pass, and if blocked off, the amount of air the brass screw allows to pass would be compensated for by the Idle Speed System when reinstalled....
So, if you're opting for utilizing the Idle Speed Screws instead, you need to either never use A/C again, or splice in an AAV to compensate. [I]Even if you retain the stock idle speed control system, you have no way of properly balancing the left and right cylinder banks without measuring CO off of each manifold! The result is one side will be too lean, and the other too rich. Aside from LAMBDA now perpetually fighting your engine, you'll be robbed of power and may do other long-term internal damage to the engine.[/QUOTE]
Not me, for reasons relate to what I sat and everything here. I just didn't have enough info on what was being tried....



I also readjusted the Microswitch as well. The factory recommends setting the stop screw to only trigger the switch once the throttle plates are back at a fully closed position. Nope, that ain't happening. I lengthened the screw and bent the bracket up to bury that damn screw almost to the bottom without interfering with the throttles fully closing.
I bent mine when I first got it. That allowed me to let the butterflies shut then turn the screw just enough to know it was not actually touching (carbs, TBs, whatever, I usually do it that way.) It looks like its broke off flush now. Like you, I have the switch setting heavy. Works for me...

======

+1 on good stuff....keep it going!
..I'm wondering what would happen if we separated WOT and Thermal Throttle Enrichment Switch circuits. Then had the WOT ground the connection for the O2 lead, which would push CO to 87° (27°-37° increase).
Maybe make it do some good holding the pedal to the floor when cold too, lol. (I'll have to think about that part ;-)

[Ron]

Ya guy's know I am going to have to sell this car, I just don't have the jack to keep it. But that being said I knew that when I bought it. Some people said sell it to someone who would love it and do a better job bringing it back to life. All I can say is I would drive my D anywhere after the work I have done on it! I am not a hack and this car is a Prime example of the D. On the upside I may have put myself in a position to actually own one that I can keep! Yes as incredible as it may seem I have a friend who has a friend who knows of another D with my name on it! Wish me luck as I am about to put 5051 up for sale. This is not a pitch but 5051 has a gorgeous underside and her topside is very very nice I am thinking of starting mid 40's any thoughts, Also everything on this car works from AC to Clock. I hope this does not violate any forum rules. It is not for sale yet! Couple light bulbs left to sort out!Sorry for the slow reply, been fighting 2 Generac generators with problems along with the normal weekend BS...love learning/sharing, but I type too slow...

Yeah, cool stuff! -- A few thoughts above.
I'm curious as to how far the brass screw wound up being open? And if the CO dwell was correct when you reinstalled the ISM, first go? I'm going to guess it was.

RE: Forum rules- Since this thread is in the General Discussion section and its your thread, you can say most any thing you want...

[Ron]

Good post. More reasons why you should have an automatic way to maintain a stable idle speed include the fact that there is no damper on the throttle so on the 5-speed, if you let off the throttle quickly, the engine could stall as the butterfly plates snap shut. The idle also must be increased for a cold engine so it doesn't stall. While it may be tempting to remove a complicated sub system, it isn't that hard to get it to work.+1

And I would say that goes for the entire system. But the electrical system is a different story!

[Giamanut]

All air that is delivered to the Idle Speed Circuit is pulled in after the Fuel Metering plate. The inlet is below the plate. All air is metered and accounted for to maintain AFR. The problem with removing the Idle Speed Motor is that there is no Auxiliary Air Valve to compensate for fast idle conditions.

With the original idle speed system, the K-Jetronic manifold here was designed to use 3 brass screws and a balancing tube. Take a look:

61763

You'll notice that aft of the butterflies on those 90° manifold tubes, there is an additional tube plumbed in which connects both sides. The 3 screws that are present in the manifold for adjusting Idle Speed are first a feeder screw, and the final two being for the left and right cylinder banks. The tube that was included above on this earlier system allows for ΔP to occur in order for the engine to naturally correct the volumes between both cylinder banks for proper emissions. It also connected to the oil filler cap for crankcase ventilation.

Here is the problem with removing the Idle Speed Motor and using those screws to set Idle:

Once you set idle speeds, you're done. You cannot increase idle speeds beyond that point on a warmed engine. Once the advent of A/C became prevalent on cars, the additional load on the engine during idle could stall them. Thus manufacturers resorted to kicker solenoids on the carburetor throttle linkage assemblies to activate whenever the compressor clutch was engaged. But rather than install a kicker solenoid, BOSCH came up with this:

61764

Whenever the engine needs a fast idle condition, that valve slides open and allows additional air to bypass the throttle plates.

K-Jetronic is a modular system that accepts add-ons, and later on BOSCH came out with the updated microprocessor-controlled Idle Speed Motor that we use now. Instead of just blindly allowing air to pass through, it controls the amount through feedback and adjusts to 775± RPMs automatically.

So, if you're opting for utilizing the Idle Speed Screws instead, you need to either never use A/C again, or splice in an AAV to compensate. Even if you retain the stock idle speed control system, you have no way of properly balancing the left and right cylinder banks without measuring CO off of each manifold! The result is one side will be too lean, and the other too rich. Aside from LAMBDA now perpetually fighting your engine, you'll be robbed of power and may do other long-term internal damage to the engine.

LAMBDA also has noting at all to do with the Idle Speed system. LAMBDA only wants to measure O2 content within the exhaust stream. It's adjustment points are within a bypass of the Primary Pressure Regulator on the Fuel Distributor by way of the Frequency Valve to control line pressures feeding the Fuel Injectors. The two systems are completely independent of on another.

Now there is another possibility for a malfunctioning Idle Speed Circuit: wiring. The test is simple though. Just test for continuity. Had this problem with my own car, grafted in a new wire to the ISM, and it worked just fine.

I also readjusted the Microswitch as well. The factory recommends setting the stop screw to only trigger the switch once the throttle plates are back at a fully closed position. Nope, that ain't happening. I lengthened the screw and bent the bracket up to bury that damn screw almost to the bottom without interfering with the throttles fully closing. The result was that the idle circuit triggered early and the RPMs would gently float down instead of rapidly falling. Same as a modern car, and far less voltage spikes from my alternator.
I put the IAC back in after I set the IDLE I blocked it off just to get everything close to 750 RPM never was
a permanent removal!

[Giamanut]

Sorry for the slow reply, been fighting 2 Generac generators with problems along with the normal weekend BS...love learning/sharing, but I type too slow...

Yeah, cool stuff! -- A few thoughts above.
I'm curious as to how far the brass screw wound up being open? And if the CO dwell was correct when you reinstalled the ISM, first go? I'm going to guess it was.

RE: Forum rules- Since this thread is in the General Discussion section and its your thread, you can say most any thing you want...

Yes it came out pretty close and ended up the screw was not out but about a turn, so I just closed it to wee what would happen and it just kept Idle at 775 maybe had 10 RPM fluctuation but pretty nice!
my Percentage was bouncing around 45 to 55 but my digital meter is a little wank.

[Giamanut]

Good post. More reasons why you should have an automatic way to maintain a stable idle speed include the fact that there is no damper on the throttle so on the 5-speed, if you let off the throttle quickly, the engine could stall as the butterfly plates snap shut. The idle also must be increased for a cold engine so it doesn't stall. While it may be tempting to remove a complicated sub system, it isn't that hard to get it to work.

I didn't Remove any system just disabled it to set idle close! The IAC is fine I did not amputate it! Just easier setting worm Idle close so when I enabled the IAC it did not have to work much and the only brass screw opened is the Idle screw.
Just to clarify.

[Giamanut]

All air that is delivered to the Idle Speed Circuit is pulled in after the Fuel Metering plate. The inlet is below the plate. All air is metered and accounted for to maintain AFR. The problem with removing the Idle Speed Motor is that there is no Auxiliary Air Valve to compensate for fast idle conditions.

With the original idle speed system, the K-Jetronic manifold here was designed to use 3 brass screws and a balancing tube. Take a look:

61763

You'll notice that aft of the butterflies on those 90° manifold tubes, there is an additional tube plumbed in which connects both sides. The 3 screws that are present in the manifold for adjusting Idle Speed are first a feeder screw, and the final two being for the left and right cylinder banks. The tube that was included above on this earlier system allows for ΔP to occur in order for the engine to naturally correct the volumes between both cylinder banks for proper emissions. It also connected to the oil filler cap for crankcase ventilation.

Here is the problem with removing the Idle Speed Motor and using those screws to set Idle:

Once you set idle speeds, you're done. You cannot increase idle speeds beyond that point on a warmed engine. Once the advent of A/C became prevalent on cars, the additional load on the engine during idle could stall them. Thus manufacturers resorted to kicker solenoids on the carburetor throttle linkage assemblies to activate whenever the compressor clutch was engaged. But rather than install a kicker solenoid, BOSCH came up with this:

61764

Whenever the engine needs a fast idle condition, that valve slides open and allows additional air to bypass the throttle plates.

K-Jetronic is a modular system that accepts add-ons, and later on BOSCH came out with the updated microprocessor-controlled Idle Speed Motor that we use now. Instead of just blindly allowing air to pass through, it controls the amount through feedback and adjusts to 775± RPMs automatically.

So, if you're opting for utilizing the Idle Speed Screws instead, you need to either never use A/C again, or splice in an AAV to compensate. Even if you retain the stock idle speed control system, you have no way of properly balancing the left and right cylinder banks without measuring CO off of each manifold! The result is one side will be too lean, and the other too rich. Aside from LAMBDA now perpetually fighting your engine, you'll be robbed of power and may do other long-term internal damage to the engine.

LAMBDA also has noting at all to do with the Idle Speed system. LAMBDA only wants to measure O2 content within the exhaust stream. It's adjustment points are within a bypass of the Primary Pressure Regulator on the Fuel Distributor by way of the Frequency Valve to control line pressures feeding the Fuel Injectors. The two systems are completely independent of on another.

Now there is another possibility for a malfunctioning Idle Speed Circuit: wiring. The test is simple though. Just test for continuity. Had this problem with my own car, grafted in a new wire to the ISM, and it worked just fine.

I also readjusted the Microswitch as well. The factory recommends setting the stop screw to only trigger the switch once the throttle plates are back at a fully closed position. Nope, that ain't happening. I lengthened the screw and bent the bracket up to bury that damn screw almost to the bottom without interfering with the throttles fully closing. The result was that the idle circuit triggered early and the RPMs would gently float down instead of rapidly falling. Same as a modern car, and far less voltage spikes from my alternator.
So like I said earlier The IAC is still there and working better than ever.
But I respectfully disagree with your first Paragraph. The sense plate Air Metering unit does not account for or even measure air, as it sends no signal to anything. It does allow a constant/Known amount of air to enter the engine for what ever position it is in. So knowing that opened half way it would allow for example would allow 300 cfm, it can be used to allow a correct amount of fuel in. It is a simple air actuated valve that allows a predetermined amount of fuel to flow to the injectors at a given position. The mixture screw is the stop it rest against and simply pushes it down to add more fuel. A/F is controlled by fuel pressure in open loop and Lambda is closed loop. CO setting does mostly nothing off idle. In closed loop CO is Lambda corrected.
IMHO. I could be wrong so I would not put a ton of stock in this post!

[Ron]

...
The sense plate Air Metering unit does not account for or even measure air, as it sends no signal to anything. It does allow a constant/Known amount of air to enter the engine for what ever position it is in. So knowing that opened half way it would allow for example would allow 300 cfm, it can be used to allow a correct amount of fuel in. It is a simple air actuated valve that allows a predetermined amount of fuel to flow to the injectors at a given position. The mixture screw is the stop it rest against and simply pushes it down to add more fuel. A/F is controlled by fuel pressure in open loop and Lambda is closed loop. CO setting does mostly nothing off idle. In closed loop CO is Lambda corrected.
IMHO. I could be wrong so I would not put a ton of stock in this post! I wouldn't say the plate allows air in depending on its position. The amount of air determines the position of the plate, which is pressed in the opposite direction by the WUR/CPR. As you say, the position will determine the amount of fuel sent to the injectors by the FD. Then the Lambda system adjusts according to what it sees after it is burned, if it is in closed loop.

[DMCVegas]

So like I said earlier The IAC is still there and working better than ever.
But I respectfully disagree with your first Paragraph. The sense plate Air Metering unit does not account for or even measure air, as it sends no signal to anything. It does allow a constant/Known amount of air to enter the engine for what ever position it is in. So knowing that opened half way it would allow for example would allow 300 cfm, it can be used to allow a correct amount of fuel in. It is a simple air actuated valve that allows a predetermined amount of fuel to flow to the injectors at a given position. The mixture screw is the stop it rest against and simply pushes it down to add more fuel. A/F is controlled by fuel pressure in open loop and Lambda is closed loop. CO setting does mostly nothing off idle. In closed loop CO is Lambda corrected.
IMHO. I could be wrong so I would not put a ton of stock in this post!

Ron summed this up quickly with this reply:


I wouldn't say the plate allows air in depending on its position. The amount of air determines the position of the plate, which is pressed in the opposite direction by the WUR/CPR. As you say, the position will determine the amount of fuel sent to the injectors by the FD. Then the Lambda system adjusts according to what it sees after it is burned, if it is in closed loop.

I think that right now we have a confusion between metering apparatuses. Specifically the Metering Plate, the Throttle Plates, and the IAC. Whatever the later two request as active meters, the former adjusts to as a passive one. Anyways...

K-Jetronic is a modular system. Keep that in mind.

The way in which it functions is that the fuel distributor is split into two haves by way of a diaphragm. The inlets to the injectors butt up against that diaphragm. The farther that diaphragm pulls away from the inlets, the wider the gap becomes into the individual injector ports becomes, and thus allows more fuel and a higher line pressure on those injectors to become. The higher the pressure, the further the spring loaded valves within the individual injectors open.

61774

As the throttle plates open up further, the CFM of air going into the engine increases, yes. As that airflow increases, it presses against the metering plate, which in turn starts to move the plunger inside of the fuel distributor.

61775

This in turn lowers the fuel pressures in the lower chambers of the fuel distributor, then allows more fuel to flow into the injectors inside of the upper chambers. It does this by bleeding out the pressures in the lower chamber to widen the gap in the upper chambers, and by both feeding that purged fuel into the upper chambers, and dumping excess into the return line to the tank.

What you're doing when you adjust the CO mixture screw is setting the AFR (preferably @ 900± RPMs as per the manual) is setting the baseline. The rest of the entire fuel map has already been dialed in by BOSCH by the port sizes and spring loads in the rest of the components. The rest of the system is purely, 100% mechanical. The Warm Up Regulator has the ability to bypass the Primary Pressure Regulator to reduce lower chamber pressures to enrich the fuel mixtures by dumping fuel back into the return line. It's internal diaphragm valve is activated either by a bi-metalic arm due to cold engine block temperatures, or by sudden huge vacuum spikes in the intake manifold due to sudden RPM accelerations.

Again, all purely mechanical.

LAMBDA is a modular add-on for K-Jetronic. What it does is it fine-tunes the emissions of the engine by monitoring the Oxygen content of the exhaust gasses. It accomplishes this by way of the Frequency Valve. Literally just a giant electronic fuel injector acting as a valve, and does the exact same thing as the warm up regulator: it dumps fuel out of the lower chambers by bypassing the Primary Pressure Regulator. If the mixture is deemed too lean, in increases flow rate to spray more fuel. If the mixture is too rich, it slows the flow rate to increase pressure in the lower chambers of the fuel distributor, in order to reduce how much fuel the injectors spray. LAMBDA has preset values for Open Loop, cold engine, warm engine, and WOT thanks to that little microswitch on the throttle spool. LAMBDA is an electronic version of the WUR, but only for fine-tuning

Now then, having said all of this... All air which is fed to the engine is 100% metered by the initial fuel metering plate. Directly below the metering plate is the inlet for the Idle Speed Circuit. A complete diagram demonstrating this is on page D:01:13 of your Workshop Manual.

[DMCVegas]

Yes. That's the way I've understood it for years. FWIW, I was asking because "taking the ISM out of the system" left me wondering if he was replacing it with a pipe or blocking it off or ???... Seemed a pipe the size of the ISM inlet/outlet would let way to much pass, and if blocked off, the amount of air the brass screw allows to pass would be compensated for by the Idle Speed System when reinstalled....
[I think you will agree? Suppose with your engine's condition, ISM, Oil, ambient temp, fuel, etc, etc., set up per the manual, it idled at 750 RPM. (That's within spec.) If you block the ISM and open the brass screw to 750 RPM, the plate will not change, because the air was metered. The WUR/CPR will not change, so pressure/fuel will not change...Except the O2 sensor will see it go leaner, because more air than the butterflies were originally allowing is getting through. The frequency valve will add fuel causing the RPMs to rise. And the Idle Speed System will bring it back down to 750.
There may be more to it, but bottom line to me is, if you have 750 RPM and the CO is correct with the manual setup and the only real change made is the brass screw is opened, it has to be the ISM compensating for it. In other words, it works!

I've never saw (or noticed) a pic with the extra lines like those in the pic. COOL!
I've read where on some Volvos that, if the balancing screws are shut, the 3rd brass screw does nothing (ie it feeds them only), but I haven't tried it on one.

No, nothing at all changes. In fact, this is what you're supposed to do when properly setting the AFR by way of the CO mixture screw. But more on that in a second. All air is completely metered, and there would be absolutely no deviation in idle speed in such an instance. The only way that we would see the RPMs change on their own is by way of a vacuum leak. That's when the mixture is lean because of the unmetered air coming in due to the leak. LAMBDA enriches the mixture, and the additional fuel then causes a surge in RPMs. Idle Speed Circuit fights the speed increase by choking air off. Now the mixture is too rich. LAMBDA kills enrichment, RPMs plummet. Now the Idle Speed Circuit has to increase air flow to raise RPMs. This causes a lean condition because of the unmetered air, so the mixture is lean because of the unmetered air coming in due to the leak. LAMBDA enriches the mixture, and the additional fuel then causes a surge in RPMs. Idle Speed Circuit fights the speed increase by choking air off. Now the mixture is too rich. LAMBDA kills enrichment, RPMs plummet. Now the Idle Speed Circuit has to increase air flow to raise RPMs. This causes a lean condition because of the unmetered air... Thus we have the surging engine idle condition. You get the picture. In a sealed system, both the butterflies, and the IAC/ISM (whatever we're calling it now) receive their air supply from the mixture chamber, aft of the metering plate.

Now, let's open up that can of worms!

Using dwell to set the CO mixture is *technically* the incorrect way to do things. At least the way that people are currently doing it.

What you want to do is actually set a complete baseline for the engine.

Cooling fans jumped by bypassing the Otterstat in order to achieve & maintain the perfect combustion chamber temperatures with no fluctuation.
Disconnect the O2 sensor to force LAMBDA into Open-Loop.
Disconnect the IAC and use idle screws to set idle speed tp 900±.


There are too many variables with people's O2 sensors, engine temps, and sooted spark plugs since lower RPMs save fuel, but run the engine dirty. You will not get a proper reading, and you don't center LAMBDA and give it enough ceiling to fully lean the engine out. See D:04:01



I bent mine when I first got it. That allowed me to let the butterflies shut then turn the screw just enough to know it was not actually touching (carbs, TBs, whatever, I usually do it that way.) It looks like its broke off flush now. Like you, I have the switch setting heavy. Works for me...

======

+1 on good stuff....keep it going!
..I'm wondering what would happen if we separated WOT and Thermal Throttle Enrichment Switch circuits. Then had the WOT ground the connection for the O2 lead, which would push CO to 87° (27°-37° increase).
Maybe make it do some good holding the pedal to the floor when cold too, lol. (I'll have to think about that part ;-)

Yep. I used to have a nasty problem with the engine rubberbanding when it dipped too low in it's Idle RPMs when coming down from high idle. Occasionally it would even stall out. I spliced in that one bad wire, and switched to the Volvo IAC, and that certainly helped. But once I adjusted the micro switch to activate the idle speed circuit early, the RPMs would just float down.

Never really saw a difference when disconnecting the WOT micro switch on the throttle spool. But again, it's for fine tuning, or at least to keep LAMBDA from fighting the WUR.

I can tell you that if you straightpipe the IAC, the idle speed goes to around 2,400± RPMs...

[DMCVegas]

Good post. More reasons why you should have an automatic way to maintain a stable idle speed include the fact that there is no damper on the throttle so on the 5-speed, if you let off the throttle quickly, the engine could stall as the butterfly plates snap shut. The idle also must be increased for a cold engine so it doesn't stall. While it may be tempting to remove a complicated sub system, it isn't that hard to get it to work.

Thank you. That does mean allot to me.

We've still never met, yet for about 20 years you've been one of my mentors. Particularly about ensuring what I already have works, and learning how to fix it instead of blaming it. After all, if I can't get what I already have working with the instructions to do so, how can I expect to get something that wasn't meant to run actually working any better without troubleshooting skills and understanding of the systems?

[Ron]

No, nothing at all changes....
I agree WRT end result. But, there is a minor change in the ISM position, to make up for the (measured) air the brass screw smuggles into the engine. :deviltail:

100% with you on the rest of the post.

[DMCVegas]

I agree WRT end result. But, there is a minor change in the ISM position, to make up for the (measured) air the brass screw smuggles into the engine. :deviltail:

100% with you on the rest of the post.

You're absolutely correct. If the bypass screws have been opened to increase airflow, the ISM will adjust accordingly by reducing it's airflow to get back to the preset idle speed.

So opening the screws has absolutely no impact on idle speed. But Opening those screws would indeed create a buffer against a rubberband effect on idle speed when the RPMs rapidly decelerate. There seems to be a delay in the activation of the Idle Speed Circuit from when the microswitch is tripped. This creates a delay where the RPMs drop way too low, and the engine can begin the lug, or in some cases such as mine, stall completely. Cracking those screws can allow for a permanent flow of air independent of the ISM, and to a degree stop the rest speed from dropping too low. The problem of course is the potential for the cylinder banks running different AFR mixtures from there on out since you don't have that balancing tube. Not that big of a deal when you're running the engine for tuning purposes. Particularly when you're already analyzing each individual cylinder bank to eliminate this issue during tuning (which we can in fact do without an infrared analyzer for less than $150). Otherwise other than long-term damage, we can end up with another situation where LAMBDA starts chasing the dragon and fights the engine for a pseudo vacuum leak. Albeit one we've created internally with those screws.

Or...we can just readjust that microswitch bracket, and possibly switch to a Volvo 760 GLE that works faster and eliminate the risk of imbalance.

[Giamanut]

I wouldn't say the plate allows air in depending on its position. The amount of air determines the position of the plate, which is pressed in the opposite direction by the WUR/CPR. As you say, the position will determine the amount of fuel sent to the injectors by the FD. Then the Lambda system adjusts according to what it sees after it is burned, if it is in closed loop.

Correct what tried to articulate is that you can know how much air can pass through it is a given position so now you can calculate the amount of fuel needed at that position and develop a way to introduce that fuel The FD and of course now you can determine the best Pressure to send to the FD but you need to compensate for atmospheric conditions so you can't use constant pressure you need to control it, enter the control pressure regulator. Then to fine tune you may need to add a little fuel to get the right air fuel mix FV comes into play there.

Now I know I am gonna get flamed for this but I think would be possible to eliminate the cold start valve by Hi-Jacking the FV just a random thought!
Anyway

[Giamanut]

Totally off the wall question but we have a vacuum line that runs from the epavp can to the cold start tube then another to the Oil filler from there. Why not a straight line to the intake? it just a vac port isn't it? Wait never mind I think if you plugged those port the Cold star valve would not work well. I guess the let air in to carry the fuel charge down the tube.

[Ron]

Correct what tried to articulate is that you can know how much air can pass through it is a given position so now you can calculate the amount of fuel needed at that position and develop a way to introduce that fuel The FD and of course now you can determine the best Pressure to send to the FD but you need to compensate for atmospheric conditions so you can't use constant pressure you need to control it, enter the control pressure regulator. Then to fine tune you may need to add a little fuel to get the right air fuel mix FV comes into play there.

Now I know I am gonna get flamed for this but I think would be possible to eliminate the cold start valve by Hi-Jacking the FV just a random thought!
Anyway
Is it me, or this is paraphrasing post 30 then the last part of post 23, minus a lot of punctuation?
:flamed:


J/K

[Ron]

Totally off the wall question but we have a vacuum line that runs from the epavp can to the cold start tube then another to the Oil filler from there. Why not a straight line to the intake? it just a vac port isn't it? Wait never mind I think if you plugged those port the Cold star valve would not work well. I guess the let air in to carry the fuel charge down the tube.There's not a lot of flow from that side- If you look closely at the ports they connect to, you'll see that the one to the oil fill tube has a tiny metering hole. It's for the crankcase which is sealed. The one coming from the canister is shut off until you break throttle.

[Bitsyncmaster]

Before I found my bad deceleration springs, I tried to plug that hose from the oil filler to get the idle RPM down. What it did was drop the idle 50 to 75 RPM. But running like that, I got some oil pulled up into the air filter so I put it back to stock. Yes that port has a small orifice and the other one in my card is a large hole. So make sure your oil filler hose goes to the port with the orifice.

[DMCVegas]

Totally off the wall question but we have a vacuum line that runs from the epavp can to the cold start tube then another to the Oil filler from there. Why not a straight line to the intake? it just a vac port isn't it? Wait never mind I think if you plugged those port the Cold star valve would not work well. I guess the let air in to carry the fuel charge down the tube.

BINGO!

As mentioned, that hose from the evap canister is for the purge line. It not only purges the vapors collected whilst the vehicle is parked, but it also prevents excess vapor pressures from building up while the vehicle is running. But that is an entirely different topic there...

I *believe* the reason the oil cap connects to the cold start tube is to facilitate air flow so that the evap system and the cold start valve can both feed their respective vapors down into the intake manifold. Part of the crank case vapors are drawn down that tube to provide an airflow that can push the evap & cold start payloads down into the intake manifold.

It's like if you were sucking on a straw that was in a milk bottle, but the bottle was perfectly sealed at the top around the straw. Without a vent hole to replace the displaced fluid inside of the container, the vacuum you create inside the bottle would pull the fluid up and prevent you from sucking out the milk. Thus you've eliminated the flow of the liquid and can't suck anymore out. BUT, poke a hole in the top to allow for airflow, then suddenly the air rushing in will allow you to suck the rest of the milk out because it will push down on the liquid and force it up the straw.

Thus the small diverted flow of crank case gasses from the oil cap helps carry the gasoline mist from the cold start valve and the carbon canister down into the intake manifold. And of course the reason you wouldn't straight pipe that is so that you don't apply full vacuum directly to the cold start valve or the carbon canisters. Just in case you're also wondering about the gasoline vapors being introduced into the manifold as unmetered fuel, never fear, LAMBDA will tweak the AFR accordingly to fine-tune things so that the engine is unaffected.

P.S. If that oil cap gets too restricted, the crank case pressures will build up enough to purge engine oil through the front main seal. Been there, done that. Overall, if someone else can correct me on any of this, please do. But I can assure you from past experience that you want to leave those hoses in place as intended.




Correct what tried to articulate is that you can know how much air can pass through it is a given position so now you can calculate the amount of fuel needed at that position and develop a way to introduce that fuel The FD and of course now you can determine the best Pressure to send to the FD but you need to compensate for atmospheric conditions so you can't use constant pressure you need to control it, enter the control pressure regulator. Then to fine tune you may need to add a little fuel to get the right air fuel mix FV comes into play there.

Now I know I am gonna get flamed for this but I think would be possible to eliminate the cold start valve by Hi-Jacking the FV just a random thought!
Anyway

No need to reinvent the wheel. All fuel metering is already done by the air flow meter in the mixture chamber, Any and all additional minuscule introductions of un-metered air from the crank case vent have already been accounted for by BOSCH. The fuel map which the engine uses was already predetermined at the factory, and the vacuum lines, vacuum delays, hoses sizes, vacuum curves, fuel flow rates, etc. have all been pre-calculated by BOSCH and are executed mechanically by K-Jetronic. It has all been accounted for, which is why there are so many different part numbers unique to the vehicles which received this fuel injection system. No need to reinvent it.

As for deleting the cold start valve and having the injectors take over it's duties, that can't be done. The mechanical system was never designed to have such a flow rate. Now, EFI injectors are, which is why modern cars don't use choke flaps or cold start valves. They're designed for those massive flow rates during cold starts. In order to accomplish such a task, you've have to graft in a second frequency valve, and then wire it to the CSV's old circuit. Then you'd need hoses, injectors, a fuel distributor, and then a new fuel pump. Because remember, this is mechanical fuel injection which requires hydraulic line pressure to activate the injectors. The wider the injectors open, the more line pressure you need to accomplish that. But even then you cannot know what parts to use until you calculate what the necessary line pressures would be, and then adjust the primary pressure regulator accordingly. Then you'd need to ensure that the warm up regulator's flow rates were acceptable, and have to then also calculate the vacuum activation pressures and delays to get that to work in harmony as well. If you were going down this kind of road, it would honestly be much easier to convert the whole engine to EFI than try and reinvent/reassign the functions of K-Jetronic.

Also, please do not think that I am trying to flame you or anything like that. By all means, please do not hesitate to ask questions. It's how I learned.

[Giamanut]

BINGO!

As mentioned, that hose from the evap canister is for the purge line. It not only purges the vapors collected whilst the vehicle is parked, but it also prevents excess vapor pressures from building up while the vehicle is running. But that is an entirely different topic there...

I *believe* the reason the oil cap connects to the cold start tube is to facilitate air flow so that the evap system and the cold start valve can both feed their respective vapors down into the intake manifold. Part of the crank case vapors are drawn down that tube to provide an airflow that can push the evap & cold start payloads down into the intake manifold.

It's like if you were sucking on a straw that was in a milk bottle, but the bottle was perfectly sealed at the top around the straw. Without a vent hole to replace the displaced fluid inside of the container, the vacuum you create inside the bottle would pull the fluid up and prevent you from sucking out the milk. Thus you've eliminated the flow of the liquid and can't suck anymore out. BUT, poke a hole in the top to allow for airflow, then suddenly the air rushing in will allow you to suck the rest of the milk out because it will push down on the liquid and force it up the straw.

Thus the small diverted flow of crank case gasses from the oil cap helps carry the gasoline mist from the cold start valve and the carbon canister down into the intake manifold. And of course the reason you wouldn't straight pipe that is so that you don't apply full vacuum directly to the cold start valve or the carbon canisters. Just in case you're also wondering about the gasoline vapors being introduced into the manifold as unmetered fuel, never fear, LAMBDA will tweak the AFR accordingly to fine-tune things so that the engine is unaffected.

P.S. If that oil cap gets too restricted, the crank case pressures will build up enough to purge engine oil through the front main seal. Been there, done that. Overall, if someone else can correct me on any of this, please do. But I can assure you from past experience that you want to leave those hoses in place as intended.





No need to reinvent the wheel. All fuel metering is already done by the air flow meter in the mixture chamber, Any and all additional minuscule introductions of un-metered air from the crank case vent have already been accounted for by BOSCH. The fuel map which the engine uses was already predetermined at the factory, and the vacuum lines, vacuum delays, hoses sizes, vacuum curves, fuel flow rates, etc. have all been pre-calculated by BOSCH and are executed mechanically by K-Jetronic. It has all been accounted for, which is why there are so many different part numbers unique to the vehicles which received this fuel injection system. No need to reinvent it.

As for deleting the cold start valve and having the injectors take over it's duties, that can't be done. The mechanical system was never designed to have such a flow rate. Now, EFI injectors are, which is why modern cars don't use choke flaps or cold start valves. They're designed for those massive flow rates during cold starts. In order to accomplish such a task, you've have to graft in a second frequency valve, and then wire it to the CSV's old circuit. Then you'd need hoses, injectors, a fuel distributor, and then a new fuel pump. Because remember, this is mechanical fuel injection which requires hydraulic line pressure to activate the injectors. The wider the injectors open, the more line pressure you need to accomplish that. But even then you cannot know what parts to use until you calculate what the necessary line pressures would be, and then adjust the primary pressure regulator accordingly. Then you'd need to ensure that the warm up regulator's flow rates were acceptable, and have to then also calculate the vacuum activation pressures and delays to get that to work in harmony as well. If you were going down this kind of road, it would honestly be much easier to convert the whole engine to EFI than try and reinvent/reassign the functions of K-Jetronic.

Also, please do not think that I am trying to flame you or anything like that. By all means, please do not hesitate to ask questions. It's how I learned.

All true but I am kind of the mad scientist type. JK about the flame part.

[Ron]

BINGO!

As mentioned, that hose from the evap canister is for the purge line. It not only purges the vapors collected whilst the vehicle is parked, but it also prevents excess vapor pressures from building up while the vehicle is running. But that is an entirely different topic there...

I *believe* the reason the oil cap connects to the cold start tube is to facilitate air flow so that the evap system and the cold start valve can both feed their respective vapors down into the intake manifold. Part of the crank case vapors are drawn down that tube to provide an airflow that can push the evap & cold start payloads down into the intake manifold.

It's like if you were sucking on a straw that was in a milk bottle, but the bottle was perfectly sealed at the top around the straw. Without a vent hole to replace the displaced fluid inside of the container, the vacuum you create inside the bottle would pull the fluid up and prevent you from sucking out the milk. Thus you've eliminated the flow of the liquid and can't suck anymore out. BUT, poke a hole in the top to allow for airflow, then suddenly the air rushing in will allow you to suck the rest of the milk out because it will push down on the liquid and force it up the straw.

Thus the small diverted flow of crank case gasses from the oil cap helps carry the gasoline mist from the cold start valve and the carbon canister down into the intake manifold. And of course the reason you wouldn't straight pipe that is so that you don't apply full vacuum directly to the cold start valve or the carbon canisters. Just in case you're also wondering about the gasoline vapors being introduced into the manifold as unmetered fuel, never fear, LAMBDA will tweak the AFR accordingly to fine-tune things so that the engine is unaffected.

P.S. If that oil cap gets too restricted, the crank case pressures will build up enough to purge engine oil through the front main seal. Been there, done that. Overall, if someone else can correct me on any of this, please do. But I can assure you from past experience that you want to leave those hoses in place as intended.
FWIW- I think you meant "was parked".

=======

What about the 3rd hole on the other side??

I believe the ISM will supply plenty of airflow to the cold start tube for the CSV, especially if you keep your foot off of the pedal...*
Excess crankcase vapors will go through the oil fill cap and then to the air cleaner and/or the cold start tube while the engine is running. It has to pass the tiny metering hole and it's dumped during off idle anyway, and as you mentioned, nothing the Lambda System couldn't compensate for.

*This made me think about the many times I have heard (and said!) that it makes no difference if you pump or hold it part throttle when cranking... Now I'm thinking the more the plates are shut, the greater the amount of air flow you will have to deliver the CSV's fuel, I.E., keep you foot off the pedal!
This is why I love decent conversation...

As far as flaming goes, who can say getting their tail feathers singed hasn't saved the thing they are attached to a few times?

P.S. I could be wrong here --- Light me up! :deviltail:

[DMCVegas]

FWIW- I think you meant "was parked".

=======

What about the 3rd hole on the other side??

I believe the ISM will supply plenty of airflow to the cold start tube for the CSV, especially if you keep your foot off of the pedal...*
Excess crankcase vapors will go through the oil fill cap and then to the air cleaner and/or the cold start tube while the engine is running. It has to pass the tiny metering hole and it's dumped during off idle anyway, and as you mentioned, nothing the Lambda System couldn't compensate for.

*This made me think about the many times I have heard (and said!) that it makes no difference if you pump or hold it part throttle when cranking... Now I'm thinking the more the plates are shut, the greater the amount of air flow you will have to deliver the CSV's fuel, I.E., keep you foot off the pedal!
This is why I love decent conversation...

As far as flaming goes, who can say getting their tail feathers singed hasn't saved the thing they are attached to a few times?

P.S. I could be wrong here --- Light me up! :deviltail:

No man, I love conversations like this. It's fun to talk of theories, and likewise just discover new things. I've been wrong before about some things, and I'll be wrong again, but I learn and enjoy the whole process. Particularly about just being nice on the internet. An exceedingly rarer opportunity nowadays.

Which 3rd hole are you talking about?


I don't think that during a cranking condition the ISM passes through as much airflow by itself as we think the engine is getting. I don't know the specific CFM values for this, but I'll lay it down like this as a formula:


x=ISM CFM
y=Throttle plates CFM
z= idle screw CFM
n=Final CFM delivered to the engine.



So lets say that the ISM when fully open delivers x value of CFM. Once the engine is running at proper idle speed, that is when the ISM can then modify the flow rate. Thus the formula is z-x=n because the ISM is subtracting airflow from the engine.

But dig this: Due to the low RPMs present during engine cranking (200±?) until engine speed goes above 775 ± 50 RPM range, then the ISM will not deviate. Thus during engine cranking, the formula becomes x+y=n because the Idle Speed Circuit is using the metric of engine RPM via the pulse signal rather than CFM via an MAF or TPS sensor like an EFI engine has. So when you hold down the gas pedal during cranking, it absolutely makes a difference because the Idle Speed Circuit has no idea what you're doing, and is totally out of range. Thus the engine gets fed a whole lot more air and fuel to start up since as n increases, so doe the fuel flow to the injectors.

I can also tell you as well that it absolutely makes a difference when cranking when you've got other problems with the engine. I once decided to try and "upgrade" my ignition coil with a POS Accel ignition coil. The damn car would not start unless I held down the gas pedal while cranking. Also had problems with my resting fuel pressures thanks to a blast of ethanol ruining my accumulator and eventually the O-rings in the PPR. Holding down the gas pedal during cranking and hard starting helps maintain airflow against the metering plate to keep the plunger pressed in to build line pressures back up quickly.

[Ron]

Which 3rd hole are you talking about?

The one going from the ISM to the Cold Start tube.

(Reading the other now...Is it getting warm in here to you? ;-)

[Giamanut]

The one going from the ISM to the Cold Start tube.

(Reading the other now...Is it getting warm in here to you? ;-)

I can't believe this thread is still going, I think I was just being random in another thread and it got split. Good stuff though!

[Ron]

Robert, to get on the same page- For Z= Idle screw CFM, you're talking the brass bypass screw that is shared by both banks. And when shut no air passes it, or the butterflies when at idle. Correct?
If so, Z-X=N would always be negative...
I'm thinking, it shut, at idle, all air goes into the intake, past the metering plate, out through the Pipe of Agony (Aux 102503), through the ISM, through the Cold Start Tube (Aux 102374), back in, and into the combustion chamber. Lots of airflow to carry the CSV spray.

Or do we need to back up?

[Ron]

I can't believe this thread is still going, I think I was just being random in another thread and it got split. Good stuff though!
+1!

I knew the conversation would take on a life of its own and split it, but who would of thought...

All I can say is, COOL!

[DMCVegas]

Robert, to get on the same page- For Z= Idle screw CFM, you're talking the brass bypass screw that is shared by both banks. And when shut no air passes it, or the butterflies when at idle. Correct?
If so, Z-X=N would always be negative...
I'm thinking, it shut, at idle, all air goes into the intake, past the metering plate, out through the Pipe of Agony (Aux 102503), through the ISM, through the Cold Start Tube (Aux 102374), back in, and into the combustion chamber. Lots of airflow to carry the CSV spray.

Or do we need to back up?

No, you're absolutely right. Both on the function, as well as the formula being incorrect.

I was counting upon the ISM subtracting CFM from it's metering due to the brass screws being opened. Which is true, but still it's still an addition. The values of the variables change, not the operations of the formula.

So let's recap the legend, add to it, and explain the functions.


x=ISM CFM
y=Throttle plates CFM
z= Idle screw CFM (including a sub operation of ratios b=left cylinder bank, c-right cylinder bank)
n=Final CFM of metered delivered to the engine.
f=Metered fuel delivery
v=Vacuum leak
Σ=Final AFR


For for the sake of simplicity, we'll break this down into the individual sub operations first. Let's go back and start with that initial brass screws vs. ISM formula. I was wrong before. What it should be is x+z=n. I'll stick with nice large, whole numbers right now, but let us say for instance that at idle speed the engine requires 20CFM of air to operate. The formula then looks like this: x+z=20. If the brass screws are fully closed, the value of z should be zero, so the equation becomes 20+0=20. But lets say that the screws are in fact opened a bit to allow for a cushion, and we allow a modest 6CFM though. The value of n still has not changed as that is our intended value we wish to obtain, but z becomes 6. So now our equation becomes x+6=20. To solve for x we know that the value is now 14, thus the equation changes to 14+6=20. The decrease in the first value of x is indicative of the ISM reducing it's air flow in order to compensate for the screws being opened.

If by chance we had the throttle plates sticking, we would simply insert y into that equation. Which *technically* it should be there anyway, and it's value should be 0 by default in an idle condition. But if not, just subtract it's value from x like we did prior to acquire it's value. IF by chance the throttle plates are too far open and the CFM exceeds the default idle value, then we're just exceeding 775±50 with the result of a fast idle condition due to stuck throttle plates. Example: 0+0+40=40 using only the ISM, or with those screws opened as per our previous scenario: 0+6+40=46.

Now let's talk about those idle speed screws. If we crack open that initial screw to allow for the aforementioned 6CFM through to the two remaining screws, we have to ensure that the screws are equally opened to flow an exact 3CFM each in order to properly balance each cylinder bank to achieve an equal AFR across all combustion chambers. Otherwise we could be imbalanced anywhere from 0:6/6:0 on the ratios between each cylinder bank. The connecting tube between the 90° manifold tubes on the B27F 2.7L PRV allowed for automatic equalization between the two banks using ΔP. The leaner side could automatically siphon off the additional CFMs from the opposing bank to naturally balance both out. The first idle screw's CFM is directly proportional to the value of z. The only thing that changes is the ratio between b & c that feeds the cylinder banks. z=b/c if I'm not mistaken. Opening one, or both of the remaining screws for the individual banks without opening the first one that feeds both has no direct impact upon anything. With unbalanced cylinder banks, one bank would be more sooty than the other, which that one would burn much hotter. The result would be fowled and burned plugs, potential engine damage, poor economy, and the fun enigmatic problem of puking both excessive hydrocarbons as well and NOx!

So in the end, we should have a formula that looks some thing like this:

(x+y+z=n)f+v=Σ

Now this is the fun part; we're only talking about air flow here. As you can already assume, the value of v should always equal zero! If it doesn't, then we've got some huge problems. On a purely mechanical system we have that unbalanced cylinder bank issue. But once we through LAMBDA into the mix...

f is influenced by many factors and actually has it's own separate formula that gets calculated first to determine it's final value before it is in turn calculated into this one. In the purely mechanical system we have the CSV to intervene during low temps, as well as the WUR...which also affects AFR during hard accelerations when significant engine vacuum is present. But LAMBDA also apes those values, as well as fine-tunes the AFR for optimal efficiently beyond what K-Jetronic itself is able to do alone. That of course hearkens back to the problem of surge idling. LAMBDA is fighting to control the AFR by adjusting fuel flow rates while unmetered air from a vacuum leak influenced by the constantly changing fuel flow rates keeps fighting those corrective attempts by changing engine speeds and sabotaging the AFR.

Two independent closed loop systems generating separate mathematical formulas that inadvertently become at war with one another. Modern cars with fully integrated PCMs would see this problem, go into a default open loop cycle choosing a stabilized condition over an optimal operating one,then just flash a Check Engine light and call it a day. But not these old modular systems.

I know it's intimidating, but it's really not. Just make sure the systems are running as intended and you won't have any problems. This is why I say that switching over to EFI if you're looking to radically change the system is a far better option. You just have to let Tuner Studio perform all of these calculations to achieve the optimal fuel map you want to generate.

It's also why anytime someone tells you, "My car runs so much better once I disconnected the LAMBDA system!" never actually solved anything. They just hit a proverbial mute button and never really fixed anything.

[Ron]

COOL!

Going with the previous assumption that the "initial screw" being shut means z=0 (I believe that is the case, but not 100% sure), I'd suggest the following:
Change z=b/c to z=b+c
Change (x+y+z=n)f+v=Σ to ((x+y+z=n)+v)/f=Σ

And, if we're going to delve deeper, being VERY picky:
Include the time duration (per minute) when defining f and v , so they will properly cancel out in the calculations.
Σ should be simply AFR (or A/F, A:F, for convention]

Not that the changes would affect your comments/conclusions -- I follow you and agree 100%! (Not intimidating...flows right along ;-)

Lots of analogies for "the war": K-Jet Storm...mechanical section brutally bombing the battlefield, in preparation; IAC's ECU commanding the troops at the front; and don't forget Lambda's special ops... There MUST be control conflict's, lol. ...It would be fun to discuss in person, but I hate typing (victim of co-ed PE v/s Typing class scheduling ;-) and OT&OTT.

FWIW: (Cut, pasted, and butchered -- Use at own risk ;-)
61788

[DMCVegas]

NOTE: If you're intimidated by all of this math, or simply don't care about these formulas, that's fine. But please skip down to the last paragraph.

We could go down an entire rabbit hole here with this. Diagram looks correct, save for y should be down on the throttle plates (and there should technically be two of them in case the butterflies are either out of sync or one is leaking, or more specifically y to represent the sum of both butterflies with each having it's on representation within a sub-operation), and where you have y, that should be n to represent the sum of all metered air flowing through.

With the idle speed screws, there is a relation to all 3 screws, which is why I wrote the formula the way that I did (which bear with me, my math is a bit rusty), and even then it's probably oversimplified and incorrect unless we want to really get into the weeds on that one. Include the first screw as a way to represent the initial air flow that must be presented to the left & right cylinder bank screws. If it's zero, then everything is zero. If it's open, unless the other screws are open, it's value remains at zero. Once a screw is open, then the screws will only take as much as they're allowed until their metering exceeds that of what the initial screw can provide. If flow is restricted on one side, then the opposite side would take up the unclaimed CFM so long as the first screw can provide it. Which going back to the previous example, if the first screw was cracked to provide 6 CFM and each screw was set to only accept 3 CFM, then we're balanced. But if we increased the opening on the first screw, then the restrictions on the second screws would still only pull 3 CFM each. Assuming of course that we actually got the screws to be that accurate. Given that BOSCH introduced that balancing tube, I seriously doubt they believed most mechanics could ever have the level of accuracy in balancing cylinder banks. Thus the original system was designed to be self-correcting. But like I say, those latter values are intended to be in relation to the first, and we ought to include the flow rate of that first screw if we really want to nail it down.

Then considering the prior formula, perhaps we should look at that again. The fuel flow rates inside of the fuel distributor may not actually be linear in relation to engine demand. They may only be proportional to air flow instead, which means we might actually have f ∝ n

f itself isn't simply an independent value. That is the entire sum of another formula, whereas both formulas are also influencing and affecting one another. Within A:F we've only looked at the A portion. x, y, & z are all items which we have control over (be it purposeful manipulation or automated control by the on-board systems). v was intentionally kept separate as it is the wild card that cannot be controlled, and is practically incalculable due to the sheer amount of variables involved with it. The value of v should always be irrelevant, because the goal should always be to eliminate it and never to accommodate it. Hence why I look down on just disconnecting LAMBDA to solve an issue, but I digress...

With f it has multiple inputs to define it's sum. Which at just a quick glance would include, but not be limited to:

Fuel pressure from pump.
Line pressure set by the primary pressure regulator
WUR line pressure.
LAMBDA modulation.
Manifold vacuum.
Fuel flow rate.


The list here goes on. But even then we have yet another formula we've not spoken about. On an internal combustion engine, combustion is the key word. The three things fire needs to occur? Air, Fuel, and Heat. A:F is no good without Heat to ignite the fuel mixture. Heat comprises both the cooling systems & ignition systems for not only their proper functions, but how they also affect the individual components of both the metered air n and metered fuel f given that temperature affects the baselines of components within those individual systems such as the ISM & Frequency Valve since Idle & LAMBDA have preset values that are temperature-dependent. So we have 3 formulas, A, F, & H which are all interacting with one another!

Last Paragraph.

Now for anyone intimidated by this, or who just doesn't care about the math, this is for you. While this has been a fun little exercise here, the point is that these formulas have always existed. Believe it or not, whenever you're troubleshooting something you're actually performing algebra. Algebra is the fundamental bedrock of troubleshooting. Now I'm not picking on anyone in particular, let alone any DeLorean owners. But far too often whenever someone has a problem, they want someone to tell them what to replace without performing any troubleshooting. I fully believe that a big part of that problem comes from weak mathematical skills which makes the troubleshooting process far too intimidating. Down below I'm going to give an example of a diagram of a simple flashlight. It contains both a schematic, as well as a pictorial diagram of a flashlight (or torch if you're from the rest of the world). The components of the circuit within the schematic tells you in what order the components would be, and the pictorial shows you what they look like and/or their physical location in the real work. Now, what is the goal of any flashlight? To product light of course! So now that you know what the purpose of this device is, you now know what it's desired sum is. And the components of the circuit become components of the formula. So a functioning flashlight should have a mathematical expression of Charged Battery+Positive Wire+Closed Switch+Second Positive Wire+Light Bulb+Negative Wire=Light Now think about what you'd do if you had a flashlight that didn't work. What would you do to troubleshoot it? And BINGO! You're using algebra to troubleshoot.

61789

[Ron]

Diagram looks correct, save for y should be down on the throttle plates (and there should technically be two of them in case the butterflies are either out of sync or one is leaking, or more specifically y to represent the sum of both butterflies with each having it's on representation within a sub-operation), and where you have y, that should be n to represent the sum of all metered air flowing through.

Yes, I have that wrong.

Anyway, two "y"s would be better, but they should be placed down under (not on) the plates. Lets call the butterfly CFM, b1 & b2.
And n can not go where y is, because it can take from 2 to 4 paths, stock, and up to 6 if all of the screws are open.
To include n, we would have to be split it into n1 & n2 and ignore v, then place each in their respective chamber along with b1 & b2, but immediately before the ports from the CSV tube.
[You didn't comment on notation/definition changes, so I'm continuing with with ((x+y+z=n)+v)/f=Σ, which don't matter because (x+y+z)f+v=Σ is wrong too...]
Note that (x+y+z=n) does not hold because that would be adding the same value (blue arrow flow) twice to z to get n. The diagram is of a stock engine at idle, I.E., the screws and butterflies are closed. So in this case, x=y=n and (x+z)=n and (n+v)/f=Σ.
Fix all that and we have:
In all cases, b1+b2+x+z=n, so (b1+b2+x+v)/f=Σ. z can be broken down into s1 and s2 where z=s1+s2. Giving, (b1+b2+x+s1+s2+v)/f=Σ.
[Makings for a diagnostic tool. But that's as deep into the rabbit hole as I want to go w/o agreeing on notation/definitions. Changing things like p=plate CFM and x to i= ISM CFM would make it easier to follow too.]


With the idle speed screws, there is a relation to all 3 screws, which is why I wrote the formula the way that I did (which bear with me, my math is a bit rusty), and even then it's probably oversimplified and incorrect unless we want to really get into the weeds on that one. Include the first screw as a way to represent the initial air flow that must be presented to the left & right cylinder bank screws. If it's zero, then everything is zero. If it's open, unless the other screws are open, it's value remains at zero. Once a screw is open, then the screws will only take as much as they're allowed until their metering exceeds that of what the initial screw can provide. If flow is restricted on one side, then the opposite side would take up the unclaimed CFM so long as the first screw can provide it. Which going back to the previous example, if the first screw was cracked to provide 6 CFM and each screw was set to only accept 3 CFM, then we're balanced. But if we increased the opening on the first screw, then the restrictions on the second screws would still only pull 3 CFM each. Assuming of course that we actually got the screws to be that accurate. Given that BOSCH introduced that balancing tube, I seriously doubt they believed most mechanics could ever have the level of accuracy in balancing cylinder banks. Thus the original system was designed to be self-correcting. But like I say, those latter values are intended to be in relation to the first, and we ought to include the flow rate of that first screw if we really want to nail it down.
I'm with you on that.


Then considering the prior formula, perhaps we should look at that again. The fuel flow rates inside of the fuel distributor may not actually be linear in relation to engine demand. They may only be proportional to air flow instead, which means we might actually have [B]f ∝ n
...
Depends on exactly what you mean by "engine demand". Assuming, for example, if at idle you stomp it, you and the engine will be demanding full power, it will not get it until time has passed. So the result is not a linear relationship... The fuel rate of the FD depends on the engine demand. But it has to be raised gradually to maintain a combustible AFR. If linear, when you stomped it, the fuel would instantly be delivered at the rate it is when max rpm is met. If you graphed the result as the pedal was pressed, it would be a rising curved line, not a straight one. (All ignoring a slew of things that would change the load...)
Ideally, we want Σ=Final AFR, to be 14.7/1, (f ∝ n). K-Jet...puts up a relatively fair fight...the Lambda system gets it closer.

61799

[Giamanut]

NOTE: If you're intimidated by all of this math, or simply don't care about these formulas, that's fine. But please skip down to the last paragraph.

We could go down an entire rabbit hole here with this. Diagram looks correct, save for y should be down on the throttle plates (and there should technically be two of them in case the butterflies are either out of sync or one is leaking, or more specifically y to represent the sum of both butterflies with each having it's on representation within a sub-operation), and where you have y, that should be n to represent the sum of all metered air flowing through.

With the idle speed screws, there is a relation to all 3 screws, which is why I wrote the formula the way that I did (which bear with me, my math is a bit rusty), and even then it's probably oversimplified and incorrect unless we want to really get into the weeds on that one. Include the first screw as a way to represent the initial air flow that must be presented to the left & right cylinder bank screws. If it's zero, then everything is zero. If it's open, unless the other screws are open, it's value remains at zero. Once a screw is open, then the screws will only take as much as they're allowed until their metering exceeds that of what the initial screw can provide. If flow is restricted on one side, then the opposite side would take up the unclaimed CFM so long as the first screw can provide it. Which going back to the previous example, if the first screw was cracked to provide 6 CFM and each screw was set to only accept 3 CFM, then we're balanced. But if we increased the opening on the first screw, then the restrictions on the second screws would still only pull 3 CFM each. Assuming of course that we actually got the screws to be that accurate. Given that BOSCH introduced that balancing tube, I seriously doubt they believed most mechanics could ever have the level of accuracy in balancing cylinder banks. Thus the original system was designed to be self-correcting. But like I say, those latter values are intended to be in relation to the first, and we ought to include the flow rate of that first screw if we really want to nail it down.

Then considering the prior formula, perhaps we should look at that again. The fuel flow rates inside of the fuel distributor may not actually be linear in relation to engine demand. They may only be proportional to air flow instead, which means we might actually have f ∝ n

f itself isn't simply an independent value. That is the entire sum of another formula, whereas both formulas are also influencing and affecting one another. Within A:F we've only looked at the A portion. x, y, & z are all items which we have control over (be it purposeful manipulation or automated control by the on-board systems). v was intentionally kept separate as it is the wild card that cannot be controlled, and is practically incalculable due to the sheer amount of variables involved with it. The value of v should always be irrelevant, because the goal should always be to eliminate it and never to accommodate it. Hence why I look down on just disconnecting LAMBDA to solve an issue, but I digress...

With f it has multiple inputs to define it's sum. Which at just a quick glance would include, but not be limited to:

Fuel pressure from pump.
Line pressure set by the primary pressure regulator
WUR line pressure.
LAMBDA modulation.
Manifold vacuum.
Fuel flow rate.


The list here goes on. But even then we have yet another formula we've not spoken about. On an internal combustion engine, combustion is the key word. The three things fire needs to occur? Air, Fuel, and Heat. A:F is no good without Heat to ignite the fuel mixture. Heat comprises both the cooling systems & ignition systems for not only their proper functions, but how they also affect the individual components of both the metered air n and metered fuel f given that temperature affects the baselines of components within those individual systems such as the ISM & Frequency Valve since Idle & LAMBDA have preset values that are temperature-dependent. So we have 3 formulas, A, F, & H which are all interacting with one another!

Last Paragraph.

Now for anyone intimidated by this, or who just doesn't care about the math, this is for you. While this has been a fun little exercise here, the point is that these formulas have always existed. Believe it or not, whenever you're troubleshooting something you're actually performing algebra. Algebra is the fundamental bedrock of troubleshooting. Now I'm not picking on anyone in particular, let alone any DeLorean owners. But far too often whenever someone has a problem, they want someone to tell them what to replace without performing any troubleshooting. I fully believe that a big part of that problem comes from weak mathematical skills which makes the troubleshooting process far too intimidating. Down below I'm going to give an example of a diagram of a simple flashlight. It contains both a schematic, as well as a pictorial diagram of a flashlight (or torch if you're from the rest of the world). The components of the circuit within the schematic tells you in what order the components would be, and the pictorial shows you what they look like and/or their physical location in the real work. Now, what is the goal of any flashlight? To product light of course! So now that you know what the purpose of this device is, you now know what it's desired sum is. And the components of the circuit become components of the formula. So a functioning flashlight should have a mathematical expression of Charged Battery+Positive Wire+Closed Switch+Second Positive Wire+Light Bulb+Negative Wire=Light Now think about what you'd do if you had a flashlight that didn't work. What would you do to troubleshoot it? And BINGO! You're using algebra to troubleshoot.

61789



I love math!

[DMCVegas]

Yes, I have that wrong.

Anyway, two "y"s would be better, but they should be placed down under (not on) the plates. Lets call the butterfly CFM, b1 & b2.
And n can not go where y is, because it can take from 2 to 4 paths, stock, and up to 6 if all of the screws are open.
To include n, we would have to be split it into n1 & n2 and ignore v, then place each in their respective chamber along with b1 & b2, but immediately before the ports from the CSV tube.
[You didn't comment on notation/definition changes, so I'm continuing with with ((x+y+z=n)+v)/f=Σ, which don't matter because (x+y+z)f+v=Σ is wrong too...]
Note that (x+y+z=n) does not hold because that would be adding the same value (blue arrow flow) twice to z to get n. The diagram is of a stock engine at idle, I.E., the screws and butterflies are closed. So in this case, x=y=n and (x+z)=n and (n+v)/f=Σ.
Fix all that and we have:
In all cases, b1+b2+x+z=n, so (b1+b2+x+v)/f=Σ. z can be broken down into s1 and s2 where z=s1+s2. Giving, (b1+b2+x+s1+s2+v)/f=Σ.
[Makings for a diagnostic tool. But that's as deep into the rabbit hole as I want to go w/o agreeing on notation/definitions. Changing things like p=plate CFM and x to i= ISM CFM would make it easier to follow too.]


n should be the sum of all air passing through the metering plate to mechanically influence the flow of fuel, regardless if which path it takes to final destination. Add up the sums of the throttle plates, the ISM, and the brass idle screws, and they should always total n. Now when it comes to calculating those individual sums, then that becomes a different story. The sum of the throttle plates/butterflies is always going to be simple addition to add the two together. The idle screws, however, that is a different kettle of fish since we're dealing with proportionality. The ratios of air passing into each cylinder bank via screws 2 & 3 is directly proportional to the constant flow rate of screw 1 which feeds both of them.

In any case, I certainly concur that the naming convention of these components absolutely needs to be sorted out first.

No need for a diagnostic tool. We already know that with the idle speed we have a maximum deviation from the mean of 50 RPMs when LAMBDA is present and active. If we exceed that, then we know there is a problem.



Depends on exactly what you mean by "engine demand". Assuming, for example, if at idle you stomp it, you and the engine will be demanding full power, it will not get it until time has passed. So the result is not a linear relationship... The fuel rate of the FD depends on the engine demand. But it has to be raised gradually to maintain a combustible AFR. If linear, when you stomped it, the fuel would instantly be delivered at the rate it is when max rpm is met. If you graphed the result as the pedal was pressed, it would be a rising curved line, not a straight one. (All ignoring a slew of things that would change the load...)
Ideally, we want Σ=Final AFR, to be 14.7/1, ([B]f ∝ n). K-Jet...puts up a relatively fair fight...the Lambda system gets it closer.

61799


Totally agree with you. But the funny there here is that we're debating two completely separate linear curves!

What I am referencing is the position of the metering plate in reference to the fuel flow rates and line pressures to maintain the desired A:F. A mechanical measurement. What you're referencing is the time in which it takes the fuel distributor to raise the line pressures to achieve those desired flow rates. A chronological measurement to achieve the aforementioned targets!

[crbritt83]

As a mathematics major, I can appreciate an attempt at talking some semblance of real mathematics in the forum.

[Ron]

n should be the sum of all air passing through the metering plate to mechanically influence the flow of fuel, regardless if which path it takes to final destination. Add up the sums of the throttle plates, the ISM, and the brass idle screws, and they should always total n. Now when it comes to calculating those individual sums, then that becomes a different story. The sum of the throttle plates/butterflies is always going to be simple addition to add the two together. The idle screws, however, that is a different kettle of fish since we're dealing with proportionality. The ratios of air passing into each cylinder bank via screws 2 & 3 is directly proportional to the constant flow rate of screw 1 which feeds both of them.
??? I believe that is exactly what I have...I think you missed where I said, "Fix all that and we have:" ...Use the definitions etc in the diagram.
(b1+b2+x+s1+s2+v)/f=Σ.


In any case, I certainly concur that the naming convention of these components absolutely needs to be sorted out first.
How about:
A/F = Air Fuel Ratio
M = Metered Air
B1 = Bank 1 Butterfly; B2 = Bank 2 Butterfly
I = ISM Air
S = Idle Screw; s1 = Bank 1 Balance Screw; s2 = Bank 2 Balance Screw;
F = Fuel; Fd = Fuel Distributor; Fx = Fuel Injector, where x = {1, 2,..., 6}; Fc Cold Start Valve;
V = Vacuum Leaks; Vx = Vacuum Leak #x, where x = {1, 2, ...}

The formula would then be (B1+B2+I+s1+s2+V)/F=A/F

I'll change the diagram once we agree...



No need for a diagnostic tool. We already know that with the idle speed we have a maximum deviation from the mean of 50 RPMs when LAMBDA is present and active. If we exceed that, then we know there is a problem.Agreed, but I meant a tool to identify the source of a problem, E.G., To determine what component(s) are causing a high A/F.
Simplest example might be: (B1+B2+I+s1+s2)/F=15.6/1 => V>0, vacuum leak!


Totally agree with you. But the funny there here is that we're debating two completely separate linear curves!

What I am referencing is the position of the metering plate in reference to the fuel flow rates and line pressures to maintain the desired A:F. A mechanical measurement. What you're referencing is the time in which it takes the fuel distributor to raise the line pressures to achieve those desired flow rates. A chronological measurement to achieve the aforementioned targets!
OK, like I said, I wasn't sure what you meant by "engine demand". I was looking at it as we set the throttle plates' position, which determines the "demand". And the metering plate measures the passing air, as it is (re)positioned by the pressure (while the demand is met. Reaction). I picked the throttle plates because it seemed that you were considering what all was/wasn't linear and/or proportional. And since the function of the FD is to deliver fuel with a mass equal to ~1/14.7 of the mass of the air measured by the metering plate, maintaining a constant AFR, it must be linear. (Of course the WUR's input, control pressure, should be considered....and other things previously mentioned, air density, etc. for K-Jet Basic, add Lambda for Ds.)

To be clear, here I can see the metering plate relationship as linear...and directly proportional.
But, I'm not seeing both curves as linear (for the reasons I mentioned earlier).
(Which, doesn't matter now that we are on the same page...me thinks ;-)

======

JUMP IN, crbritt83!!!

[Ron]

Just read through everything and the NPQ firefighter instructor in me spotted a FWIW:

The three things fire needs to occur? Air, Fuel, and Heat.
The Fire Triangle is obsolete. It has been updated by a Fire Tetrahedron to include "an uninhibited chemical chain reaction (sufficient exothermic reaction energy to produce ignition). :tongue2:

[DMCVegas]

??? I believe that is exactly what I have...I think you missed where I said, "Fix all that and we have:" ...Use the definitions etc in the diagram.
(b1+b2+x+s1+s2+v)/f=Σ.


How about:
A/F = Air Fuel Ratio
M = Metered Air
B1 = Bank 1 Butterfly; B2 = Bank 2 Butterfly
I = ISM Air
S = Idle Screw; s1 = Bank 1 Balance Screw; s2 = Bank 2 Balance Screw;
F = Fuel; Fd = Fuel Distributor; Fx = Fuel Injector, where x = {1, 2,..., 6}; Fc Cold Start Valve;
V = Vacuum Leaks; Vx = Vacuum Leak #x, where x = {1, 2, ...}

The formula would then be (B1+B2+I+s1+s2+V)/F=A/F

I'll change the diagram once we agree...

I can dig it. What I was doing previously was breaking down the individual values of the unique butterflies into a total sum for Throttle. Just as I was also doing for the idle screws. The reason being is that the total is appropriate for an engine with no vacuum leaks when determining the full force of air flow acting upon the metering plate. Once we start chasing imbalances in the cylinder banks, those individual values then become relevant in terms of identifying vacuum leaks versus leakage of those 4 points. Which in the grand scheme of things itself may not actually figure into the actual troubleshooting per se, but it is important to the mathematical theory at work here. Speaking of which...



Agreed, but I meant a tool to identify the source of a problem, E.G., To determine what component(s) are causing a high A/F.
Simplest example might be: (B1+B2+I+s1+s2)/F=15.6/1 => V>0, vacuum leak!

Like I say, unless we have individual MAF sensors on each cylinder bank, it's impossible to determine if these items are leaking by themselves. Modern engines don't even bother with this. But what method modern engines use to monitor for cylinder bank imbalances, we can certainly utilize. https://www.innovatemotorsports.com/products/lc2.php

Using something such as DPI's exhaust with the optional second O2 bung on the passenger side, we can easily have two of these units running in tandem to show us differences in oxygen level variations in the exhaust gasses to determine imbalances and vacuum leaks. Plus I love this thing because it also has an analog output to mimic the stock narrowband sensor to drive the existing LAMBDA controller. Although any old AFR meter setup would do.



OK, like I said, I wasn't sure what you meant by "engine demand". I was looking at it as we set the throttle plates' position, which determines the "demand". And the metering plate measures the passing air, as it is (re)positioned by the pressure (while the demand is met. Reaction). I picked the throttle plates because it seemed that you were considering what all was/wasn't linear and/or proportional. And since the function of the FD is to deliver fuel with a mass equal to ~1/14.7 of the mass of the air measured by the metering plate, maintaining a constant AFR, it must be linear. (Of course the WUR's input, control pressure, should be considered....and other things previously mentioned, air density, etc. for K-Jet Basic, add Lambda for Ds.)

To be clear, here I can see the metering plate relationship as linear...and directly proportional.
But, I'm not seeing both curves as linear (for the reasons I mentioned earlier).
(Which, doesn't matter now that we are on the same page...me thinks ;-)

======

JUMP IN, crbritt83!!!




Yup, we're almost on the same page. What I was referencing wasn't the FD's ability to maintain the 14.7:1 but rather the internal line pressures to the injectors themselves. Which have an operating rage of 54-80 PSI if I remember correctly. The pressures themselves I don't believe are linear, but absolutely the ability to feed the engine is.

[DMCVegas]

Just read through everything and the NPQ firefighter instructor in me spotted a FWIW:

The Fire Triangle is obsolete. It has been updated by a Fire Tetrahedron to include "an uninhibited chemical chain reaction (sufficient exothermic reaction energy to produce ignition). :tongue2:

UGH! I despise the tetrahedron model!

If the values of the individual components are sufficient enough to induce oxidation, that in and of itself should be considered the chain reaction which completes the triangle to cause a fire. Otherwise, you ain't got no fire! The tetrahedron model completely ignores the individual values of the contributing components and acts as if the chain reaction phase is itself some sort of magical primer that initiates the entire thing, instead of actually being the end product of fire. Likewise extinguishing any blaze isn't simply stopping the chain reaction by itself, but removing one of the three components by removing one or more of them from the chemical reaction.

I feel like the tetrahedron model was nothing more than trying to make fire safety more "hip".

[Ron]

I can dig it. What I was doing previously was breaking down the individual values of the unique butterflies into a total sum for Throttle. Just as I was also doing for the idle screws. The reason being is that the total is appropriate for an engine with no vacuum leaks when determining the full force of air flow acting upon the metering plate. Once we start chasing imbalances in the cylinder banks, those individual values then become relevant in terms of identifying vacuum leaks versus leakage of those 4 points. Which in the grand scheme of things itself may not actually figure into the actual troubleshooting per se, but it is important to the mathematical theory at work here. Speaking of which...

Like I say, unless we have individual MAF sensors on each cylinder bank, it's impossible to determine if these items are leaking by themselves. Modern engines don't even bother with this. But what method modern engines use to monitor for cylinder bank imbalances, we can certainly utilize. https://www.innovatemotorsports.com/products/lc2.php

Using something such as DPI's exhaust with the optional second O2 bung on the passenger side, we can easily have two of these units running in tandem to show us differences in oxygen level variations in the exhaust gasses to determine imbalances and vacuum leaks. Plus I love this thing because it also has an analog output to mimic the stock narrowband sensor to drive the existing LAMBDA controller. Although any old AFR meter setup would do.




Yup, we're almost on the same page. What I was referencing wasn't the FD's ability to maintain the 14.7:1 but rather the internal line pressures to the injectors themselves. Which have an operating rage of 54-80 PSI if I remember correctly. The pressures themselves I don't believe are linear, but absolutely the ability to feed the engine is.
As promised:
61865

I can agreed on all...assuming you agree we do have enough already to determine if there is a vacuum leak w/o individual MAP sensors, just not where it is coming from (can't be the areas defined so far, since they are measured). Maybe we need to make some more assumptions, for now, to avoid the can(s) of worms. Expanding with the same example, a perfect stock engine with a perfect setup could have a high A/F caused by a fuel system leak getting into the air intake....endless rabbit holes going to China.

[Ron]

UGH! I despise the tetrahedron model!

If the values of the individual components are sufficient enough to induce oxidation, that in and of itself should be considered the chain reaction which completes the triangle to cause a fire. Otherwise, you ain't got no fire! The tetrahedron model completely ignores the individual values of the contributing components and acts as if the chain reaction phase is itself some sort of magical primer that initiates the entire thing, instead of actually being the end product of fire. Likewise extinguishing any blaze isn't simply stopping the chain reaction by itself, but removing one of the three components by removing one or more of them from the chemical reaction.

I feel like the tetrahedron model was nothing more than trying to make fire safety more "hip".LOL...I've heard that before and I see where you are coming from. If you view fire as a thing, at a given moment in time, rather than as an event, it makes sense.

Another way: The triangle 'theory' states that given the three things in sufficient quantities, you WILL have fire. If you add halon, for example, there will be no fire although the three are still present in the same quantities. Contradiction.

[DMCVegas]

As promised:
61865

I can agreed on all...assuming you agree we do have enough already to determine if there is a vacuum leak w/o individual MAP sensors, just not where it is coming from (can't be the areas defined so far, since they are measured). Maybe we need to make some more assumptions, for now, to avoid the can(s) of worms. Expanding with the same example, a perfect stock engine with a perfect setup could have a high A/F caused by a fuel system leak getting into the air intake....endless rabbit holes going to China.

I agree. Looking at the engine as-is in pure stock form, we have the visual error of a hunting idle speed that deviates more than 50 RPMs from the mean, which manifests itself as continued severe looping in speeds.

Absolutely we could have a high A/F ratio due to a malfunctioning fuel delivery system that could cause this same symptom. If it were an external vacuum leak, then we could proceed with the traditional smoke test, or just go right for the throat and spritz starting fluid or carb cleaner around various parts of the intake system to observe a jump in RPMs. Wherever we spray and get the surge, that's where we've found our leak.

Now if it were an issue of an unmetered fuel system leak, we'd have both the idle issue, as well as stumbling. It is at that point we might start with leak down tests, as well as visual inspections and mechanical testing of the system as a whole, and the individual components. Spray patterns, popping pressures, injector flow rates, distributor flow rates, etc.

Tests that may even need to be performed with modern OBD-II EFI systems. The only difference here is we're observing physical symptoms as the warnings, and not the glow of an idiot light.

[DMCVegas]

LOL...I've heard that before and I see where you are coming from. If you view fire as a thing, at a given moment in time, rather than as an event, it makes sense.

Another way: The triangle 'theory' states that given the three things in sufficient quantities, you WILL have fire. If you add halon, for example, there will be no fire although the three are still present in the same quantities. Contradiction.

It is matter of perception though, I suppose. The triangle is the rule of thumb, but in order to complete a triangle geometrically, no matter which type, all three sides must be touching. Much akin to how dropping a lit match into a full jerrycan will simply extinguish the match since the fuel smothers it, but a jerrycan full of fumes will result in ignition. In the prior example, not all 3 sides touch as the heat is disbursed and the flame is smothered. In the latter, all 3 sides touch at once, and fire.

I also understand how use of the chain reaction within the tetrahedron model helps to also explain smoldering. But still, even then I don't believe it to be apropos as it also leaves out other factors. Granted we can even challenge the validity of the fire triangle at this point as well. In the end, I just feel that the 4th side obfuscates the safety factor for the layperson. If people could be trusted to better understand fire, or matters of safety in general, we wouldn't have to teach people mantras or jingles to keep them safe.

[David T]

Things like the fire triangle and such are useful simplifications to get a quick handle on a situation. Like ABC for first responder EMT's. Easy to remember like the food pyramid. Not completely accurate but covers most situations. As for "hunting", with several control systems that are not talking to each other except through the thing they actually control, they each have their own hysteresis causing the idle to go up and down as they interact with each other. That is why there is the + - 50 RPM. Add vacuum leaks and that reduces the range (power) the idle system has over control of the system. Now add worn tired parts and bad electrical connections and then the systems lose what little control they had.