mixture setting/rich smell?

[rjd2]

Hi all-first off, a MASSIVE thank you to the good people on this forum. You've helped me on numerous occasions; this is an invaluable resource for DMC-12 owners.

I recently completed a thermister replacement job, and now have the car back sorted-no fuel leaks, smooth idle, everything seems to be working properly. The last goal I had before buttoning it all up was to make sure the mixture was set properly. I sealed off the "cabin" side of the vacuum hose at the oil filler cap location, and hooked up my dwell meter. Here's what I get:

1-idles solidly at ~750-800 or so.
2-starts immediately, holds with no goose-ing of throttle.
3-dwell meter holds at ~42-45
4-when the "cold start" phase hands off to the normal hot run phase, I get some idle hunting, one or two WILD swings, and then it stabilizes right around 42-ish. With a hot engine, I have yet to see any idle hunting at all.

I can't seem to shake the feeling it's still running rich, mainly from the smell. I don't expect the car to smell like a 2017 Accord, but the richness smell is intense. If I do a 5 minute test routine, my clothes will smell for the rest of the day. If this is par for the course, then no problem. Should I aim to tune it leaner even if my readings seem to look ideal? Thanks for the advice!

Oh, and to all the folks down south, good luck, and stay safe...

[Bitsyncmaster]

Your dwell setting should be less than the dwell you read on the cold start. Starting the cold engine the system sets the mixture rich so the cold engine will run.

So your dwell on the hot engine should be set to 30 to 35 degrees.

[NckT]

You really need to start with fuel pressure diagnostics first to ensure that the Control Pressure Regulator (CPR/ WUR) is operating within specification, and that the vacuum pipes to the CPR operate correctly from the thermo-vac switch. (Assuming that the cold start valve is not permanently operating. Try removing the blue plug from this cold start valve)

[David T]

It will run (and smell) rich when cold and till it gets into "Closed Loop". You will know it is in closed loop because you will see the dwell meter start swinging. In open loop the dwell is a steady reading. Once the dwell starts swinging (easy to see on an old, analog dwell meter) set it to between 35 and 45. If it doesn't "swing" the WOT switch could be stuck, the O2 sensor may be bad or disconnected, or the frequency valve is not working. Before trying to adjust the mixture screw you must make sure EVERYTHING, including fuel pressure, valve adjustments, injectors, spark plugs, all of it, is set perfectly. You cannot "adjust" the motor to run right if something is not right. If the tamper proof plug is still in the mixture unit you are better off just leaving that adjustment alone. It is VERY sensitive. A tiny turn can throw it way off and it is probably closer to where it should be than if you try adjusting it.

[NckT]

It will run (and smell) rich when cold. .....

It shouldn't do if set correctly.

[David T]

The catalytic convertor and the O2 sensor are not effective till they warm up and til they warm up you will get more smells. That being said you should not run the motor in a garage where the exhaust can accumulate. Run it outside and it should be less noticeable. If it is bad maybe you have exhaust leaks and the exhaust is getting into the cabin?

[NckT]

.......

I can't seem to shake the feeling it's still running rich, mainly from the smell. I don't expect the car to smell like a 2017 Accord, but the richness smell is intense. If I do a 5 minute test routine, my clothes will smell for the rest of the day. If this is par for the course, then no problem. Should I aim to tune it leaner even if my readings seem to look ideal? Thanks for the advice!

..........

David, please stop quoting the manual of operation with respect to the dwell cycle etc and read what the original Op has said what is wrong. This very rich smelling condition is due to incorrect fuel pressures or very less likely mixture adjustment, especially considering that the amount of unburnt fuel coming out of the exhaust is enough to make the clothes smell. Item 4 description of symptoms is clearly when the lambda switch open circuits and the fuel ecu references the oxygen sensor and the wild swings described is the result of the o2 feedback to lean out the over fuelling condition. This is why I advised to check the fuel pressures first.

I'm not into the childish immature 'last post wins' mentality either so I'll only respond to the rjd2's original problem.

[FABombjoy]

If cold dwell == warm dwell - especially with no fluctuation when warm - the WOT switch could also be stuck down. Or the thermistor is shorted, wire pinched, etc.

You could quickly test for this permanent enrichment condition using a multimeter to check resistance of the WOT switch to ground. The light green wire should read continuity to ground when cold (coolant is 59F or lower), but should not read continuity to ground when warm.

[David T]

From reading his first post it sounds like he is in open loop. He needs to figure out why it doesn't get into closed loop. It could be fuel pressure, it could be a lot of things. I gave him some suggestions to start looking. I didn't suggest checking fuel pressure because he had no other symptoms to suggest that.

[MikeWard]

Running in open loop does not cause strong fuel smell.

I agree with Nick, fuel pressures should be checked to confirm the CPR and PPR are functioning correctly.

I had a similar issue on my car which turned out to be my CPR was out of spec, and caused a strong smell of fuel which burnt your eyes if you were anywhere near it. Nick very kindly recalibrated the CPR for me and that resolved my issue.

[FABombjoy]

FHe needs to figure out why it doesn't get into closed loop. It could be fuel pressure, it could be a lot of things.
There are only a couple of reasons to be in open loop mode:
1: Bad/disconnected O2 sensor
2: Cold engine
3: WOT switch engaged.

2 & 3 both relate to the same wire being grounded

If you're crazy rich the dwell meter will fall to nearly nothing as the system attempts to lean out the mixture
If you're crazy lean the dwell meter will spike high

AFR on a warm engine stuck in enrichment mode will dip as low as 10:1

If the dwell is fixed in the middle and doesn't vary the system is electrically stuck in open loop. I wouldn't chase down a fuel pressure solution until the electrical component has been ruled out.

Disconnect the WOT switch and see if the dwell starts to vary (WOT switch stuck). If not, see if the light green wire on the connector has continuity to ground (thermistor stuck grounded).

[MikeWard]

Read the original post, he states it's dwell is steady between 42 and 45 which is bang on where it should be.

If it's swinging like this then it's fine. Mine was a tad high here as it was going up to 47, but it's in the ballpark, and doesn't smell of fuel either...
https://youtu.be/DB5pSdGc2fM

Fuel pressure test is the way to go, if I was a betting man my money is on out of spec CPR.

[FABombjoy]

Read the original post, he states it's dwell is steady between 42 and 45 which is bang on where it should be.
If the exhaust is eye-watering rich it should be attempting to lean the mixture and would not be anywhere near 45.

[NckT]

If the exhaust is eye-watering rich it should be attempting to lean the mixture and would not be anywhere near 45.

Not when the engine is cold due to the lambda switch or if the oxygen sensor is not hot enough to give signal feedback

[NckT]

http://dmctalk.org/showthread.php?15567-Proofing-my-work-after-fuel-distribution-re-assembly&p=224064#post224064

Rjd2, is your cold start fuel pressure still as per your video on the above post on another thread?

If it is, the control pressure should not additionally drop when you rev it. The vac delay valve should keep the control pressure lower for between 8 to 12 seconds.
Disconnect the vacuum pipes to the cpr and block the river/ silicon pipes that have been removed. Also disconnect the grey electrical plug off the CPR

You'll have to include more information on the cold start (engine left overnight), such as the temperature of the control pressure regulator (CPR / WUR).

[FABombjoy]

Not when the engine is cold due to the lambda switch or if the oxygen sensor is not hot enough to give signal feedback
O2 should be active within the 5 minute window given by the OP. Lambda switch should be off well before 2 minutes.

Some reference data:

Cold start (25f morning, -4c)
Orange trace is duty cycle (dwell-type spec * 10%)

Difference here is that I'm running a WBO2 controller which will have an active signal before the stock 1 wire, hence the immediate reaction shown in chart #2

Thermistor transition out of cold start visible. Jetronic computer immediately attempts to lean a very rich mixture. DC goes from 63% to 15%
53353

CPR transitioning toward warm pressure range. Mixture leans and frequency valve is now within operating range. DC starts to swing.
53355

Warmed up, DC fluctuation is now minimal. Total time under 5 minutes. RPM is high as I was adjusting the throttles / idle switch.
53354

[NckT]

Thank you for your data. This does not equate to the fact that the cold start is very rich to the point of saturating hydrocarbons on clothes. .. within that 5 minute window as you say.

I've lost count the number of CPR's I've overhauled and re calibrated because of this issue, as well as fuel diagnostics.

I'll let you help the rjd2 out, no problem.

[FABombjoy]

I'll let you help the rjd2 out, no problem.
That is the goal, right? To be helpful?

All I am suggesting is to test for an electrical short before proceeding to fuel system tests.

rjd2's issue mimics my own experience with a stuck WOT switch, as has also been experienced by others on 'talk. First post indicates he was just working with the thermistor circuit which is directly involved.

[Bitsyncmaster]

I also run a wideband unit so I go closed loop in about 15 seconds after engine start (wideband is a heated sensor).

But even with that quick controlled mixture, sometimes if the wind is just right blowing into the garage I do have that rich mixture smell right after starting. I remember running a tail pipe wideband test and the mixture does start out in the 10 AFR range (i think).

[rjd2]

Ok, a bit more data:

-I have a BRAND new CPR unit in place as of January 2017. I'm not sure how to test it, but for the first half of the year, it showed no problems.
-I can tune the dwell to hang right around 40, and it won't oscillate at all. When I tuned it any leaner, what I got is the idle jumping up to 1500-1700; it would skip any "warm up" period and go straight to HIGH. I tuned it back rich, to where the idle seemed solid, and left it.

For now, I've got the car able to hang right at ~750ish RPM at idle, and a dwell around 40. I haven't put it through any longer test periods than 5-10 minutes; I'll get some regular usage in, some longer run periods, and see how it goes. (as an aside, at this point I believe the O2 sensor is the only thing in the fuel system that HASN'T been replaced.) Thanks for the help everybody!

[David T]

You test the CPR by observing the fuel pressure as the motor warms up. If the dwell does not fluctuate you are NOT in "closed loop". It takes more than 5-10 minutes to get into closed loop depending on the ambient temperature and if you are just idling or driving the car. The dwell should be moving up and down about 10 degrees in closed loop.