3.0L MegaSquirt w/ EDIS, starts and idles... what next?

[dn010]

This is where the cylinder comes in handy as you can set the injector up to spray with the fuel pump running and know exactly what the output is. You can then set your injector voltage table. I’m sure others didn’t have a problem in that area but I did, and maybe it’s because I am running MAF based, but my car did not like the data I gave it until I tested the injectors. I am using pink top Bosch 440s.


Sent from my iPhone using Tapatalk

[jangell]

EDIS needs to know where the crankshaft / Top Dead Center is in order to work. You can't weld on a trigger ring to the pulley anywhere you feel like it and put a sensor on the engine anywhere you want. The ring needs to be arranged so that the missing tooth is 6 teeth away from the sensor when the crank is at TDC. [https://www.diyautotune.com/support/...rank-trigger/]

Right, I've been over those docs quite a bit in the last two days, and the EDIS docs in the online MegalManual.

The VR sensor, missing tooth gear, main pulley and VR sensor bracket are all things I bought from Josh, which saved me the trouble of actually getting everything mounted and aligned myself. I'm trusting that his work is good, since he had it on a running engine before he upgrade to his LS1. While I was aware of the specific requirements for how the gear was attached to the pulley and the placement of the VR sensor, I didn't think about them too much because I already had Josh's hardware. But I do know I need to get the VR sensor aligned properly so the EDIS has a signal that correctly aligns with TDC, as described here.

But since we are DIY, this arrangement may not be perfect and we may have installed the ring and sensor a few degrees off. This is why you need to check the timing and see what it is. You can change the timing offset in MS, but you can't change anything inside the EDIS box, it is expecting the missing tooth angle to be correct.

It also occurs to me that that Josh may have used a timing offset that I'm not aware of to compensate for the mounting position, in addition to what I think is a 120 degree offset and a BAC firing order to allow for an alternate mounting position for the VR sensor.

Speaking of which, this is the best picture I have of my VR sensor placement (again, from Josh's known working setup). I can't get a better picture than that now that the muffler is installed.

IMG_0299.jpg

The reason for my confusion is that the EDIS docs in the online MegaManual (from my post yesterday) suggest that I need to do VR sensor adjustment on a running engine, and that this is how I would fix the timing associated with EDIS. I think what you and the MS3 docs are suggesting is an alternate way to compensate for VR sensor misalignment through TunerStudio in software, by testing the timing on a cranking engine instead of requiring a running one.

At least I think these both accomplish the same task, right? I just want to understand what I'm doing.

You've stated already you're not very familiar with EDIS. You'd really benefit from looking up MS & EDIS information so you know what is going when your engine is running - makes it easier to diagnose and tune, etc.

Agreed.

-- Joe

[FABombjoy]

Actually, once I connected the MAP I saw the AFR go pretty crazy still, up over 22 at one point before it stalled out. I'm guessing other factors are at player here.

That's to be expected. Doing that shifted you out of the WOT section and into something leaner.

It looks like you just need more fuel. Once you've double-checked that timing is good I would richen the whole VE table until it runs and gets up to temp. Once hot, get your VE table sane and tune warmup enrichment the next day. While hot you can quickly test your IAC valve by tweaking the warm # of steps and ensuring RPM reliably follows suit. Or use the IAC testmode and visually check to see it move.

I wouldn't spend much time worrying about injector characterization at this stage. You can tweak those values, tune dead times by using the changing squirts trick, or maybe find a table online. I spent a fair amount of money having my 30lb injectors characterized but the voltage curve isn't much different from default and I still ended up tweaking the base dead time anyway. Knowing your precise fuel pressure and exact injector flow rate is less important than getting it to run with reasonably AFR readings and monitoring injector duty cycle to ensure you aren't close. Just watch your instruments and listen.

When I first finished my turbo 2.8 I just got it running on a big 'ole fatty-fat fuel table, tore around the countryside for 30 minutes leaning it out, and promptly drove 600 miles while lightly abusing autotune. Once it's started and runs to temp I highly recommend just diving in

[Josh]

Im a little late to the party.

Since the package got lost in the mail, i had to substitute in larger injectors. I made the original tune with 19# injectors, the ones you have a 22#. Some changes will need to be made.

It was not clear to me if the firing order issue was rectified. I know I had a hard time with this. I do not recall what i settled on to confirm your findings. I do know it was different than any information given the Mega manual. You may be onto something with my crank trigger positioning.

There is a lot of adjustment in the VR sensor bracket, no need to correct it on the software side (I do not think this is even possible as the VR sensor is fully interpreted by the EDIS module). To confirm the alignment of the sensor you must do two things:
-Unplug the edis unit from the MS and start the engine. It should run, and you should find it is giving a solid 10deg with your timing light. This is its error state when it cannot communicate with the computer.
-Position the crank sensor as close to the trigger wheel as possible. Pull the fuel pump relay and power to the coil and turn over the engine with your DMM probes across the terminals of the crank sensor. You will be looking for an AC voltage of around 3-4V. Anything less suggest the sensor is too far away and will be getting a weak signal.

[dn010]

While you can change offset in TS with EDIS, here is a little more in depth reason as to why it's not proper-

The EDIS module will only change timing if it receives an order from the MS unit to do so. I stated earlier that my timing was off by 3 degrees. I could change the offset in TS to show 3 degree offset and my timing would be solid where I need it to be. BUT, if the EDIS module ever loses that signal with the MS unit, the module will default to 10 degrees timing, expecting my VR sensor is in the correct position. Meanwhile, my engine would be running at 7 degrees because it's no longer getting that offset message from MS.

So, you really need to make sure you have that base timing correct with the MS unplugged. Then you can see if you need to adjust your sensor and get where you need to be. THEN when it is timed properly while unplugged, plug it back in and command the MS unit to run at a fixed timing, confirm the timing is correct with your light and you know your MS is telling the EDIS the proper timing and without MS it will be 10 degrees.

[opethmike]

Also, have you tried hugging the MS ECU? It may be lonely, and in need of reassurance.

[jangell]

That's to be expected. Doing that shifted you out of the WOT section and into something leaner.

It looks like you just need more fuel. Once you've double-checked that timing is good I would richen the whole VE table until it runs and gets up to temp. Once hot, get your VE table sane and tune warmup enrichment the next day. While hot you can quickly test your IAC valve by tweaking the warm # of steps and ensuring RPM reliably follows suit. Or use the IAC testmode and visually check to see it move.

Ok, cool. I might pop the IAC off and visually check it, since that'll be easy enough to do. I set up that VE table a a couple of years ago now, so I'll have to re-read the docs and make sure I richen it correctly. I'm guessing "sane" is "the car idles at 775 RPM properly"?

I wouldn't spend much time worrying about injector characterization at this stage. You can tweak those values, tune dead times by using the changing squirts trick, or maybe find a table online. I spent a fair amount of money having my 30lb injectors characterized but the voltage curve isn't much different from default and I still ended up tweaking the base dead time anyway. Knowing your precise fuel pressure and exact injector flow rate is less important than getting it to run with reasonably AFR readings and monitoring injector duty cycle to ensure you aren't close. Just watch your instruments and listen.

Good to know. I'll still adjust for 21 lb/hr, since that fits with what I found online and the actual fuel pressure I'm seeing, and it's a simple enough change. I mostly want it "good enough". I mean, I have an automatic; I'm not going to be going to races in this thing, and I'd probably tweak for fuel economy over power anyway (although I realize that having everything dialed in properly will improve both).

When I first finished my turbo 2.8 I just got it running on a big 'ole fatty-fat fuel table, tore around the countryside for 30 minutes leaning it out, and promptly drove 600 miles while lightly abusing autotune. Once it's started and runs to temp I highly recommend just diving in

You may be braver than I. The garage I rent sits right on the sidewalk, with a 2" bump that I have to power the car over before it drops 4"+ (I have wooden ramps to help with that) before I get onto the narrow road, which is connected to two fairly busy roads just a few hundred feet away. I'm fully expecting to stall in the middle of the road and block traffic. . But once I'm past that, I do hope to just drive around and let auto-tune do its work.

Thanks!

-- Joe

[jangell]

Im a little late to the party.

Welcome!

Since the package got lost in the mail, i had to substitute in larger injectors. I made the original tune with 19# injectors, the ones you have a 22#. Some changes will need to be made.

I never did get your tune (I forgot to grab it before I sent you back the extra MS unit), so that wasn't a problem; I set up my own tune as best I could. Seems it needs more work.

It was not clear to me if the firing order issue was rectified. I know I had a hard time with this. I do not recall what i settled on to confirm your findings. I do know it was different than any information given the Mega manual. You may be onto something with my crank trigger positioning.

Just to sum up my most recent theory:

Engine cranks clockwise when looking at it from the back of the car. On a V6, the sensor should be 6 teeth ahead of the VR sensor. But the firing order you have on your wiring diagram is BAC, which is 120 degrees off from the EDIS-6 ACB firing order. Which means the sensor should be located 180 degrees off (since it's 10 degrees per tooth) -- right at the bottom of the pulley. However, the sensor is located at about 240 degrees (about 7:30 on a clock).

I'm fairly sure I have something wrong there, since obviously this wo.

There is a lot of adjustment in the VR sensor bracket

I think at least 20 degrees worth. Which is probably a significant part of my problem -- I have the sensor currently sitting at the very end of the bracket, as I didn't realize just how much adjustment there is (now that I realize that each tooth/gap set is 10 degrees, it's pretty obvious). Simply centering it might improve things considerably, and then I can time it and tweak it from there.

no need to correct it on the software side (I do not think this is even possible as the VR sensor is fully interpreted by the EDIS module). To confirm the alignment of the sensor you must do two things:
-Unplug the edis unit from the MS and start the engine. It should run, and you should find it is giving a solid 10deg with your timing light. This is its error state when it cannot communicate with the computer.
-Position the crank sensor as close to the trigger wheel as possible. Pull the fuel pump relay and power to the coil and turn over the engine with your DMM probes across the terminals of the crank sensor. You will be looking for an AC voltage of around 3-4V. Anything less suggest the sensor is too far away and will be getting a weak signal.

Sounds good. I'm sure I can get the VR pretty close to the gear, but I'll definitely test it. I'll do this next time I can, which probably won't be for about 2 weeks, unfortunately.

Thanks!

-- Joe

[jangell]

While you can change offset in TS with EDIS, here is a little more in depth reason as to why it's not proper-

The EDIS module will only change timing if it receives an order from the MS unit to do so. I stated earlier that my timing was off by 3 degrees. I could change the offset in TS to show 3 degree offset and my timing would be solid where I need it to be. BUT, if the EDIS module ever loses that signal with the MS unit, the module will default to 10 degrees timing, expecting my VR sensor is in the correct position. Meanwhile, my engine would be running at 7 degrees because it's no longer getting that offset message from MS.

So, you really need to make sure you have that base timing correct with the MS unplugged. Then you can see if you need to adjust your sensor and get where you need to be. THEN when it is timed properly while unplugged, plug it back in and command the MS unit to run at a fixed timing, confirm the timing is correct with your light and you know your MS is telling the EDIS the proper timing and without MS it will be 10 degrees.

Even if adjustment in TS is an option, I'd prefer to get the VR sensor as closely to calibrated as possible first. I definitely want limp-home to work if the MS signal is ever lost. This is definitely the first thing all do next, especially now that I've realized just how much adjustment Josh's bracket has.

Thanks again

-- Joe

[dn010]

Firing order doesn’t matter for this part. Put the crank pulley at TDC / 0 degrees and post a picture of the toothed wheel and sensor.

It doesn’t matter if the sensor is 120, 240, 90 degrees, you can mount the wheel and sensor at any degree you want, as long as the missing tooth is 6 away from the sensor when the crank is at TDC.


Sent from my iPhone using Tapatalk