Hey Guys,

So I finally got Megasquirt to start my car! Thanks to a distress call to Marc (Bullett) it turns out the car was not starting because the wires to the coil pack were reveresed.

Well, I finally got it started using Symtech's 420A base map. I am using MS2/Extra v3.2.4.

The only thing I changed were the required fuel and injector settings.

The car turns over now and runs for about 2 or 3 seconds before stalling. If I hit the throttle a bit here and there I could keep it alive for a little long but not much more.

Here are my log and msq files.... (I might have throttled during the log).


https://docs.google.com/file/d/0B22tb0XxveQ8UlBOY1dkT3FNbGM/edit?usp=sharing
https://docs.google.com/file/d/0B22tb0XxveQ8VDNpU1huR2FSM0E/edit?usp=sharing

Any ideas?!

Oh yeah.. I have a built bottom end (0.020 over) and 83 lb/hour Low Z injectors.

Just a few things I noticed quick when looking at the log, you need to calibrate you TPS. Also you are extremely lean which is probably why you are stalling out. Make sure your req fuel is set correctly and then if add some fuel to the map to try and keep the car running. Also check your WBo2 calibration to the gauge to make sure MS is reading the same number. Remember to check at a rich and lean point. Check around 11.0 and around 18.0

Originally posted by SilvrEclips
Just a few things I noticed quick when looking at the log, you need to calibrate you TPS. Also you are extremely lean which is probably why you are stalling out. Make sure your req fuel is set correctly and then if add some fuel to the map to try and keep the car running. Also check your WBo2 calibration to the gauge to make sure MS is reading the same number. Remember to check at a rich and lean point. Check around 11.0 and around 18.0

Thanks for looking at it! I won't be able to try some new things for at least a week (the car is at my parents house) but I will look into those.

The required fuel was calculated by TunerStudio. I used 2000cc, 83 lb/hr flow, and a 14.7 AFR. It calculated a required fuel of 3.8?? Is that not right? Is it safe to tweak that number or would you suggest I change the VE table below say 2,000 RPM to richen it a bit?

Also, I know I calibrated my TPS before logging this so perhaps what you are seeing is me throttling to try and keep it alive? Or are you saying that because the minimum is reading as 9%?

Also,

I compared the stock 420A base map from Symtech with some other 420a msq files in the depository and noticed each one had a different setting for the "sensor calibration". Right now mine is set to MPX4250... Will these settings have an effect on the idle/issue?

Yes the throttle should be close to 0% and 100% on the min. This will effect things like over run and closed loop idle if you choose to use those features.

I would leave the required fuel where it set it to and adjust your VE table. If you are running a base map I would scale the whole map when you make changes to try and richen it up. Then once you get the car idling you can see where you need to make changes to.

That sensor is your map sensor, you need to set it to whatever sensor is on the board.

Your running too lean and your Throttle position needs to be calibrated properly. Make sure your WBO2 sensor is calibrated correctly since its only reading Lean even when your PW go up. if the O2 is reading correctly then you need more fuel.

Also what are you using to control idle?

Have you set your injector dead time correctly also? that will screw up the injector PW if not set properly.

My vote is the injector dead time.

Are those high or low Z injectors? If they are low Z, you may not be opening them with enough hold current, and they will not open all the way, so MS thinks it is putting in a lot of fuel when really not much is getting in there.

Also make sure if you are running a fuel pressure other than what the injectors were rated at that you convert the flow over. I think 43.5 is typically used for flow specifications, so if you are running something other than that as your base pressure, convert the flow over to what it will really be. (Calculators online).

Oh yea, and what are you using for idle air control? Mine would start but die in a few seconds until I got my IAC set properly (Stock Stepper Based IAC). You can block it off and crack the throttle blades more to rule that out.

Thanks for all the tips guys. I was at my parents briefly the other day and gave it a try. Most of my time was taken up by trying to figure out why my wideband gauge was reading 14.8 for the couple seconds it idled but MS was reading 19. I think it was because I never grounded the wideband directly back to MS. I ran an alligator clip to the ground on the wideband and then to a ground straight back to MS and it looked like it was trying to read similar to the gauge. Unfortunately, the car wont idle long enough to see how close.

I have not touched the IAC setting yet (so they are set to whatever the base map from symtech is).

I will take a look at blocking the IAC and seeing what that does.

I have been reading about injector dead time and A majority of the information I am finding show how to calculate it based on a log (with a graph and etc..)... Can you guys point me to how I can calculate it prior to logging (since the car wont start).

Thanks agaiN!

Are you on High or Low Z Injectors?

I did not buy from symtech, but I believe their guides imply using a pwm idle valve, not a 4 wire stepper like the 420a has stock... you may or may not have the hardware on your board to run the stepper based one.

Additionally, it could be open or closed. If you pull the throttle body off, it is easy to see where the valve is in its cycle, but that is a pain. You can block it with a vacuum plug or some-such, just don't let the plug get sucked in the TB :P

Some info on Stepper IAC:

YouTube.com

I managed to use a trimmed down vacuum plug on the throttle stop to achieve about 7-8% tps when off the throttle. Of course, I disabled all the closed loop idle and such.

Originally posted by Tired2
Are you on High or Low Z Injectors? I did not buy from symtech, but I believe their guides imply using a pwm idle valve, not a 4 wire stepper like the 420a has stock... you may or may not have the hardware on your board to run the stepper based one. Additionally, it could be open or closed. If you pull the throttle body off, it is easy to see where the valve is in its cycle, but that is a pain. You can block it with a vacuum plug or some-such, just don't let the plug get sucked in the TB :P Some info on Stepper IAC:

YouTube.com

I managed to use a trimmed down vacuum plug on the throttle stop to achieve about 7-8% tps when off the throttle. Of course, I disabled all the closed loop idle and such.

They are Accel Low Z 83lb/h injectors.

Thanks for the information and video!

By the way.. Iw atched all your videos not too long ago before I installed MS... Thanks for them! I think the last one I saw was episode 23??

Cool, I'm glad you found them useful. Hopefully there are not too many holes/misinformation in them... I'm new to this as well.

If you have low Z injectors, from what I understand, when you set up the software peak and hold for MS2, you can get away with like 90 or 100% hold times as long as the engine is at idle or low load. Don't load the motor up like that though. Since the load is low, the fuel pwm duty will be low, so they will have time to cool down between squirts and not overheat. If you run them 100% current limiting (no current limiting), they should stay open while you tune in the idle.

Once you have it at a steady idle, you can start decrementing the 'hold' number of the peak and hold settings. Decrement by about 5%, burning the ecu each time, and give it a moment to settle out. If the idle is still good, continue to decrment until you find the percentage that the idle degrades. Then, go up like 3% from there and you've tuned in the peak and hold for your particular injector characteristics.

My hold percentage came out to like 75%, which is higher than I expected, but I'm not sure I'm entirely done with that setting since my idle was not completely stabilized.

Check out: Youtube: "Evan's Megasquirt Build Ep 25 - High vs Low Z Injector Sizing / Driving / PWM Basics" if you have 20 minutes to kill and don't have a full understanding of low Z peak and hold.

Originally posted by Tired2
Cool, I'm glad you found them useful. Hopefully there are not too many holes/misinformation in them... I'm new to this as well. If you have low Z injectors, from what I understand, when you set up the software peak and hold for MS2, you can get away with like 90 or 100% hold times as long as the engine is at idle or low load. Don't load the motor up like that though. Since the load is low, the fuel pwm duty will be low, so they will have time to cool down between squirts and not overheat. If you run them 100% current limiting (no current limiting), they should stay open while you tune in the idle. Once you have it at a steady idle, you can start decrementing the 'hold' number of the peak and hold settings. Decrement by about 5%, burning the ecu each time, and give it a moment to settle out. If the idle is still good, continue to decrment until you find the percentage that the idle degrades. Then, go up like 3% from there and you've tuned in the peak and hold for your particular injector characteristics. My hold percentage came out to like 75%, which is higher than I expected, but I'm not sure I'm entirely done with that setting since my idle was not completely stabilized. Check out: Youtube: "Evan's Megasquirt Build Ep 25 - High vs Low Z Injector Sizing / Driving / PWM Basics" if you have 20 minutes to kill and don't have a full understanding of low Z peak and hold.

So after watching your video and reading some more of the manual I just realized I don't remember ever installing a wire to the IAC... I am assuming it is the "FIdle" wire from the harness? Which wire should I solder it to on the IAC?

So, there are two (well more like 4 or so) types of IAC out there.

The stock one, at least on my '97 420A is a 4 wire stepper based IAC. To run it, you'll need to jumper in a few wires in the MS2 unless that has already been done. See the assembly guide, or contact the place you bought your MS from to see if you have support for it.

The other common type is a PWM one, that yes, will go on the FIDLE output. I'm seem to recall that the FIDLE needs a TIP120 on it because the transistor does not have enough umph to get it done. You'll have to read up on the assembly guide to figure that out though, I can't recall. Or, watch my first 6 videos or so, I think I talk about FIDLE back when I thought mine was a PWM single wire one.

The Symtechlabs 420A wiring diagram should get you close if you have a 4 wire IAC.

... The top left 4 wires that say IAC2B, etc. I don't have the wire colors handy, but it is easy enough to just ohm them out to find out which two pairs make a coil. Then, just use the test mode from my IAC video "IAC Physical Test" to make sure you have good movement. The Throttle body is easy to pull off... 5 minutes or so. I broke an IAC bolt trying to get it out, so just pull the TB and look inside.

Originally posted by Tired2
So, there are two (well more like 4 or so) types of IAC out there. The stock one, at least on my '97 420A is a 4 wire stepper based IAC. To run it, you'll need to jumper in a few wires in the MS2 unless that has already been done. See the assembly guide, or contact the place you bought your MS from to see if you have support for it. The other common type is a PWM one, that yes, will go on the FIDLE output. I'm seem to recall that the FIDLE needs a TIP120 on it because the transistor does not have enough umph to get it done. You'll have to read up on the assembly guide to figure that out though, I can't recall. Or, watch my first 6 videos or so, I think I talk about FIDLE back when I thought mine was a PWM single wire one. The Symtechlabs 420A wiring diagram should get you close if you have a 4 wire IAC. ... The top left 4 wires that say IAC2B, etc. I don't have the wire colors handy, but it is easy enough to just ohm them out to find out which two pairs make a coil. Then, just use the test mode from my IAC video "IAC Physical Test" to make sure you have good movement. The Throttle body is easy to pull off... 5 minutes or so. I broke an IAC bolt trying to get it out, so just pull the TB and look inside.

Okay, I will check it out... Will MS run without controlling the IAC? Is there a way to check if this is the issue before I do all the wiring??

You can adjust the throttle stop screw to open the throttle plate slightly to get enough air in so it can idle.

The throttle screw is covered and/or non existent on these... so, I just cut a vacuum plug down and put it on the little nub and got 6-7% throttle. I used another vacuum plug to block off the IAC tunnel in the throttle body since I was unsure if it was open or closed. Yours could be open, so I'd plug it, just don't let the plug go in the intake.

You can run without one, but you'll probably idle poorly when the motor warms up.

Originally posted by Tired2
So, there are two (well more like 4 or so) types of IAC out there. The stock one, at least on my '97 420A is a 4 wire stepper based IAC. To run it, you'll need to jumper in a few wires in the MS2 unless that has already been done. See the assembly guide, or contact the place you bought your MS from to see if you have support for it. The other common type is a PWM one, that yes, will go on the FIDLE output. I'm seem to recall that the FIDLE needs a TIP120 on it because the transistor does not have enough umph to get it done. You'll have to read up on the assembly guide to figure that out though, I can't recall. Or, watch my first 6 videos or so, I think I talk about FIDLE back when I thought mine was a PWM single wire one. The Symtechlabs 420A wiring diagram should get you close if you have a 4 wire IAC. ... The top left 4 wires that say IAC2B, etc. I don't have the wire colors handy, but it is easy enough to just ohm them out to find out which two pairs make a coil. Then, just use the test mode from my IAC video "IAC Physical Test" to make sure you have good movement. The Throttle body is easy to pull off... 5 minutes or so. I broke an IAC bolt trying to get it out, so just pull the TB and look inside.

What did you have your Idle Control Setting too? The base map for 420a is on/off.. did you do PWM warmup? Stepper Closed, open??

Just thought I would update everyone on my progress.

I got the car to idle!!! Granted, it is super rich and at 1200 RPM but it is staying on!

Doing some research on MSExtra I found this site:

http://injector-rehab.com/shop/lag.html

Completely lost and frustrated I used the table on that website for my Accel 83 lb injectors with a value of .85 for lag time (dead time). Right away the car held idle.

Of course, it stayed on but was surging from about 300 RPM to 1500 RPM.

Using what you guys said in this thread I then blocked the IAC. Right away the car idled smooth with no surge (with a 4% throttle).

My next step is to hook up the IAC to MS and find some issues with wiring... I am getting 19.2 AFR on Megasquirt with my AEM Uego but my AEM Uego gauge is reading 11.2 AFR at idle.

Additionally, my temperature and MAP reading on MS are reading much higher than they actually are:

Coolant Temp: Stock Reading = 202, MS Reading: 293
Intake Air Temp: Stock Reading = 109, MS Reading: 169
MAP Pressure: Stock reading = 13?? (Don't remember, but it was much lower than MS), MS Reading = 39

Do you think the temps and MAP are due to wiring or calibration? Same with the UEGO??

Thanks for all the help guys!

You need to tell MS what wideband you have so it can convert the raw analog value it reads to the AFR the wb02 is trying to use. The setting is under EGO control I belive, but it took me a while to find it last time I needed to change it.

If you are using stock clt temp sensor, you'll need to copy in the correct values for it, the MS expects a GM ones by default. Symtechlabs has the table for the stock one that you can fill in. Same goes for IAT if you use the stock sensor.

As for MAP... you should not be using the stock MAP, as it will not read up into boost well. Your MS unit should have a MAP onboard with a vacuum line going to it from the intake manifold... So, I'm not sure what is going on there. Find out what MAP is in your MS ECU and set that correctly as well.

Are you sharing sensors with the stock ECU? Or using GM sensors? The GM CLT and IAT are set up by default, so if you have those, go verify your selection in MS.

Separate GM sensors for IAT and Coolant.

I have a vacuum line into the MS unit and set it to the correct 3 Bar map that is on board.

I thought I set the EGO controller correct in Tunerstudio but I am not sure. All I could find for examples were in Megatune. I anyone running the Uego so I can verify my settings?

I found this post last night... Haven't tried it on the car yet...

http://msextra.com/forums/viewtopic.php?f=101&t=35521

This could help with the AFR's for sure. I know I did not set a custom wideband and used one of the default single widebands in tunerstudio.

Yea, give that a shot. As for the CLT, i think your MS one is off, if it was really 220, you'd be boiling coolant out of the spill tank.

Yeah... When the coolant temperature read 269 degrees I know something was up! I will keep you guys posted...

Is there any calibration of the temperature gauges for the GM sensors or should I check the wiring itself?

There is some good info here:

http://www.msextra.com/forums/viewtopic.php?f=101&t=45400

I think you can 'calibrate' it, but you should not need to if you know it is a good GM sensor. I'd guess wiring.

Did you follow the guide for proper grounding and bring all the sensors back to the MS unit instead of just wherever they land?

Yeah, I will check wiring before tweaking anything.

I brought them all the sensors back to a spot on the intake manifold where they meet up with all the grounds from MS... I also sent a ground straight to the battery from that same spot... I must have saw it this way somewhere and copied it haha.

They are brand new temp sensors from symtech so I am assuming the sensor itself is fine.

I did ground the MS to the block and to the maninfold, but I ran all my sensors directly back to the MS box. Supposedly grounding sensors at common engine ground points can give inaccurate readings, especially when other systems are pulling a lot of current. I would try to get each sensor, or maybe start with CLT and get it back to the ECU and see if that helps. If you built your own ECU, check the bias resistors for those sensors, if they are the wrong value that will mess you up also.

Both sensors are important, CLT will help calculate how much warm up enrichment you need, and how open/closed your IAC should be (it closes more the warmer the engine)... The IAT is very important also, because it helps calculate the mass of the air going into the system. PV=nRT, what I have affectionately named "perv-nert" is the ideal gas law equation.... the T is the temperature converted to kelvin. So, if you have an artificial increase in T, you'll have a proportional decrease in 'n', which is the amount of air you have in moles. That artifical decrease will make the ecu put in less fuel. So, if you get your AFR correct with bad sensor data, when the sensors start working correctly later on your tune will be junk again. I'm not sure this is new, but just a good chance to bring it up for those who have not researched it. P is pressure (MAP), and V is volume of your cylinder(s). R is the ideal constant.

Originally posted by Tired2
I did ground the MS to the block and to the maninfold, but I ran all my sensors directly back to the MS box. Supposedly grounding sensors at common engine ground points can give inaccurate readings, especially when other systems are pulling a lot of current. I would try to get each sensor, or maybe start with CLT and get it back to the ECU and see if that helps. If you built your own ECU, check the bias resistors for those sensors, if they are the wrong value that will mess you up also. Both sensors are important, CLT will help calculate how much warm up enrichment you need, and how open/closed your IAC should be (it closes more the warmer the engine)... The IAT is very important also, because it helps calculate the mass of the air going into the system. PV=nRT, what I have affectionately named "perv-nert" is the ideal gas law equation.... the T is the temperature converted to kelvin. So, if you have an artificial increase in T, you'll have a proportional decrease in 'n', which is the amount of air you have in moles. That artifical decrease will make the ecu put in less fuel. So, if you get your AFR correct with bad sensor data, when the sensors start working correctly later on your tune will be junk again. I'm not sure this is new, but just a good chance to bring it up for those who have not researched it. P is pressure (MAP), and V is volume of your cylinder(s). R is the ideal constant.

Cool I will check it all out this weekend. I will run the two gauge grounds directly to MS and not the common point between MS and the battery.

Thanks for the advice and the physics lesson ;) I actually understood all of what you said! haha

Originally posted by Tired2
There is some good info here: http://www.msextra.com/forums/viewtopic.php?f=101&t=45400 I think you can 'calibrate' it, but you should not need to if you know it is a good GM sensor. I'd guess wiring. Did you follow the guide for proper grounding and bring all the sensors back to the MS unit instead of just wherever they land?

By the way... for those of you following this thread and for future reference. The temperature issues I was having were because I never set the sensor types in Tunerstudio. Everything I read says GM is the default but I still had to go to "Calibrate Thermis..?" in Tunerstudio and select the GM sensor for both the IAT and Coolant sensors. Fixed my temp issues right away.

Originally posted by RfOeCnKdOeNr

Originally posted by Tired2 There is some good info here: http://www.msextra.com/forums/viewtopic.php?f=101&t=45400 I think you can 'calibrate' it, but you should not need to if you know it is a good GM sensor. I'd guess wiring. Did you follow the guide for proper grounding and bring all the sensors back to the MS unit instead of just wherever they land?

By the way... for those of you following this thread and for future reference. The temperature issues I was having were because I never set the sensor types in Tunerstudio. Everything I read says GM is the default but I still had to go to "Calibrate Thermis..?" in Tunerstudio and select the GM sensor for both the IAT and Coolant sensors. Fixed my temp issues right away.

Strange, I never had to set mine, maybe your settings got removed somehow. .. Or maybe I did, I dunno.

As for testing the wbo2, you can mount it in the tail pipe of another car, you could rig up a mount of some sort to get it in there just a bit... but I'm not sure why it would really ever output 5.8v on the analog. If that is the case, there is almost definitely something wrong. All they really do there is bring the reading in analog, condition it a bit maybe, and use a digital-to-analog output on some microcontroller to output the AFR value scaled to 0-5v.

Originally posted by Tired2

Originally posted by RfOeCnKdOeNr

Originally posted by Tired2 There is some good info here: http://www.msextra.com/forums/viewtopic.php?f=101&t=45400 I think you can 'calibrate' it, but you should not need to if you know it is a good GM sensor. I'd guess wiring. Did you follow the guide for proper grounding and bring all the sensors back to the MS unit instead of just wherever they land?

By the way... for those of you following this thread and for future reference. The temperature issues I was having were because I never set the sensor types in Tunerstudio. Everything I read says GM is the default but I still had to go to "Calibrate Thermis..?" in Tunerstudio and select the GM sensor for both the IAT and Coolant sensors. Fixed my temp issues right away.

Strange, I never had to set mine, maybe your settings got removed somehow. .. Or maybe I did, I dunno. As for testing the wbo2, you can mount it in the tail pipe of another car, you could rig up a mount of some sort to get it in there just a bit... but I'm not sure why it would really ever output 5.8v on the analog. If that is the case, there is almost definitely something wrong. All they really do there is bring the reading in analog, condition it a bit maybe, and use a digital-to-analog output on some microcontroller to output the AFR value scaled to 0-5v.

Maybe it is the sender then. Just to make sure... You are only using 2 wires from the wideband to MS right? The signal and the ground?

I remember when I hooked it up I had to search a bit to find the pinout of the wideband.

Yes, that is correct. The wbo2 input on the MS ECU is a 0-5v analog input. Technically you only need the white wire since the gauge will be grounded at the car, and the MS will be grounded at the car, but any difference in potential between those ground points will show up as error in your voltage reading. For instance, if the white wire had 4.8v with respect to gauge ground, megasquirt may see 4.6v with respect to the MS ground. So, there would be a .2v difference in potential between your two grounds, which is bad. So, you can avoid that by grounding all MS sensors right to the ECU, which should give them all a near-identical ground reference voltage. It sounds like you did this already, however.

The pinouts I saw in the UGEO manual show the blue wire is serial. That is another way to test the sensor if you'd like.. Find a DB9 female connector:

(Via the manual):
"The BLUE wire from the AEM UEGO Gauge shall be connected to Pin #2 (RX) on the serial port for receiving data.
Page 7
Pin # 5 (GND) on the serial port shall be grounded."

Here is the manual: http://forum.aempower.com/forum/index.php?action=dlattach;topic=21559.0;attach=4370

Then you just use Hyperterm to view the serial stream, it will output the AFR. MS won't make use of the serial stream, and it will be slower anyway I imagine, but it could be a good sanity check.

Update with issues...

As you know I finally got the car to turn on and stay on. The car is idling smoothly (no oscillation) at a high RPM (1200-1500) and the AFR reads super lean. Well, off the charts apparently. The AFR on tunerstudio reads ~19.5:1 and my UEGO guage reads nothing (out of rang I assume). Additionally, when I go to VE Analyze Live it reads "O2 Out of Range".

I did some research and found a couple threads on MS Extra that mostly pointed to wiring issues. The AEM wideband currently has the single signal wire going to the MS unit and the ground wire going directly back to the MS Unit. I also hooked up a voltmeter when the car was running and it read roughly 5.8 volts... (which is about 20:1...about what Tunerstudio says).

Here is a log of the car at idle.

https://docs.google.com/file/d/0B22tb0XxveQ8ampUQlMxR1VEX3M/edit?usp=sharing

I forgot to take the MSQ file off my laptop so I will post that later if you guys need it.

Also, I set up the EGO control to be a single wideband, and I calibrated the VE table based off the AEM UEGO settings.

Do you guys think this is a hardware or software issue?? I tried richening the tune and the AFR never changed...

Currently I use IAC Moving only, open loop. You can go closed loop once you get it close I think, but I've not really got it there yet. Then I just started with the default curve on the Idle Step Bins control.. (It goes froma bout 40 steps at -40F to 160 steps at 155F with a slight bend in the curve. Set the IAC mode and then start your car so it homes the stepper on key-on. After that, just try to get it to run ,and watch the X axis on the steps graph and modify the steps at the temp you are at until it idles well. Once it idles well there, let it warm up a bit more and keep tweaking. Of course, you may not have really cold temp right now, so this may need tweaked in the winter months, but it should be very close.

As for your wb02 sensor... The white wire should be 0-5v, so if you see 5.8 on it something is wrong. I'd disconnect it until you figure it out.. I'm not sure the inputs on the MS are protected from overvoltage, and they should see 5v max. (I would hope they are protected, but I'd have to check the schematics).

Originally posted by Tired2
Currently I use IAC Moving only, open loop. You can go closed loop once you get it close I think, but I've not really got it there yet. Then I just started with the default curve on the Idle Step Bins control.. (It goes froma bout 40 steps at -40F to 160 steps at 155F with a slight bend in the curve. Set the IAC mode and then start your car so it homes the stepper on key-on. After that, just try to get it to run ,and watch the X axis on the steps graph and modify the steps at the temp you are at until it idles well. Once it idles well there, let it warm up a bit more and keep tweaking. Of course, you may not have really cold temp right now, so this may need tweaked in the winter months, but it should be very close. As for your wb02 sensor... The white wire should be 0-5v, so if you see 5.8 on it something is wrong. I'd disconnect it until you figure it out.. I'm not sure the inputs on the MS are protected from overvoltage, and they should see 5v max. (I would hope they are protected, but I'd have to check the schematics).

Great. Thanks for the info about the IAC.

Does anyone know how I can test the UEGO to make sure the sender is or isn't shot? I'd rather make sure it is ruined before I buy a new sender for 80 bucks.

I decided to just suck it up and order a new wide and sensor for my UEGO... Should be coming in next week...

Silly question I thought of while pondering my issues... Dd you guys hook up your MS signal wire to the wide band before or after the plug? I don't know if it would have an effect but mine is hooked up after the sensor plug (so, sensor, plug, MS (tap-in), and AEM gauge)....

If I understand what you are saying, you might have it wrong... according to the manual, the white wire you want comes out of the gauge, so you should not have to 'tap' the wire anywhere. If you hooked in between the sensor and the gauge, that is incorrect. The wires that go to the sensor are either heater wires or unconditioned / unscaled raw output from the sensor. The gauge typically contains the controller that runs the heater for the sensor, and reads in the raw voltage from the sensor and displays it and converts it to 0-5v and serial stream. On the 4 pin connector on the back of the gauge, a white wire should come out. Hook that to the MS Ecu directly. If you hooked into the sensor white wire, I'm not sure what it does, but that would explain your high voltage reading on it. I'm not sure how protected that input is on the MS ECU, so you might have toasted it. Not that big of a deal probably if it is toasted, because there should be a spare analog in you can use (maybe), or you could toss on a canbus module if it is hosed to get some more inputs.

Oh, and I noticed it has calibration options on it for the analog output. Make sure that the rotary switch in the back is on setting P0 or P1, otherwise you'll be on a different voltage scale that the ECU does not expect. 0-5v is the default (P0/P1)

Originally posted by Tired2
Oh, and I noticed it has calibration options on it for the analog output. Make sure that the rotary switch in the back is on setting P0 or P1, otherwise you'll be on a different voltage scale that the ECU does not expect. 0-5v is the default (P0/P1)

Thanks... and thanks for the clarification above. It never crossed my mind until you posted the PDF for the UEGO. I installed the wideband over 4 years ago so I completely forget about the two pin outs! Well, let's just hope the MS didn't fry.

Thanks again and I will keep you all posted! I hope this discussion helps someone along the way as well.

Hey guys,

Just a little update. I happened to be in the area last night so I stopped at my parents house to get a little work done on the car.

I disconnected the MS unit from the Wideband PLUG and reconnected it to the signal wire from the GUAGE (which I so conveniently had already run under the dash and ready for a connection!. The car started up right away and the UEGO guage and megasquirt had the same reading!!! ####!

Well, it was #### for a minute of so...

After less than a minute of idling at about 14.8 AFR (showing both on the gauge and MS) the AFR slowly creeped up to 16, then 17, then 18.90 and went "out of range"...

Each time I turned the car off and on it would do the same thing.

Any ideas as to what would cause this? My thought is that it is still a hardware issue (wiring?) Because I think it even did it when the ignition was on and the car wasn't even on.

Also, I did run the ground from the gauge directly to the MS Unit.

Could it be the actual sensor causing this?

Here are my latest MSQ and log files (with the raising AFR)

https://docs.google.com/file/d/0B22tb0XxveQ8ampUQlMxR1VEX3M/edit?usp=sharing
https://docs.google.com/file/d/0B22tb0XxveQ8cFZrR25BcmJsdFU/edit?usp=sharing

Thanks in advance!

I don't have time to check the logs right now, but it could be your sensor heating up too much or something.

Technically the car should be running while the sensor is powered up or it could be damaged... at least according to my innovative gauge manual. Apparently they need the exhaust going to by to cool them or whatnot. So, just try not to leave it in the 'on' position for a very long time without running the motor. Seems like they would give you a tach input or something if it was that big of a deal, but who knows.

As for the numbers going up, it could be the sensor warming up, but it should hold steady while running. If you turn the car off, it will creep up as the mixture in the exhuast creeps more and more toward standard atmosphere, which is infinity to 1 or something AFR, so that is probably okay. The reading going up while you run sounds like possibly it is leaning your out because warm up enrichment is not adding as much fuel as the car gets warmer. You should be able to view that field in the log to see if that is changing while you go lean. Check the curve and make sure it has sane values in it, I think very cold gives 150% fuel, all the way down to 100% fuel at operating temp.

You can also check to see if it is the wbo2 causing a problem by looking at the pwm of your injectors to see if they are constant. If RPM, MAP, and IAT are pretty steady, a constant PWM should yield a constant AFR. That assumes the injectors are working as they should of course.

Originally posted by Tired2
I don't have time to check the logs right now, but it could be your sensor heating up too much or something. Technically the car should be running while the sensor is powered up or it could be damaged... at least according to my innovative gauge manual. Apparently they need the exhaust going to by to cool them or whatnot. So, just try not to leave it in the 'on' position for a very long time without running the motor. Seems like they would give you a tach input or something if it was that big of a deal, but who knows. As for the numbers going up, it could be the sensor warming up, but it should hold steady while running. If you turn the car off, it will creep up as the mixture in the exhuast creeps more and more toward standard atmosphere, which is infinity to 1 or something AFR, so that is probably okay. The reading going up while you run sounds like possibly it is leaning your out because warm up enrichment is not adding as much fuel as the car gets warmer. You should be able to view that field in the log to see if that is changing while you go lean. Check the curve and make sure it has sane values in it, I think very cold gives 150% fuel, all the way down to 100% fuel at operating temp. You can also check to see if it is the wbo2 causing a problem by looking at the pwm of your injectors to see if they are constant. If RPM, MAP, and IAT are pretty steady, a constant PWM should yield a constant AFR. That assumes the injectors are working as they should of course.

Yeah... i still think it's something with the gauge. Over at MSExtra they told me NOT to ground the guage to MS. SO I will try that next.

The log does have a constant PW, IAT, RPM, and MAP yet reads 19.8 AFR...

The PW is steady around 1.4 ... Does that sound correct for 83lb injectors?

I'm not sure if that is ideal or not for PW. You can start scaling the table though while it is running... select the whole thing, and increase the values. They should take effect right away, the higher you go, the more fuel you'll get, and the lower the AFR will go... See if you can get it adjusted.

Once you do that, you can determine if you want to recalculate your req_fuel, but if you know the injector value is correct it should be right.

If you can get the AFR back in range by scaling the table just a bit, you might be okay to go with that and start the rest of the tuning process. Don't drive it that lean though, that would not be good.

Another update for anyone following...

I finally have the wideband hooked up and working correctly! I hooked the white signal wire to the MS unit and the black ground to it's own ground (NOT MEGASQUIRTS GROUND). The AFR reads identical on my guage as it does on Tunerstudio. The slowly rising AFR doesn't happen anymore either.

I started the car and attempted to tune idle by adjusting the injector characteristics a percentage and millisecond at a time (just as the manuals say to). I eventually got a decent idle but it got worse as the car warmed up. I continued tweaking and got a smoother idle but it had a slight oscillation.

Additionally, the car would studder when given throttle. The AFR's would richen and the car would almost shut off before the RPMs would raise and continue running.

I decided to check the IAC valve that was now hooked up thinking that is may be the issue. I took the throttle body off and used the IAC test mode.

No luck.

I then remembered in Evan's youtube video and in the manual it stated that the polarity of the pins might be backwards. So I made some "jumper" wires to switch the polarities of the plug. No luck. I tried every single combination of the IAC plug and never got a single bit of movement.

Am I missing something here? Is there a way to test if the MS unit is giving the sensor the signal it needs? I have Tunerstudio set to IAC Stepper Moving Only. Do I need to tell Tunerstudio anything else about the sensor? Is there a place to do so?

Could the fact that the IAC is slightly open have to do with the studdering on throttle?

I have to say the MegaSquirt is getting extremely frustrating to me... Extremely...

Hang in there, it sounds like you are making progress at least. Keep in mind that the stock ECU took much engineering to get right initially, and you are trying to replace most of its functionality. It does not come easily, but the things you are learning here apply to most every car on the road. You may have noticed that a good bit of the old school mechanics are bowing out with statements like "I can fix a car, but I not when all these computers get in the way"... or something like that. That is because it is not easy... you are attempting one of the most difficult parts of working on cars that most people pay a professional to do.

Now that that is out of the way...

You might try disabling Acceleration Enrichment, and be sure EGO control is turned off for initial testing. At least, that is my approach... limit the system to as simple as you can get it, then enable features one at a time making sure they don't mess it up.

For the IAC, are you sure your ECU has the proper jumpers in place to run the IAC? Your vendor would have set it up this way. There is an IAC test mode that I went through in my video, but it is not that great. When you power on the ignition, you should see it move to 'home', then move to where it is set to go in the IAC steps table for your current engine temp. After that, it will close further and further (with the 'base' IAC steps table) as the engine warms up.

You might also check the log for warm up enrichment if you have not. It will add fuel while the engine is cold to keep it running well, then get back to 100% of reqfuel once you are at operating temp.

I don't think your studder on throttle problem is the IAC though. Once you open the throttle blades it basically bypasses the IAC. Having a stuck or non-functional IAC that is half open is just like cracking the throttle blades a bit or using an idle air screw to adjust idle speed.

I wish I could give more advice, but I'm basically at the same point you are, only my stock ECU is being a Bitch right now. I don't have any time to work on it right now, but as I try more things I'll post more videos (hopefully).

To the OP now with some more progress can you post an updated Tune file and Log?
You are getting closer though dont give up, as I try to remember I too used to have these same issues when I first when standalone. But like stated above, turn everything off like EGO control and Acceleration enrichment. For idle settings you should try Closed-Loop always on instead of moving only ive always had better results with. You definitely need to know for sure if your iac stepper motor is working or not.

Thanks for the encouraging posts guys... I will post a log and tuning file ASAP. The car is located an hour drive away at my parents house so that is adding to the frustrations!

I am going to email sum tech with the serial number and find out what kind of IAC is on the board if any... The harness had all four IAC wires soldered in and ready to roll...

I'll keep everyone posted.

Also, I have been calling some shops in my state and none of them want to deal with megasquirt. Any suggestions? My plan was to get the best tune I can and then bring it to a dyno with someone who tunes cars all day to fine tune it.

I just talked to Symtech and apparently my EXU only has the FIdle drive installed!!!

So I guess I need to re-hookup the IAC to the stock ecu and splice the FIdle wire into one of the signals?? Any idea which pin to splice into for our cars??

Originally posted by RfOeCnKdOeNr
I just talked to Symtech and apparently my EXU only has the FIdle drive installed!!! So I guess I need to re-hookup the IAC to the stock ecu and splice the FIdle wire into one of the signals?? Any idea which pin to splice into for our cars??

So, if you have FIDLE only, you have a few options...

Your stock ECU has no use for the IAC, and probably will cause you more trouble than it is worth even if it did want to control it still. It would probably be hard to tune MS when MS was not controlling it.

So, you can remove the IAC if you'd like and install and FIDLE Valve. There are a few types, but probably you'll want a 2 or 3 wire PWM (I think both work fine). I know bosche makes one, there is info on the megamanual about them. You can probably find one at a junk yard or order one online pretty cheap. See if you can find some info about how to convert over the IAC to PWM FIDLE on our cars, I never really looked into it. Maybe Symtech can help. It should be as easy as finding one that fits and mounting it up.

The wires are in the harness because the IO can be used as spares for other stuff.

I'm not really sure why symtech did not fit the IAC stuff since the 420A is factory IAC, and I have no problems controlling it with the stock MS2 hardware. If they give you the answer to that, please post it back here... I'm not a symtech customer, so I could never really ask.

One other option is a FIDLE non-pwm valve. A lot of racing guys use this because it is simple... It is just on/off. If it will be a daily driver, PWM would be nicer, but you could get away with an on/off style as well. Like other IAC valves, it makes a vacuum leak, but instead of the amount of 'leak' being controlled as with a PWM or Stepper based IAC, it is just 'leak or no leak'.

One final option would be to find out what it takes to enable the onboard stepper control for your ECU. It should be as easy as jumpering in the 12v power up to the stepper chip on the MS2 (S12 comes to mind but you'd need to check the manual)... Then you just have to jumper in the pins near the DB37 on the board to use IAC.
Then, the FIDLE wire could run cooling a fan relay or AC, etc.

Originally posted by Tired2

Originally posted by RfOeCnKdOeNr I just talked to Symtech and apparently my EXU only has the FIdle drive installed!!! So I guess I need to re-hookup the IAC to the stock ecu and splice the FIdle wire into one of the signals?? Any idea which pin to splice into for our cars??

So, if you have FIDLE only, you have a few options... Your stock ECU has no use for the IAC, and probably will cause you more trouble than it is worth even if it did want to control it still. It would probably be hard to tune MS when MS was not controlling it. So, you can remove the IAC if you'd like and install and FIDLE Valve. There are a few types, but probably you'll want a 2 or 3 wire PWM (I think both work fine). I know bosche makes one, there is info on the megamanual about them. You can probably find one at a junk yard or order one online pretty cheap. See if you can find some info about how to convert over the IAC to PWM FIDLE on our cars, I never really looked into it. Maybe Symtech can help. It should be as easy as finding one that fits and mounting it up. The wires are in the harness because the IO can be used as spares for other stuff. I'm not really sure why symtech did not fit the IAC stuff since the 420A is factory IAC, and I have no problems controlling it with the stock MS2 hardware. If they give you the answer to that, please post it back here... I'm not a symtech customer, so I could never really ask. One other option is a FIDLE non-pwm valve. A lot of racing guys use this because it is simple... It is just on/off. If it will be a daily driver, PWM would be nicer, but you could get away with an on/off style as well. Like other IAC valves, it makes a vacuum leak, but instead of the amount of 'leak' being controlled as with a PWM or Stepper based IAC, it is just 'leak or no leak'. One final option would be to find out what it takes to enable the onboard stepper control for your ECU. It should be as easy as jumpering in the 12v power up to the stepper chip on the MS2 (S12 comes to mind but you'd need to check the manual)... Then you just have to jumper in the pins near the DB37 on the board to use IAC. Then, the FIDLE wire could run cooling a fan relay or AC, etc.

Haha I am just happy to know that it wasn't my wiring... Man, i sat there for hours trying to figure out why the IAC wasn't moving.

I emailed Symtech back with some of the questions... I will keep you guys posted.

Jeffrey,

Installing a FIDLE valve on a 420A is pretty straightforward. There are some generic instructions here:

http://www.megamanual.com/v22manual/mwire.htm#idle

Not all 420A/Neon’s have a stepper motor IAC. For those that do, though, the stepper IAC circuit is available as an option through our option service:

http://www.symtechlabs.com/catalog/megasquirt-option-service-p-43.html

The FIDLE driver is not compatible with a stepper motor IAC. If you would like to attempt to modify your ECU yourself to use a stepper motor IAC, please see step 26i here:

http://www.msextra.com/doc/ms2extra/build_manual.htm

Your MegaSquirt ECU currently uses the stepper motor driver for the tachometer output and general purpose outputs, so you will lose these functions if you install the stepper IAC circuit. You can, however, use the FIDLE circuit as a general purpose output and install a stand-alone tachometer output circuit.

Best regards,



SymTech Labs Support

It sounds like you would be best off just finding the stock FILDE valve off another 420A. Not sure what your budget is like, but I'm sure it won't break the bank too badly.

Mine had Stepper IAC, so I assumed they all did. That always confused me, turns out they just have one or the other. Maybe someone else can chime in?

Originally posted by Tired2
It sounds like you would be best off just finding the stock FILDE valve off another 420A. Not sure what your budget is like, but I'm sure it won't break the bank too badly. Mine had Stepper IAC, so I assumed they all did. That always confused me, turns out they just have one or the other. Maybe someone else can chime in?

Seems like it.... Also sounds like it can only be an open/close valve and not PWM or am I reading that wrong?

Anyone else running with the FIDLE instead of IAC stepper??

Originally posted by RfOeCnKdOeNr

Originally posted by Tired2 It sounds like you would be best off just finding the stock FILDE valve off another 420A. Not sure what your budget is like, but I'm sure it won't break the bank too badly. Mine had Stepper IAC, so I assumed they all did. That always confused me, turns out they just have one or the other. Maybe someone else can chime in?

Seems like it.... Also sounds like it can only be an open/close valve and not PWM or am I reading that wrong? Anyone else running with the FIDLE instead of IAC stepper??

The 'official' FIDLE pin on MS2 is PWM capable, so it will do either type.

I did not find this for the longest time, but when I finally did it was helpful: http://www.msextra.com/doc/ms2extra/MS2-Extra_Hardware.htm#ms2options

Originally posted by Tired2

Originally posted by RfOeCnKdOeNr

Originally posted by Tired2 It sounds like you would be best off just finding the stock FILDE valve off another 420A. Not sure what your budget is like, but I'm sure it won't break the bank too badly. Mine had Stepper IAC, so I assumed they all did. That always confused me, turns out they just have one or the other. Maybe someone else can chime in?

Seems like it.... Also sounds like it can only be an open/close valve and not PWM or am I reading that wrong? Anyone else running with the FIDLE instead of IAC stepper??

The 'official' FIDLE pin on MS2 is PWM capable, so it will do either type. I did not find this for the longest time, but when I finally did it was helpful: http://www.msextra.com/doc/ms2extra/MS2-Extra_Hardware.htm#ms2options

Just an update... I am going to enable the stepper motor IAC on my ECU. I looked into the FIDLE option and I just didn't like it. Here is the email from Symtech when I asked if the MSExtra instructions for doing the IAC stepper setup on my own were correct:

"Your MegaSquirt ECU uses a v3.0 board. You will find the only jumpers you actually need to install are JS1 to IAC1B and JS3 to IAC2B (the rest are already installed). Again, you will lose the tachometer and general purpose output functions by doing this, though."

Sounds simple enough?! I am going to try it next weekend when I go to parents house.

I will keep you guys posted!

Get it going yet?

I finally got back to my project... hopefully I'll be running it again soon for tuning.

I wanted to post this warning here.. though the thread is still panning out, I found a problem with the wiring I used on my car, which is very similar to what symtech labs suggests.

Basically, if you pull the MS ECU 2A fuse from the circuit, the P-Type resistors used in the MS ECU no longer are pulled high, creating a short to ground. So, if you have low Z injectors like I do, you get 55 Watts to each one (picture the heat a 60W light bulb makes)... Anyway, pull the fuse, you can toast your injectors, and maybe your coil.

The 'right' answer is to follow the MS3 'external wiring' layout instead, giving 12v to your coil and injectors downstream of your fuel pump relay, which is controlled by the MS ECU.... so they can never be powered unless the ECU is.

I'm not sure why they don't just use N-Type transistors here. I'll try to figure that out, but it is not really relevant because they just don't.

Everything should still work, just be aware your 2A fuse could blow and toast your coil, injectors, and possibly your battery if you don't notice it has a new 20A draw on it.

High Z injectors will be better off, but still not good.

Originally posted by Tired2
Get it going yet? I finally got back to my project... hopefully I'll be running it again soon for tuning. I wanted to post this warning here.. though the thread is still panning out, I found a problem with the wiring I used on my car, which is very similar to what symtech labs suggests. Basically, if you pull the MS ECU 2A fuse from the circuit, the P-Type resistors used in the MS ECU no longer are pulled high, creating a short to ground. So, if you have low Z injectors like I do, you get 55 Watts to each one (picture the heat a 60W light bulb makes)... Anyway, pull the fuse, you can toast your injectors, and maybe your coil. msextra thread here: http://www.msextra.com/forums/viewtopic.php?f=132&t=50803 The 'right' answer is to follow the MS3 'external wiring' layout instead, giving 12v to your coil and injectors downstream of your fuel pump relay, which is controlled by the MS ECU.... so they can never be powered unless the ECU is. I'm not sure why they don't just use N-Type transistors here. I'll try to figure that out, but it is not really relevant because they just don't. Everything should still work, just be aware your 2A fuse could blow and toast your coil, injectors, and possibly your battery if you don't notice it has a new 20A draw on it. High Z injectors will be better off, but still not good. Maybe symtech will respond to this post or my post on MSextra.

So, if the 2A fuse blows... the Injectors could be toasted??

I hit another snag last weekend. I decided to enable the stepper IAC driver myself because symtech said it was just jumping two things. Well, what they said and what my board looked like wasn't jiving...

Here is the thread with images on their response...

http://msextra.com/forums/viewtopic.php?f=132&t=50741

I called and emailed symtech and they finally emailed me back a few hours later. By that time I fried the stepper chip from trying to figure it out.

The new chip just came in and I will attempt all of the above again soon! lol

Oh, that sucks... frying chips.

So, I made a mistake when initially reading the wiring guide for 420a from symtech, as well as various other guides I have read.

Hopefully you got it right, but I made a mistake.

My mistake, which I'll hopefully fix in my youtube videos soon, is that I power the MS, coils, and injectors all from the same ignition switched line. (via a relay I added).

The problem is that if the MS loses power (by pulling or blowing the 2A fuse), it will allow a ground path to coils/injectors, which is bad, because that is basically the "on" state, which can blow up injectors, especially low Z.

The solution is simply to follow the guides and don't mess up like I did. That means wiring the system so that the fuel pump relay provides the power to the injectors, the coils, and of course the fuel pump. On our cars you can use the ASD relay as well, but I won't go into that, symtech has a writeup.

So, that way, if the MS ECU ever loses its power, the fuel pump relay will not have power. As I'm writing this, however, I'm realizing that the fuel pump relay pin is a P-Type as well, so wtf, I'm losing my mind I think.

Okay, sorry if this is off topic... but it is still relevant to MS users...

I got the issue all sorted out in this thread:

http://www.msextra.com/forums/viewtopic.php?f=132&t=50803

Basically, your main relay should not power up everything right from the ignition switch, if you do that, and your main ECU fuse blows or is removed, the injectors and coils can get shorted to ground through the PNP transistors they use.

So, the fuel pump relay output uses an NPN transistor, which will not allow ground connection through the ECU if the ECU loses power.

Symtechlabs has it right on the 420a schematic, so make it match that and you'll be fine.

If you switch everything on with one relay, just don't remove the power to the ECU with they key on or you could toast your injectors.

In summary:

Power ECU from key switch. It will enable the fuel pump relay. Power everything else using the 12v power that comes out of the fuel pump relay. (Ideally this would be just powering up your 'main' relay... in essence making it less 'main' than once intended.)

How is your iac fix going? Get the chip replaced?

I realized tonight that the idle steps settings did not match my system when using the symtech base msq... (probably because they use pwm valve)... anyway, it was actually opening more as it got warm. Luckily I had my video from testing it out and set the steps to go from like 160 to 250 or so instead of the other way around. Just keep an eye out for that... your best bet is to pull the TB off and run the test sequence and see the step count where it starts, then ends, etc.

Originally posted by Tired2
How is your iac fix going? Get the chip replaced? I realized tonight that the idle steps settings did not match my system when using the symtech base msq... (probably because they use pwm valve)... anyway, it was actually opening more as it got warm. Luckily I had my video from testing it out and set the steps to go from like 160 to 250 or so instead of the other way around. Just keep an eye out for that... your best bet is to pull the TB off and run the test sequence and see the step count where it starts, then ends, etc.

Hey, I am expecting the ECU back from symtech this week. They appear to be having issues with USPS so I don't have a tracking number.

Thanks for the heads up on the IAC steps. I do plan on taking the TB off and testing it similar to your video. Hopefully I will have an update after the upcoming long weekend (I should get the ECU before then).

Originally posted by Tired2
I'm not sure why they don't just use N-Type transistors here. I'll try to figure that out, but it is not really relevant because they just don't. Everything should still work, just be aware your 2A fuse could blow and toast your coil, injectors, and possibly your battery if you don't notice it has a new 20A draw on it.

The injector drivers do use N-channel MOSFET's. The main problem is that power could backfeed through the flyback damping circuit, keeping the ECU powered up as long as the injectors are powered.

The MOSFET's might occasionally stick open because there aren't pull-down resistors to dissipate the gate charge, but that isn't a common occurrence. You could install pull-down resistors (10k or 100k ohm from R15 and R20 to ground), but as you noted, it's much better practice to just ensure the ECU controls power to the injectors and coil.

FINALLY!!!!

I got the ECU back from Symtech and hooked it up this morning... Well, the idle sounded like it was skipping a coil. So after much frustration and comparing my MS wire diagram with the one Symtech sent I realized I had to change one of my spark pins!

With the new IAC stepper motor driver installed I was able to set up the idle a little better (all though it is still idling pretty high... almost like a vacuum leak... I'm thinking the IAC is open).

Either way... I got the car idling smooth tweaking the injector settings (hopefully they are safe settings... Please, if you know anything about this stuff feel free to take a look at my MSQ and Tunerstudio file and let me know if I am doing something wrong).

Once idled I turned on VE Analyze Live and took to the streets. Slowly filling in the map the car was RUNNING!

After taking it around locally I continued VE Analyze live onto the highway. Made sure the numbers didn't look too crazy and burned it here and there.

Finally, I even drove it on an hour long trip from my parents house to my apartment.

Here are the logs and tunerstudio file from that hour long trip. I haven't sat down and studied it too hard yet but watching my UEGO gauge I noticed the car was running pretty rich when cruising in 5th gear at about 70MPH.

Here are the files... Constructive criticism extremely welcome (although 95% of the tune right now is VE Analyze Live)

https://docs.google.com/file/d/0B22tb0XxveQ8Qm9NOVFJeTE0ODg/edit?usp=sharing
https://docs.google.com/file/d/0B22tb0XxveQ8WGZaR25EMlFOcUU/edit?usp=sharing

Congrats on getting it going. I took a glance at your settings. Why do you have fuel signal from MS3X? do you have an ms3x board? it should be on standard fuel if using injector drivers off the mainboard. Also you might want to zero out your priming pulsewidth since your just starting out, if you keep turning the key on and off it will be spraying fuel into the cylinders which you really dont need right now. And under idle control try Always On instead of Moving Only, you could lose steps when in moving only.
Also if you turn on idle advance you could keep the idle at a specific advance somewhere between 5-10* is normal, this could help keep the idle steady and not so high.
And under idle control you have your 'valve closed step' at 100 which should be lower than your 'min pid steps'..if your 'valve open' steps are 60 then your 'valve closed' steps should be less, so try 20 since thats what you have under your 'min pid steps'

Everything else looks okay, You may need to play with your step sizes and PID settings to get your idle dialed in perfect.

Originally posted by Gruff511
Congrats on getting it going. I took a glance at your settings. Why do you have fuel signal from MS3X? do you have an ms3x board? it should be on standard fuel if using injector drivers off the mainboard. Also you might want to zero out your priming pulsewidth since your just starting out, if you keep turning the key on and off it will be spraying fuel into the cylinders which you really dont need right now. And under idle control try Always On instead of Moving Only, you could lose steps when in moving only. Also if you turn on idle advance you could keep the idle at a specific advance somewhere between 5-10* is normal, this could help keep the idle steady and not so high. And under idle control you have your 'valve closed step' at 100 which should be lower than your 'min pid steps'..if your 'valve open' steps are 60 then your 'valve closed' steps should be less, so try 20 since thats what you have under your 'min pid steps' Everything else looks okay, You may need to play with your step sizes and PID settings to get your idle dialed in perfect.

Awesome! Thanks for all the advice! I will give it all a try and let you know how it goes. I am not ashamed to say I am a complete #### so anything will help.

Originally posted by Gruff511
Congrats on getting it going. I took a glance at your settings. Why do you have fuel signal from MS3X? do you have an ms3x board? it should be on standard fuel if using injector drivers off the mainboard. Also you might want to zero out your priming pulsewidth since your just starting out, if you keep turning the key on and off it will be spraying fuel into the cylinders which you really dont need right now. And under idle control try Always On instead of Moving Only, you could lose steps when in moving only. Also if you turn on idle advance you could keep the idle at a specific advance somewhere between 5-10* is normal, this could help keep the idle steady and not so high. And under idle control you have your 'valve closed step' at 100 which should be lower than your 'min pid steps'..if your 'valve open' steps are 60 then your 'valve closed' steps should be less, so try 20 since thats what you have under your 'min pid steps' Everything else looks okay, You may need to play with your step sizes and PID settings to get your idle dialed in perfect.

A couple questions... I am working offline getting some of these settings going before I run outside and plug in the laptop.

First, I have MS2/Extra... where in tunerstudio are the settings for the fuel signal that you are talking about.?

Second, if I turn on the idle advance settings I can set different parameters for when it come on, which is fine, but where do I set the actual idle advance for when these paramaters are met and it turns on?

Thanks!

Once you turn Idle advance on, you'll see a new option light up on the "Extended" menu at the bottom... "Idle advance Timing".

Oh, and check this out if you have not...

http://www.diyautotune.com/tech_articles/megasquirt_idle_tuning.html

Gives a better idea of what you are trying to accomplish using idle advance (though it does not mention that actual setting, same thing I think)

Edit: And this:

http://www.msextra.com/doc/ms3/Idle_Advance.html

via above link: "In general, the most stable idle is reached by decreasing the idle timing, and increasing the amount of air entering the engine (using an idle air valve or similar). As such the idle advance timing should be as low as possible while retaining a smooth idle. Since less timing is used during normal idle conditions, as load increase, the timing should also increase to counteract RPM decrease when the load increases."

Originally posted by RfOeCnKdOeNr

First, I have MS2/Extra... where in tunerstudio are the settings for the fuel signal that you are talking about.?


Sorry my fault you dont have that option..when i opened your tune file my computer opened it like a ms3x tune. lol disregard that part of my comment


But for idle you basically drop timing really low so you can open the IAC more to let more idle air in. This is mainly to keep your # of idle steps higher so you dont need to run really low steps. Some stepper motors will lose steps when trying to get the stepper motor to close to 0 steps.

Okay. We are getting somewhere....

Idle is much smoother and around the 900-1200 RPM range! ####!

One problem, maybe you know which of the tweaked settings could cause this. When I rev the motor a little but at a certain point it will automatically rev up to around 3500 RPMs for a few seconds and then return to idle (even if I take my foot off the accelerator).

Any ideas?

Thanks!

Ase settings?

Originally posted by Tired2
Ase settings?

I never changed those... Here are my ASE settings..

https://docs.google.com/file/d/0B22tb0XxveQ8NE5keDdISUpZYUE/edit?usp=sharing

Would that cause a rise in RPM from 1200 to 3400 ish??

It didn't do the rise and hold RPM thing until after I tweaked the IAC settings to something similar to your settings and the idle advance settings.

Do the attached ASE settings look okay to you guys??

Sorry, I meant accel enrichment, Ase only lasts a few seconds after starting.

That's kind of what I figured... I just finished reading this:

http://www.diyautotune.com/tech_articles/megasquirt_acceleration_enrichment_tuning.htm

I'm going to take a look at that but would adding fuel during a quick accel make it hold a 3000 rpm for a few seconds ? Wouldn't it need air to hold? Like the iac being open?

I would think it would but iac does not change with tps, only temp. Once warm iac steps should hang still. Check the log and see what iac steps are at when this event goes down.

I just went out to datalog it and the second I started it the car ramped up to 3400 RPMs and held it.... I let it go for a bit to datalog and shut it down...

Here is the tuning file and datalog...

https://docs.google.com/file/d/0B22tb0XxveQ8el9JMmhueFpzUXM/edit?usp=sharing
https://docs.google.com/file/d/0B22tb0XxveQ8S3d4Q3pMdG1IRUU/edit?usp=sharing

Let me know if anything looks out of the normal. The only changes I made were the IAC settings and ignition advance. Not sure which one would cause this.

Sorry when I said 'Always On" I meant 'closed loop always on' since you were already running closed loop..You will get better results with closed loop.
But the reason your idle is so high is because when you start it up your IAC is opening up to 150 steps and increases with Coolant temps. This is because since you switched to regular always on you made it switch over to read from the 'idle steps' graph. If you go to the 'iac steps' menu under your idle settings you will see you have the # of steps increasing greater as CLT temps increase, which is all wrong; the # of steps should be decreasing as CLT temps increase.(as stated in the manual for stepper control, 0-steps=closed; 255-steps=fully open. So when you start it up your giving it tons of air so its high revving.
And turn your warmup temp cutoff temp down to around your thermostat temp rating. 180 will work. You have it set at 208 to cutoff so it may take a while to get to 208 and while driving at high speeds if/when CLT temps drop below say 200 then your WUE will be cutting on and off while driving.

Originally posted by Gruff511
Sorry when I said 'Always On" I meant 'closed loop always on' since you were already running closed loop..You will get better results with closed loop. But the reason your idle is so high is because when you start it up your IAC is opening up to 150 steps and increases with Coolant temps. This is because since you switched to regular always on you made it switch over to read from the 'idle steps' graph. If you go to the 'iac steps' menu under your idle settings you will see you have the # of steps increasing greater as CLT temps increase, which is all wrong; the # of steps should be decreasing as CLT temps increase.(as stated in the manual for stepper control, 0-steps=closed; 255-steps=fully open. So when you start it up your giving it tons of air so its high revving. And turn your warmup temp cutoff temp down to around your thermostat temp rating. 180 will work. You have it set at 208 to cutoff so it may take a while to get to 208 and while driving at high speeds if/when CLT temps drop below say 200 then your WUE will be cutting on and off while driving.

Gruff511,

If you watch this, starting at about 2 min, you can see mine closes as steps increases. I'm not sure if wiring can cause it to 'run backwards', but mine definitely is about closed at ~250.

http://youtu.be/CLRyLVdAT2c?t=2m

Yeah i remember seeing over on msextra some of ms2 codes are actually programmed so the IAC steps are higher when closed..when the later firmwares are opposite where fully closed is 0 steps.
But either way the OP idle is high revving because its getting too much air, probably since he has closed loop off and other parameters are working his idle. While if running closed loop the ecu will adjust steps to steady idle

I was playing with my idle tonight... essentially we are at the same point, I just have not yet drove mine around. I idled it for about 30 minutes tonight now that I have my cooling fan going. It runs about 193 or so. What does your run at?

As fro your log... Does it only do the 3400 RPM thing when you start it warm? I noticed compared to mine.. my car idles around 1100-1200 with IAC steps maxing out at about 155 or so. Yours goes on up from there, which is "more closed".

I think your advance may be too high, mine rides about 20deg at idle, and I started with the stock symtech tune. Check that your idle advance is working as expected and have a look at the spark table.

I'm not sure exactly how idle advance is supposed to work, but it does not have the effect I'd imagine. As load goes up, I advance, but the advance does not seem to increase from normal even though the conditions should be met. Maybe I missed a condition on mine.

Originally posted by Tired2 I'm not sure exactly how idle advance is supposed to work, but it does not have the effect I'd imagine. As load goes up, I advance, but the advance does not seem to increase from normal even though the conditions should be met. Maybe I missed a condition on mine.

Seems like you may be confused on what 'idle advance' does..its not supposed to change your timing as load increases. It merely just fixates to a specific degree of timing while only in idle..when you give it gas you timing will switch back to the set timing in your spark table. So idle advance deactivates when you press the throttle

Originally posted by Gruff511

Originally posted by Tired2 I'm not sure exactly how idle advance is supposed to work, but it does not have the effect I'd imagine. As load goes up, I advance, but the advance does not seem to increase from normal even though the conditions should be met. Maybe I missed a condition on mine.

Seems like you may be confused on what 'idle advance' does..its not supposed to change your timing as load increases. It merely just fixates to a specific degree of timing while only in idle..when you give it gas you timing will switch back to the set timing in your spark table. So idle advance deactivates when you press the throttle

That is basically what I thought it did, but when you enable the setting, you get a 'curve' to define. So, I have mine set so that as load increases while idling, the timing jumps up a bit to steady the idle. I agree this may not be the best thing to do, I'm just going off of diyautotune's suggestion of creating a spark map the increases just a bit along the far left side of the table to help steady the idle. It was my assumption that 'idle advance' was a special function that extended the functionality of simply setting up your table like diyautotune suggests (linked a few posts up).

Do you have any suggestions on what the spark table should look like at idle? 20 deg, 30 deg?

Mine is doing okay, but stumbles a bit once in a while. I'm still on 'open loop always on', so maybe I just need to spend some time with the closed loop settings.

Not sure which firmware your running but some people have had better luck then others with different settings. I first started off running open loopbut james over on msextra says closed loop should respond better with controlling the stepper. I've always had better results with closed loop anyways

Okay, I did some reading. I think mine just runs backwards, I'll probably need to switch a coil around on the IAC to make it run the other way.

I think I'm fine for open loop still, and probably would be good for closed loop, but I imagine I'd have to invert my gains and such, which might not be worth it since I should only have to swap one set of wires.

I'll be waiting though, until I can pull the TB off again though to really map out the steps/location.

OP ----

Just be sure you verify yours, if yours is wired differently than mine is (very possible), you could be adjusting the wrong way.

This may cause problems with homing it as well.

Originally posted by Tired2
Okay, I did some reading. I think mine just runs backwards, I'll probably need to switch a coil around on the IAC to make it run the other way. I think I'm fine for open loop still, and probably would be good for closed loop, but I imagine I'd have to invert my gains and such, which might not be worth it since I should only have to swap one set of wires. I'll be waiting though, until I can pull the TB off again though to really map out the steps/location. OP ---- Just be sure you verify yours, if yours is wired differently than mine is (very possible), you could be adjusting the wrong way. This may cause problems with homing it as well.

Thanks guys.. Valuable discussion. I am going to try Gruff's idea for the IAC tonight and see what happens... It does seem like the thing is wide open... I will keep you posted later.

So I flipped it to Closed Loop always on and the idle was about 1200 now... And it stopped doing the 3400 RPM garbage.

But then I started tweaking the start, open and closed positions... It was running pretty good until I turned the car off. Then it wouldn't start (got about 600 RPM and studdered to a stall). I'm assuming the IAC is closed.

I've read through this a few times now:

http://www.msextra.com/doc/ms3/Idle_Control.html

And there is something that is just not clicking in my brain. Let's assume mine is NOT working in reverse.

The Start Value is the fully closed position? Let's say it's 160 steps.

The open steps... Would then have to be 160? or -160?

The close steps would then be 0 if it is starting in the closed position, correct?

That is how my brain is understanding it... Tell me if I am wrong... Please lol

I think you should pull the throttle body, it only takes a sec. Then use the test mode to get a better idea of what steps mean full open, full closed, etc.

Negative steps would be on the other side of a home position, which would not make sense in this case.

Since there is no encoder feedback on these, it must move all the way to one end first to get an idea of where it is... It does that when you key on. I think it will call that 0 steps, then go from there.

Don't quote me on that, but i think that should be right.

Originally posted by Tired2
I think you should pull the throttle body, it only takes a sec. Then use the test mode to get a better idea of what steps mean full open, full closed, etc. Negative steps would be on the other side of a home position, which would not make sense in this case. Since there is no encoder feedback on these, it must move all the way to one end first to get an idea of where it is... It does that when you key on. I think it will call that 0 steps, then go from there. Don't quote me on that, but i think that should be right.

Yeah, I am definitely going to take the TB off and run the valve test tomorrow.

So, on "key on" the number of steps, again lets say 160 for arguments sake, will bottom out the IAC. Now that it's closed and the car is on this position is considered "0" (I guess I quoted you anyways, haha) so then "open steps" could be 160... would that make closed 0? Because it takes 0 steps to close?

Can you post your latest .msq? Can you get a log of a cold start during warmup? As well as a log of the no start?

I don't want to say 100% sure this is how it is working, but this is my guess without looking more into it. I've been avoiding just going and reading the code for some reason...

Here is what I would try if I was coding a stepper...

key goes on>
What position am I at?
No clue.
Step back a lot from where I am so I bottom out.
Now I know this is home, (typically closed all the way).
I'll call this "0".
Move to 150, which is 150 steps from 0.
(pulse 150 steps forward)
Now I'm at 150.
Move to 160, which is 10 steps further from here
(pulse 10 steps forward)
Now I'm at 160.
Move to 100, which is 60 steps backward from here
(pulse 60 backward)

... etc.

So, if you were to assume the MS does this, since I'm 95% sure these are not absolute positioning motors, it would have to do something similar. The only absolute positioning you can really run is if you define a home position first and keep track of where you are. (Hence the trouble with "moving only" setting, if it creeps around when unpowered, you lose track of where it is and can't position it anymore. "Always on" though is not great because it keeps torque on the motor, which usually means excess heat.

I havent looked at the manual in over a year so im a little rusty as well on the steps it takes but that sounds pretty close. It mostly just takes some playing around and constant watching of the steps on the laptop to figure out the exact step range that works for you.
Some stepper motors will need more steps than others to close and vice versa. If the stepper motor gets warm then thats fine. but if it gets TOO hot to the touch then you shouldnt run always on, or just used a different brand stepper motor thatll work with always on.

Originally posted by Gruff511
Can you post your latest .msq? Can you get a log of a cold start during warmup? As well as a log of the no start?

I will log cold start, warm-up, and idle tonight when I get home from work.

So, I pulled my TB off tonight to switch the direction my valve moves... I got it now where higher step count is more open, which is higher RPM. It was backward before (just switch one coil polarity if this happens to you). 0 steps is all the way closed (after homing to the closed position). It bottoms out on the other side, all the way open at around 230 or so.

One thing I noticed:

From the manual:
"Start Value - This is the number of steps the valve will move (in the closing direction) on boot so that the rest of the code can rely on the start position being accurate."

I did not realized this was the home move distance parameter until tonight. I went ahead and set mine to 255 for now, and I'll drop it down to the max I end up letting the PID use in closed loop mode.

The value the motor "starts" with is from the cranking steps curve.

If "start value" is left too small, it may not fully reset back to 0 on next key on, and you'll be really confused probably.

It is too late to run mine tonight, but I'll give it a shot tomorrow night hopefully. On my drive last night, I underestimated how important a decently tuned idle is when driving around. Got some strange looks in my neighborhood doing 400 rpm and clunking to a stop at every stop sign.

I dont know if you guys have found these threads but they may give some more detail then whats just in the manual.

http://www.mx5nutz.com/forum/index.php?showtopic=46823

http://www.msextra.com/forums/viewtopic.php?f=91&t=30441&sid=84c8a3d2dc5c138aa7497cdf3ba86199

Thanks for the links, I had not seen those.

Did you get yours going?

I setup my idle almost 2 years ago using 'closed loop always on' after setting up PID settings I havent had any issues. It idles perfect still and mine is my daily driver as well as weekend race car. To be honest it idles steady and smoother than my stock ecu ever did. Once you get PID set it works great.

Gruff,

Would you mind posting your closed loop idle settings? Specifically I'm interested in seeing if your steps min, steps max, rpm closed, and rpm open somewhat matches mine.

Gains would be nice for comparison as well if you wouldn't mind.

After some playing tonight, I got idle working much better with open loop, and found at about 88 steps I get a pretty bad ~900 RPM idle. With it at 110 steps, I get around 1600rpm, so I used those values.

I need more time tuning the PID though. I think it is the fault of my P-Gain being too high, but when I burn the settings, rpm drops a bit, which sends steps up to like 115 from 94, and the RPM starts into the 2000s and I have to kill it. The next start will settle out, and loading up the motor by turning the wheel works decently well, but still drops a lot in RPM, which is why I jacked up the P gain in the first place.

Yea ill post my msq when i get to my computer but what is your idle timing set at? Sounds like your idle spark advance is high. Do you have your idle timing around 5-10 degrees like it should

Originally posted by Gruff511
Yea ill post my msq when i get to my computer but what is your idle timing set at? Sounds like your idle spark advance is high. Do you have your idle timing around 5-10 degrees like it should

I second that... Seeing your values will help me wrap my mind around it.

I started her up today... Took a log of the warm up.

Even after warmup with closed loop I couldn't get the idle down below 1200/1300...

Just for fun I tried open loop and ran a log. The motor was warm but I ran open loop with a start value of xxx. I then went to the end of the idle steps graph (because the car was warm and began lowering or raising the steps to see what happened. This was the only change throughout my entire process that actually began getting real time results...

As I lowered the final step at the end of the graph the RPMs began to decrease. At about 20/25 for the final parameter on the graph I was idling smooth at 900/1000 RPMs. I tweaked the idle advance a bit and it was even smoother...

Could this be right?? I even drove around a bit and it always settled back to a nice 900-1000 RPMs.

Here is my first log (closed loop warm up):

https://docs.google.com/file/d/0B22tb0XxveQ8Wmp5MkE4R2tTTzA/edit?usp=sharing

Here is my second log (open loop already warm):

https://docs.google.com/file/d/0B22tb0XxveQ8U3M1YTZvNUJadG8/edit?usp=sharing

I didn't get a warm up log yet with the open loop because the car had no time to cool down.. But here are the open loop settings and I will get a warm up log tomorrow ...

https://docs.google.com/file/d/0B22tb0XxveQ8VU1zOTZtUHlYWHM/edit?usp=sharing

I basically chose a number for the start value based on other runs and said 110. Then I lowered the last bin until I got an idle I liked. Then scaled the other bins up. Obviously this need sot be tweaked considering if I continued to scale up I would pass my start value... Or can I?

I don't know... Hopefully this will spark some ideas in you guys as to how I should approach this.

Thanks!

Here is my msq. If you can, open it with an example MS3 project as some settings may say different with an ms2 or ms1 project.

https://docs.google.com/file/d/0B28Mu9h7EyFWNUJsQzA3TUh2VHc/edit?usp=sharing

Originally posted by Gruff511
Here is my msq. If you can, open it with an example MS3 project as some settings may say different with an ms2 or ms1 project. https://docs.google.com/file/d/0B28Mu9h7EyFWNUJsQzA3TUh2VHc/edit?usp=sharing

Thanks Gruff, I'll have a look.

I got mine going pretty well with:
Open steps: 116
Closed steps: 88
Min steps for PID: 82
P-Gain: 175
I-Gain: 50

I am having a strange problem where the steps jump way up high (like 149) when I burn changes to the ECU, but then when CL idle kicks back in it is good to go... Just an annoyance though really.
http://www.msextra.com/forums/viewtopic.php?f=4&t=51319

Otherwise, it holds idle pretty well. Not perfect, but it does not die or really stumble much on loads from power steering and stuff. MSQ and log in the msextra post above if anyone wants a look. My fuel table is a mess though, and I'm running like 12.0 under load... still working on that.

Originally posted by Gruff511
Here is my msq. If you can, open it with an example MS3 project as some settings may say different with an ms2 or ms1 project. https://docs.google.com/file/d/0B28Mu9h7EyFWNUJsQzA3TUh2VHc/edit?usp=sharing

Dude,

Thanks for posting this! It helped out a ton! I started with your exact settings for IAC and they worked great. I had to tweak the idle advance a little. I took her out and around town a bit to see how the new settings were and she always returned to 900/1000 RPMs.... Awesome.

Now that it at least idles nicely I plan on using the tuning recommendations in the manual to dial in PID for my car. There are still events (Like A/C on) that cause a stumble in idle.

By the way... the organization of MS3 for tunerstudio is SO much nicer! I almost wish I went MS3 just for that! It just makes sense...

If you get the gslender mod of the ms2 code he has added all the ms3 menus.

No problem glad it all worked out for you

Originally posted by Gruff511
No problem glad it all worked out for you

Quick question that one of you may know... I got the idle pretty smooth now, it returns to a nice idle after driving, A/C, warm up, etc... Overall, running great.

The only thing I notice is that at the best idle I am at 30-31 steps. That yields a nice 900 RPMs and 13.7ish AFR. But after a few runs up and down the street I noticed it settles in at about 21-24 steps causing a slightly Rich idle (12.4-12.6 AFR)... What is controlling this? Why is it not returning to 30 steps?

That's normal because after driving and everything heats up the idle will need to reduce the amount of hot air entering the motor to get the idle down so it rudeces the steps closing off some air. Its richer because your idle falls into a different map%. You can fine tune it in the 'idle ve' feature

Originally posted by Gruff511
That's normal because after driving and everything heats up the idle will need to reduce the amount of hot air entering the motor to get the idle down so it rudeces the steps closing off some air. Its richer because your idle falls into a different map%. You can fine tune it in the 'idle ve' feature

That's kind of what I figured... It makes sense... And the fact that the valve closes more restricts the amount of air making it Rich...

Is the idle VE feature in MS2? I am having trouble finding information on it...

I think it is in your main VE table near the bottom left, just see what cells you are in when cold, then when hot.

I don't think MS2 has the idle VE table.

Originally posted by Tired2
I think it is in your main VE table near the bottom left, just see what cells you are in when cold, then when hot. I don't think MS2 has the idle VE table.

Gotcha... I will give it a try tonight. I was looking for something like an idle VE table or separate area...

I could be wrong, but I think that is a feature of MS3, a "finer" table just around Idle that it switches to under certain conditions. You won't need it anyway probably, I think I read that it was for trying to tune in cams that don't want to idle nicely.

Originally posted by Tired2
I could be wrong, but I think that is a feature of MS3, a "finer" table just around Idle that it switches to under certain conditions. You won't need it anyway probably, I think I read that it was for trying to tune in cams that don't want to idle nicely.

Yeah, I will just keep an eye on what bins I am in during which stages of idle. Thanks, I will let you guys know how it goes...

Originally posted by RfOeCnKdOeNr

Originally posted by Tired2 I could be wrong, but I think that is a feature of MS3, a "finer" table just around Idle that it switches to under certain conditions. You won't need it anyway probably, I think I read that it was for trying to tune in cams that don't want to idle nicely.

Yeah, I will just keep an eye on what bins I am in during which stages of idle. Thanks, I will let you guys know how it goes...

Quick update...

I kept and eye on the bins and tweaked the VE table a tad bit... Worked so far! The car has been running nice... Once it is warm I am much more confident driving it and it is feeling similar to stock ECU. Still fine tuning the VE table at the top end (haven't been able to make many ramp runs).

Quick question...

During warm up... The car idles smooth (no oscillation). Starts at around 1200/1300 RPM (depending on temperature) and slowly creeps down with coolant temps rising... Where I get into trouble is when the coolant temp kicks the idle advance. Let's say idle advance is set at 160. When the temp hits 160, the advance kicks in. And then the PID freaks out and oscillates. The only way I am able to get out of it is to tweak the PID gains (lower them slightly) until it stops oscillating.

Do I just keep repeating this every warmup cycle until it doesnt oscillate when idle advance kicks in or am I doing something wrong?

Check out around the 426 second mark on this log:

https://docs.google.com/file/d/0B22tb0XxveQ8QmhIYlQ3cTNaamM/edit?usp=sharing

Notice the drop in advance to 6 and then the oscillation begins...

Any ideas?

Its oscillating because IAC steps are moving around causing the rpm's to bounce around. Its basically not responding and recovering to the sudden spark advance drop.You need to dial in the PID settings more.