2002 Megasquirt EFI Turbo tuning questions

[Nateb603]

Hi Guys, 

I am finally getting my ugly duckling put together and looking for some tuning advice and help getting my megasquirt project sorted out. This is my first experience with megasquirt so I am trying to learn it as I go. The car will start and somewhat idle but I do not want to try to drive it until the idle is steady and have a base tune figured out. I am starting from scratch so any advice or time saving tips would be greatly appreciated. It runs rough below about 1000 rpm and I'm trying to figure out how the idle tuning should be approached. The Idle vacuum also seems low at 13-16 in. and wants to stall when warmed up. I have adjusted the valves and verified the cam/ crank timing. I am wondering what idle controller settings most people are using and if I need to have the throttle cracked at all.

 These are the specs of the engine 

 

IE 292 billet cam

Bosch 24lb injectors

318i intake manifold

02 again stepper motor adaptor with jeep stepper

02again  crank trigger wheel and VR sensor mount

Ford EDIS

E30 325 throttle body 

Nissan Z31 throttle position sensor and 02again adaptor

fuel labs AFPR set at 43lbs

in-tank electric fuel pump

Holset HX30W turbo with turbosmart internal wastegate

Ebay 50mm BOV

front mount intercooler

megasquirt II v3 with msextra 3.4.4

original green board map daddy

Ross Racing pistons with stock rods

MLS headgasket and ARP headstuds

AEM wideband

 

 

If anyone can post pictures of their general settings, with a similar set up, that would be great.  I know idle tuning greatly relies on fueling and timing but I would like to have all the  stepper motor setting figured out before I start adjusting fuel and timing. Thanks

 

-Nate

 

 

 

 

 

 

 

 

[irdave]

Hi Nate.  It's simple, but there's so much to it.

 

You're running throttle by wire?

 

There'll be a couple guys on here pretty quick to be able to help you much better than I, so in the mean time we'll bump it so they see it.  @AustrianVespaGuy@jimk

[jimk]

I don't have anything to help with the Megasquirt, the program operation is not the same as other manufacturers.   Watch some on-line webinars from Motec, watch to learn the basics, not their software.

 

In general the non-boosted operation (manifold pressure no greater than atmospheric) will be very similar to other natural aspirated (NA) tunes.  So find some starting parameters from another M10 engine. 

 

Tune the non-boosted area of the fuel and ignition advance maps first and move into boosted in small steps.  Things can go to hell in a hurry in boost and it isn't very sympathetic to piston damage.

 

The injector size you listed is 1.2 times those for a NA M10, so watch the duty of the injectors as you do the boost tuning.  Has the injector size been checked for the amount of power you are anticipating?  There are on line calculators.  MSInjector program for a modest 5psig boost/intercolled shows 260cc/min minimum size.

Edited July 12, 2021 by jimk

[AustrianVespaGuy]

YAY MEGASQUIRT!!! Thanks for the tag @irdave! And no worries @Nateb603, I gotcha covered!  And everything @jimk is correct, first and foremost being those injectors are probably too small for a forced induction build.  But now to the fun stuff!

 - FIRST: Is your Req_Fuel correct? For an M10 w/24# injectors should be 13.3, but double check before we end up chasing our tail!

 - You said you checked timing, but have you REALLY checked timing? This means a timing light! Use the 'Fixed Advance' option in the ignition settings menu to set the timing to a known, fixed value, and use the timing light to make sure that's ACTUALLY when it's firing (pretty easy to see with that 36-tooth wheel, since each tooth = 10 degrees ) Idle is probably shit at 0deg BTDC, but try 10 and 20 and make sure they're both right/change as expected. You can then adjust the offset as needed to get it spot on.  In my experience, the M10 wants idle advance to be in the mid-teens.  Start with 15 deg, keep it STEADY in the idle cells of the table, and play around one degree at a time if you want.

 - Now for the amount of air at idle, those steppers can sometimes be a bitch.  Pull the adapter block off the manifold but leave the stepper in it.  Power on the car and play with the stepper settings until you are SURE you know A.) it's moving and B.) which way it moves, i.e. whether *MORE* steps = more closed or more open.  Finally, you'll want to get an idea of the total number of steps of its range (I think mine is on the order of 170-180) between fully closed and fully open.  THEN you can proceed with making the warmup curve so that it runs to a partially-open position for cold cranking, then tapers down to closed by the time it's warmed up.

 - Next, have you confirmed that your AEM is working properly, and calibrated so that Megasquirt reads it properly?  Once your confident there and have a decent idle, it's time for the best part of all: DATALOGS!!! ? Take some, and post/pm them to me along with your tune, and I'll teach you what to look for in them so that we can start tuning your VE table.  

[Dudeland]

In the FWIW category, you can buy the auto learn function in the megasquirt tuning software.  So once you have the basics worked out, and a decent timing curve picked, you can set your desired AFR and see a fuel map begin to form.  Having had to tune my sniper efi, sharing data logs is a great way to get some seasoned eyes on what is happening.  

 

Being a new build it may make sense to test for vacuum leaks as well.  

 

What boost controller are you using?  Have you ran it enough to start to produce boost? 

 

A detailed set of pictures of your setup may also help, as those folks with turbo experience may catch something that can save you heartache, money and time. 

 

I am looking forward to seeing more. 

 

[irdave]

To add; I had so many dumb things I did wrong setting my (non- MS) ECU...  I didn't have the signal ground wire from the AEM WBO2 connected to anything, let alone the ECU ground, I had my stepper control wires for the IAV connected to the wrong outputs so it would for a minute then not, my VDO oil temp gauge was feeding voltage to the sensor to measure resistance but I also had that wire running to the ECU for oil temp and that voltage was causing problems, many more.  So many little things to sort.

[Nateb603]

@irdave  @jimk @AustrianVespaGuy @Dudeland

 

Thank you guys for your replies and your help. I will try to answer some of your questions as I'm sure ill have a few more of my own. The 24lb injectors I have installed right now are definitely too small for the boost range, I was told larger injectors are very difficult to get to idle correctly because of the smaller amount of on pulse you have. I mainly installed them just to see if I could get it running and see if I had other mechanical issue to sort out. I have a set of 42lb bosch that I will be swapping back in. I realize there is no point to start trying to tune this, only to have to change everything again when I replace the injectors. 

 

As for the timing I have not verified the timing with a light because I was unsure if it was possible with the EDIS and waste spark. I know the coil will be firing twice as much as if checking a typical distributor setup. I added a paint mark to the tooth that is over the pointer at TDC so I will try to verify with the timing light. Did any of you have to add in a timing offset to correct the actual spark timing, and would this be adjusted in tuner studio via the "trigger wizard" ?

 

I have removed my stepper adapter and checked the steps with it in test mode - I believe mine is about the same. I will pull it off and test it again because I don't remember the exact amount of steps and need to write it down again. I believe it was 180 open to 50 closed.  I was following Vespaguys great write up ( thank you for making it)  but the thing I'm still confused about is setting the stepper to close completely when warmed up? the table stops at about 175 degrees and I just wondering how it is supposed to idle if the valve is fully closed and cutting off the air supply to the engine. Do I need to have the throttle blade slightly cracked? I assumed by using the stepper I would have better airflow control so I just didn't understand closing it when warmed up. Is there another table the IAC uses when warmed or are you using closed loop control when warm?

 

I believe the AEM wideband is reading correctly. the model I have is not listed in tuner studio so I have it set for custom linear and used the table provided with the gauge. The MS and the gauge seem to match but its hard to glance back and forth when the AFRs are bouncing a bit. 

 

it does not have a boost controller and the turbo is set at 14 psi based only on the spring of the wastegate. I have not tried to rev it or get into the boost areas at all. 

 

My other question is about the map settings. I am using the original green circuit board map daddy and from the original documentation it shows a barometric correction for total vacuum % of 147 and rate % of -47.  Is this correct?  I tried running it with the corrections at 0 and it seemed to run better. 

 

my other concern is the low vacuum at idle. The car doesn't want to idle unless I have a tiny bit of pressure on the throttle and gets rough at lower rpms. seems to run in the 14-16 in hg. or 50ish kpa. I was expecting to see something closer to 35 kpa at lower rpms.  I checked my valve timing and it seems to be correct and adjusted the valves to .008" per the manual. I do not believe there is a vacuum leak but I could be wrong. I am assuming this could also be a timing issue if I do need to change the timing offset.

 

My next steps will be reinstalling the 42lb injectors, rechecking my IAC steps, and verifying the timing is correct.  Thank you all so much for the assistance. I added some current pictures of the engine and step up. please excuse my ugly engine bay and all the dust, the car is pretty rough still and has not ran in a very long time. 

 

 

  

  

 

 

 

 

 

Edited July 13, 2021 by Nateb603

[AustrianVespaGuy]

6 hours ago, Nateb603 said:

Do I need to have the throttle blade slightly cracked?

Yes. When the car is fully warmed up, you want the throttle plate stop screw set to hold it cracked open to your desired idle speed.  The stepper motor valve is then only used to let in additional air during warmup, but you want it closed and out of the equation by the time you're up to temperature. UNLESS you're running air conditioning, in which case you can also use it to bring up the idle speed when the compressor is running. But for the general sake of simplicity, just use it for warmup only! I close mine down by 170.

6 hours ago, Nateb603 said:

I am using the original green circuit board map daddy and from the original documentation it shows a barometric correction for total vacuum % of 147 and rate % of -47.  Is this correct?

If you really have the old/original green board (and both sensors have barbs on them) and not the new bluish board (where the baro sensor has no barb), then yes those weird 147/-47 settings should be correct, BUT, since you're running Extra firmware, you then need to zero out the Barometric Correction curve (which is usually defaulted to 100). OR, you can zero out the total vacuum and rate in the calibration, and then set up the correction curve instead.  But if you have the 147/-47 in the calibration and 100 in the correction curve, yeah that'll make things *not* right. (There's a super tiny text box in the bottom of the curve window that explains this).

 

6 hours ago, Nateb603 said:

As for the timing I have not verified the timing with a light because I was unsure if it was possible with the EDIS and waste spark. I know the coil will be firing twice as much as if checking a typical distributor setup.

 

Even though it's firing twice, that's still only twice per 720 degrees of crank revolution (compression and exhaust strokes), so it's still always firing at the same time every 360deg of crank revolution and so works just fine with a timing light.  Dizzy fires a given plug just once every OTHER crank revolution.  And it's very possible that you'll need to adjust the trigger angle offset some depending exactly how your crank, wheel, and sensor positions all stack up.  That's why it's important to do this exercise, so that when your MS tells EDIS to fire at say 20deg BTDC, it then fires when the crank is ACTUALLY AT 20 deg BTDC, and not 13deg (<- 7deg offset)!

[jimk]

7 hours ago, Nateb603 said:

I realize there is no point to start trying to tune this, only to have to change everything again when I replace the injectors. 

If MS has to be retuned with an injector change, then isn't keeping up with technology.  I made a size change with the Haltech and it only needed the parameters for the injector size and dead time changed.  No retuning.

 

7 hours ago, Nateb603 said:

The MS and the gauge seem to match but its hard to glance back and forth when the AFRs are bouncing a bit. 

If MS displays the % short term correction, then it is the parameter to watch.  It tells how close the tune is.

 

7 hours ago, Nateb603 said:

I was expecting to see something closer to 35 kpa at lower rpms. 

45-50 kPa is normal.  35 kPa is reaced only on hard deceleration on closed throttle.

 

7 hours ago, Nateb603 said:

rechecking my IAC steps

Doesn't MS IAC control in closed loop?  It should control idle speed to the target rpm.

[AustrianVespaGuy]

2 hours ago, jimk said:

I made a size change with the Haltech and it only needed the parameters for the injector size and dead time changed.  No retuning.

Yep, it's as easy as just changing the size so that it recalculates req_fuel for the new injectors, and shouldn't need to mess with VE table.

 

2 hours ago, jimk said:

Doesn't MS IAC control in closed loop?  It should control idle speed to the target rpm.

It certainly can do closed loop, but I still recommend that as a second step after getting open loop working properly and a decent tune first.  Too often people will have some VE or timing variations in the idle areas of their tables, and between those and the closed loop stepper trying to control air, the idle ends up hunting because everything is 'changing.'  Also, that Jeep stepper for a 4.0L is on the big side for the half-sized M10, so even a small number of steps can sometimes make a pretty big change in idle speed.  It'll all still work just fine of course, I just think it's a lot to tackle for someone just starting out with an EFI conversion, whereas the open loop method is a bit simpler to get going with, and then come back to closed loop later after they've gotten the hang of things.  Same goes for starting with time-based accel enrichment before trying to tackle wall-wetting algorithms!

[jimk]

7 minutes ago, AustrianVespaGuy said:

Too often people will have some VE or timing variations in the idle areas of their tables, and between those and the closed loop stepper trying to control air, the idle ends up hunting because everything is 'changing.' 

I use a zero throttle ignition map to stop that.

[tzei]

Some notes about on the hardware-side, if don't mind...

 

Ad some heat shielding under edis and fpr. I'd chage fpr location more to the cold side of the engine. Both of 'em thanks you later.

I'd route Turbo's oil feed a bit differently - a way easier to get to valve train to check things etc.

ebay bow may do it's job or could struggle (leak, jam) at some point so keep that in mind when tuning or getting into head scratch situations. My bow is before cooler - i like the idea of dumping hot air out of the system better than dumping cold air out. 

[AustrianVespaGuy]

3 hours ago, jimk said:

I use a zero throttle ignition map to stop that.

Ahh, that's clever! I suppose it's just based on a threshold of TPS% < X => use Table #2?  I know Megasquirt supports tableswitching, but I always thought it was just for Nitrous or Ethanol or whatever, so I'll need to play around some with this idea!  Can you do it for fuel too, or just timing?

Then again, how does that end up different from just dialing in the right values in the main tables? I mean, when you're at 800-1000 RPM, 0 throttle, does it really matter whether your 45VE and 15deg is in Table1 or Table2? I think I must still be missing something here. Always more to learn!

[jimk]

1 hour ago, AustrianVespaGuy said:

Can you do it for fuel too, or just timing?

Fuel also with a fuel zero throttle map, but self tuning finds the best values in the fuel table.

1 hour ago, AustrianVespaGuy said:

Then again, how does that end up different from just dialing in the right values in the main tables?

Saves a lot of manual trial and error.   Recall, this is a Haltech Elite ECU.

 If the main ignition table is using a MAP sensor then the manifold pressure bounces around a little and the advance will move around interpolating from the adjacent cell values..  The zero throttle table can have the same advance over several rpm cells for stable timing.

Begin with typical values in a base IAC position table, from then on, the IAC learns the correct IAC values to achieve the target rpm and saves the correction in a long term correction table.  The long term corrections can be incorporated into the base table with a software click.   I use a two axis table, coolant temperature and oil temperature axis. 

[Nateb603]

13 hours ago, tzei said:

Ad some heat shielding under edis and fpr. I'd chage fpr location more to the cold side of the engine. Both of 'em thanks you later.

I'd route Turbo's oil feed a bit differently - a way easier to get to valve train to check things etc.

ebay bow may do it's job or could struggle (leak, jam) at some point so keep that in mind when tuning or getting into head scratch situations. My bow is before cooler - i like the idea of dumping hot air out of the system better than dumping cold air out. 

I will try to add some heat shielding. I put the FPR on the turbo side just because of space constraints.  it sticks out too far with the pressure gauge to swap it to the other side because of the manifold and the plug wires are not long enough to reach really anywhere besides where the coil pack is mounted. I will be replacing the bov. I got the current one for free from a friend but have not decided on what I will replace it with. I am leaning towards the new style synapse bov since it should bolt up to the same location. I have just heard over the years that placement on the cold side towards the TB is best but there are many debates on placement and again it came down to space for me. The charge pipe from turbo to IC is very short and fitting a bov to that pipe would be difficult. The oil feed to the turbo only requires loosening both AN fittings and removing the valve cover normally. maybe takes an additional 60 seconds.   

19 hours ago, AustrianVespaGuy said:

Yes. When the car is fully warmed up, you want the throttle plate stop screw set to hold it cracked open to your desired idle speed. 

I cracked the throttle body to 4% and recalibrated the TP. I'm sure Ill need to adjust the throttle opening but haven't been able to run the car again just yet.

 

I set the map corrections back from the original documentation to 147% and -47 and set the Baro correction curve to 0 across the board.  

 

I reinstalled the 42lb "green giant" injectors and reset the req_fuel.

 

I removed the stepper with the adapter still on it.  For the homing steps I got a number of 160 fully closed which showed 0 on the gauge and the valve step number of 140 to fully open.  This is the setting I'm getting confused on and I'm still not sure what the tuner studio setting should look like. I have the homing setting to fully closed. Does this then mean I have 140 steps from fully open to closed on the curve?  

 

I have read that the closed loop idle control is very difficult if the base idle tuning is not correct and can complicate things, so I will be leaving it in the open loop and trying to make the controls simple. The main thing I needed to know was that I need the throttle cracked a bit if closing the IAC when warm.  

 

I also tried to check my timing but only had a few minutes to do it after work and want to recheck it again. I was trying to look at the mark I made on the trigger wheel but should have verified it with the TDC mark on the flywheel also.

I watch my kid in the mornings and work nights so I don't really have any free time on weekdays to run the car. I will double check everything again this weekend when I actually have a chance to let it run and warm up. I should have some time to devote to tuning it this Sunday. I mainly just wanted to make sure I have the basic configuration correct.

 

Thank you all for your advice and help. I wish I had more time to devote to this project but life is hectic and I'm just excited to see it start at this point. 

 

Edited July 14, 2021 by Nateb603