🔧 BUILD Turbo 3800 T56 Swapped G-body Grand Prix w/ Holley EFI

[bs009]

Took the car for a test drive today. Pretty sure my tape measure alignment is way better than the alignment I had done at a shop in Michigan because the bump steer is completely gone and the tires no longer feel like they're fighting each other during turns. Might still get an alignment done, but I think I can probably get away without having a pro do it for power tour.

First time it's been outside in almost 2 months

Transmission works flawlessly, clutch seems like it's making a little bit of noise though at idle, kind of a clicking sound every once in a while, not consistent at all so that's strange. Running the car at speed at low RPM does shake the car a bit, but it has no problem accelerating in 5th gear at 60 mph around 1400 rpm. The rear 2.41:1 gear ratio is definitely the worst part about the car right now though. Too bad my new axle didn't get here in time because that would've solved that problem too.
I'm so glad I don't have to deal with that slippery unpredictable automatic anymore.

Going to do some tweaks to the tune because with a manual trans the pops and burbles I put in there are excessive AF


Also the WOT box works pretty well, but I am noticing an oddity with it all. When you kill power to the ICM the PCM doesn't seem to get an RPM signal anymore and won't fire the injectors.
I'm realizing that this must only be an issue when the WOT box is trying to no-lift shift because it completely kills power to the ICM instead of pulsing it. Sometimes it worked flawlessly and sometimes it would kill power to the ICM and cause a huge backfire once the PCM realized the engine was running. I'm thinking that if it's pulsing power the ICM it has just enough to give the PCM enough of an RPM signal to keep the injectors firing.

Watch the tachometer in the video to really see what I'm saying. Fair warning: I sound like a giddy school girl in the video




Wonder if I can run another ICM in parallel for the sole purpose of sending an RPM signal to the PCM while the other ICM is getting it's 12v signal pulsed/killed by the WOT box. Wonder if the crank/cam sensors will care or not?
I'm thinking I would just run every wire in parallel except for the power wire. New ICM would just act as a dummy without actually firing a coil and the old ICM would be there to fire the cylinders.

In that run there was one point where the RPMs hit 6,500 before the PCM caught on (the limit in the tune is 6200 rpm). Definitely want to make sure the PCM knows what's going on so I don't over rev the motor if possible.

 

[Turbocharged400sbc]

Yeah that's the problem with the system is that when you kill power to the ICM it kills the 7x composite signal that the ICM generates from the 3x and 18x crank sensors and the cam sensor. The PCM requires that signal to know which cylinder is at TDC and what RPM it's spinning at.
Until the p c m gets one full cam Revolution [2 crank revolutions] to know which cylinder is cylinder one and its at TDC.... your engine fires in batch fire mode..... before it switches over to sequential so yes you'd be wreaking havoc with the icm and it's one reason why I never went with that route.

The ignition coils are fed constant power through the icm power pin... and the ICM has current limited ground switch controls of the coil primaries.

You would have to do something similar to my ICM stacker set up where you can supply the coils with their own separate power from the icm feed, that way the wide open throttle box would actually only kill the coils.....

Keep in mind the other main issue with this is if it kills power to a coil that is currently energizing and it kills the power little earlier than what the PCM would you end up with a cylinder that fires at 40 degrees before top dead center when it really should be firing at 12 degrees to keep you from pushing the rod bearings into the oil pan.

Yet another reason why I do not like the wide open throttle box systems.

It's one reason why I'm going with the ls X style coils and the adapter module..... that way at least you can do the math for coil frequency at you're limiting RPM and make sure that the switching frequency of the logic kill circuit allows it to be something more like a soft-touch rev limiter

 

[Turbocharged400sbc]

I always thought about using Arduino code with a controller to be able to do a frequency and duty cycle type controller to allow tuning for the off time to be for a full firing event.

 

[sktchy]

I've wondered about these WOT boxes, kinda glad somebody else is tryin to figure out how to make it work right tho. I really think down the road I'd like to scrap the po4 altogether if I stay manual.

This thing is still awesome btw

 

[Turbocharged400sbc]

Some of the more complex Arduino type computers have enough inputs and outputs I could use it to monitor actual engine RPM and track tdc1 via inputs and use the six pwm outputs to individually kill the logic signal 2 the six LS coil packs in a truly random non-sequential method that prevents any coil energizing...

Which is what I gather the Lingenfelter units do with the LS V8 systems.

Something like that would be easy to set conditions and parameters for when to engage RPM limiting and even taking over control and retarding timing for anti lag.

I actually found some firmware that would let me turn a duet 3D printer board into a rudimentary EFI system lol

 

[bs009]

Okay wow this is getting complicated fast then. Hopefully I haven't damaged the motor then or anything. I'm down to try modifying an ICM to divorce power for the coils from the rest of the module, but if there's still a possibility of it firing one with 40* of advance like that then that's pretty concerning. I'm surprised the WOT box doesn't do a lot of that calculation on its own too like the Lingenfelter box does.
I'll try and see if I can get them to describe how it cuts power to the coils

I don't think I've ever used an Arduino before but I'd be down to try it. I studied computer science with a lot of ECE classes so I should be able to figure it out.
I've also heard megasquirt 2 could be used to basically replace the ICM too for something like this, Never looked into it myself though.

I'm planning on converting the car to Holley HP EFI over the winter though so I don't want to put a huge amount of effort into setting this up. Holley EFI should have 2 step built into it already from what I've heard and I'd just run LS coils with it at that point.

 

[sktchy]

Okay wow this is getting complicated fast then. Hopefully I haven't damaged the motor then or anything. I'm down to try modifying an ICM to divorce power for the coils from the rest of the module, but if there's still a possibility of it firing one with 40* of advance like that then that's pretty concerning. I'm surprised the WOT box doesn't do a lot of that calculation on its own too like the Lingenfelter box does.
I'll try and see if I can get them to describe how it cuts power to the coils

I don't think I've ever used an Arduino before but I'd be down to try it. I studied computer science with a lot of ECE classes so I should be able to figure it out.
I've also heard megasquirt 2 could be used to basically replace the ICM too for something like this, Never looked into it myself though.

I'm planning on converting the car to Holley HP EFI over the winter though so I don't want to put a huge amount of effort into setting this up. Holley EFI should have 2 step built into it already from what I've heard and I'd just run LS coils with it at that point.

From everything I've seen anyway. After spending some time in e38s and e92s our pcm is so limited and dumb by comparison. I've seen megasquirt run Mac valves for boost control with my own eyes and speed density would also be a real nice option to have. I'm assuming the holley would be at least on par.

 

[Turbocharged400sbc]

I'm screwed though I need transaxle controls

 

[bs009]

For sure, Holley is basically the state of the art from what I've seen so far. Seems like you can do pretty much anything you want to with a Holley Dominator except charge the battery with the ECM connected

HP EFI is a bit more limited than the Dominator, but a bit cheaper. It has less inputs/outputs to play with and no trans controller capability, but I think it's going to be sufficient for what I want to do with the car for now and upgrading isn't a big deal between the two if I want to do it later on.
I'm going to talk to the Holley folks at power tour if they're though and pick their brain about the systems to see if they can push me over the edge to switching.

Holley is really user friendly and it can self-learn fueling with its wideband sensor so lopey cams are a lot easier to work with. Running E85 and open loop for tuning a big cam on the stock PCM sounds like a nightmare to me for the way I use the car with all of these road trips. I've been waiting to do a cam swap until I switch over because of that.
The HP EFI lets you program 4 generic inputs and 4 pwm/switching outputs. The Dominator lets you run something like 40 inputs and outputs so it's a monster. The Dominator can also control a 4l60e/4l80e and a few other transmissions too.
Since I'm using a manual trans the HP EFI is going to be enough for me too so I can save a few hundred dollars.

HolleyEFI HP or Dominator EFI System – Which is best for me? - Havoc Speed

This guide assumes you have an existing, fundamental understanding of the general Holley EFI system. Trying to determine which system your build is going need? Have questions? We’ll do our best here to guide you through the differences, and technical specs to aid your decision. This guide will...

www.havocspeed.com

Megasquirt 3 is cheaper but it's a lot more work from what I've heard so far. The Holley seems like it's going to be a lot easier to work with, but maybe I need to go check out megasquirt and play with their software to see what it can do.

I'm screwed though I need transaxle controls

Wonder if you could fool the dominator into controlling the 4t65e as if it were a 4l60e with some tweaks? Suppose there are also stand-alone controllers out there too

 

[bs009]

Oh I also learned just now that my ICM wasn't plugged in all the way during my test run so that probably wasn't helping the way the car ran.
I think I'm going to still try to run 2 ICMs though or disect one to split power for the ICM and coils to see how that works.

N2MB emailed me back and claimed the WOT Box won't interrupt a coil that's firing, but I doubt they actually know that that's true for every engine out there. All it asks for in the tune for the box is the number of cylinders and where the RPM signal is coming from.

 

[sktchy]

From the little I've messed with tunerstudio which is what I've seen all the squirt guys run here its pretty straightforward. Not sure if there's another software for it or not. But I've also had guys in here renting time on the rollers taking videos with their phones of what stuffs doing and adjusting that way instead of actually datalogging it and I know a guy with a 9 second ls that eyeballs his wideband and dials it in from there.

Reminds me how much I still suck every time I get somethin like that in here and I'm over here rolling through hpt watching for kr and those guys don't even bother putting knock sensors in ?

As far as the big cam if your drive by cable I think some of it will be in your throttle blade. I've been doin some experimenting with mine and base timing trying to smooth out some annoyances I get once in a while missing the airflow tables from an e38 like crazy lol

 

[bs009]

Looking at my service manual now, it's looking like I probably shouldn't just be killing power to the ICM while the engine is running so it looks like I'm going to have to dissect an ICM to try to get the WOT box to work here.
ICM apparently plays a much larger role than I had thought it did before and if I have two ICMs seeing different things because one isn't powered or is restarting the PCM is going to get weird signals

 

[v6buicks]

Looking at my service manual now, it's looking like I probably shouldn't just be killing power to the ICM while the engine is running so it looks like I'm going to have to dissect an ICM to try to get the WOT box to work here.
ICM apparently plays a much larger role than I had thought it did before and if I have two ICMs seeing different things because one isn't powered or is restarting the PCM is going to get weird signals

I'll be really interested to see what you come up with since I wanted a WOT box as well. This would be another good opportunity for a write-up!

 

[bs009]

Okay so it's looking like separating the ICM power from the coil power is going to be really simple, just difficult execute.

I dissected a spare ICM I had lying around and just mapped all of the pins to the large input/output traces above the (let's call it) logic circuit.

1 pin on each coil seems to have constant power all of the time, the other goes to the large chips on the right side of the board. I had to confirm that though so I went a little deeper.

I could keep going but I'm pretty confident I'll find a circuit like this:

So it's looking like the simplest way to do this would be to mount the coils somewhere else where you can feed each one the pulsed power from the WOT box, and run a wire to each coil from the ICM for the other post on the coil that actually controls it. Would need to run good wire and not go to far though to make sure the extra wire isn't causing a signal loss.

 

[bs009]

Or another idea would be to just cut the lead into the circuit board for the power signal, then solder in a wire that runs outside the ICM directly to the ignition voltage source before the WOT box. Would be a cleaner install but would still mean modifying an ICM and cutting the lead could be kinda iffy. It would be the quickest solution though

 

[v6buicks]

I'm happy to see somebody else experimenting with this before I end up drawing the roadmap myself. lol

I think your second idea could be done really cleanly after refilling the space with potting compound. ?‍ Congrats on getting the transmission conversion done by the way!

 

[Turbocharged400sbc]

ICM "stacking" aka multiple ICM's in parallel...

so while I was scratching my head over my radiator fan shroud on the Toyota... I started thinking about the one gentleman who joined up... I completely forget his screen name... who wired ICM's in parallel.... now my inital thoughts were to use our ICM as a logic driver for LSx coils ran in...

www.thechicagogarage.com

 

[Turbocharged400sbc]

And yes you are correct all of the coils are fed battery voltage. The ground side of each coil is controlled by a current limiting chip on the circuit board

 

[bs009]

I've considered ICM stacking, would love to be able to run less spark and maybe have a bit better idle, but if I'm jumping to LS coils down the road I'm not sure if it's worth it. Might go order some aluminum though and play with it after power tour. It would mean making this same modification to both ICMs too unless I run the coils over somewhere else remotely with a diy bracket which does sound pretty tempting now that I'm thinking about it

 

[Turbocharged400sbc]

I figured it would also give you an idea on how to make it stack up the way you want it to.
I gots betta ideas

V6 DIS to LSx coil system

 

[bs009]

I figured it would also give you an idea on how to make it stack up the way you want it to.
I gots betta ideas

That looks pretty dang good! I haven't been able to find anything useful about swapping over to the LS coils with our ICM but I think I could probably figure it out now that I've got one disassembled haha!
things to think about for sure

 

[bs009]

I now pronounce the ICM and the coil power signals divorced!

RTV is curing right now

Also added a 1 pin harness for the ICM and the car so I can actually remove it easily if I want to. I even ran a fresh 18awg power wire for the coils separate from the ICM so maybe I can get a bit more juice out of it if it's not already limited by the ICM.

Should actually work better this time this way

 

[v6buicks]

You're making it look easy. If I get some free time, I know what I'm doing to one of my spare ICMs.

 

[v6buicks]

So uhhh... It's been a couple days...

How's that ICM?

 

[bs009]

So I haven't actually been able to test it yet, been working on brakes, wheel studs, and cleanup the last few days. New brakes are on now for good, but have been having some trouble bleeding them so I've been chasing that the last few days.

I did at least fire it up and the car does still run now with the new ICM but I haven't been able to to test out the two step since the car's stuck in the garage right now. Don't really want to set the garage on fire ?
I have no doubt it's going to work, but I might still add a relay to retard all of the timing when in anti-lag just in case it might be able to fire 20* advanced like turbocharged400sbc mentioned

Leaving early in the morning tomorrow. A bit worried since I've done so much work to it and have only driven it around the block so far. Been trying to make sure I haven't forgotten anything and have been combing over things lately. Feels like a lot has changed even though it really hasn't been that much but it has a long drive ahead of it tomorrow.
New wheels are also here now. Spent all of last night ceramic coating them. Rear end didn't make it in time though as anticipated, but came really close