Help me finish my race car

Many of you are following my race car build and I'm almost at a standstill over a computer problem, and I need your help to move the build forward. The engine is a 2013 Ecotec LHU out of a Buick Regal Turbo that came with an automatic transmission but I now run a manual trans (BW Super T-10). The wiring harness was modified by enginewiring.com to eliminate the Body Control Module, Transmission Control Module and Fuel Pump Control Module. Enginewiring.com are friendly with Trifecta and eliminated these things with consultation from Trifecta. The engine starts and runs but only allows full throttle briefly before the ECM ruins the fun and pulls throttle back to 30%. I am told that this is because the ECM believes that the car is in Park or Neutral. I have a Vehicle Speed Sensor going through a Dakota Digital Box that can convert the signal to the most common kind of signals that an ECM wants to see. I am using HP Tuners to fine tune the engine but HPT is unable to get around this VSS problem. Trifecta apparently has a way of hacking the operating system of the ECM but they lock out any possibility of fine tuning it myself and charge more money every time they change the tune. So far I have heard nothing but bad things about their tune for this particular engine and many report that Trifecta just stops responding to requests for improvements. I have attempted to contact them but so far have zero response. There are rumors of people with special talents that know how to hack the operating system but so far they fall into three categories: 1)I know how, it will cost you a bunch of money, no guarantees, send me money first 2)I know how, I have signed non-disclosure agreements and can't help you 3)I know how, I have no time for you So, The GRM community is huge, knowledgeable and I know would want to see my race car completed. Can someone please help me? Here's the thread for my build: http://grassrootsmotorsports.com/forum/grm/scca-emod-mgb-gt-v12-build/26822/page1/

I can't really help your specific problem, but a few notes...

First- find second opinion on why the throttle is cutting back. While there are totally different limits on a manual vs. auto trans car, most of them are engine speed limits- so an auto in neutral would go to 50% redline or something like that. It's possible that it sees this issue, but it's also possible that it's something deeper- as in an electronic throttle safety item- that the engine isn't making more torque than it should.

Second- if it's the simple fix, I wonder if GM has a simple set of switches in their code to change the transmission settings. Although, by 2013, that code would go pretty deep. For sure, there's a lot of torque control code that goes really deep.

Unless there's a disgruntled former GM employee from 2012 out there, it may be really hard to get a solid fix on via the GM module.

with basically zero knowledge about things of this depth, if the ecm thinks its in park/neutral would it be possible to hook up a neutral safety switch and manually turn the switch to the 'drive' position to make the ecm happy?

fIND PIN ON PLUG FOR the neutral/park sensor. find out what signal it generates. (probably constant on or grounded when in park). Replicate signal.

Easiest and cheapest I bet.

I would find a willing owner of a donor car, and borrow an escilloscope, and check what signal it's being sent. It might be simple, and you can then replicate it. It might just be on/off. With a car and a scope, you can find out in a few minutes.

Here's the thing about just suggesting making a neutral/park signal and that should fix it.

In those two states, the engine should not make any torque. No reason to. The system is smart enough to know that. So when it does make more, it will go into limp mode, aka- over torque protection mode.

Good thing to try, but be ready to go a lot deeper.

edit- this isn't a fun police error, it's a electronic throttle safety device- preventing error states that would make the car go off on it's own- which is bad. ETC will always check and make sure it's not doing more than it's supposed to (the idea of preventing a throttle cable from sticking). It's got back up systems and multiple sensors just for this reason (just like modern electronic controlled planes).

Another suggestion, are you using the throttle pedal (or throttle position sensor) from the donor vehicle? Because those are designed to output two separate signals (on two separate wires) to measure throttle position. This is required by FMVSS regulations for double redundancy. The thing is, different manufacturers use different "spacing" between those signals. One may have the signals travel from 0-5V and 0-4V respectively, while another may use 0-5V and 0-2.5V. If the ECU doesn't see the proper spacing between the two TPS signals it will trigger and error and limp home mode.

Good luck.

In reply to alfadriver:

Perhaps we're not speaking the same language. I was advocating for a signal to convince the ECU that the transmission was in drive. I imagine this could be easily toggled with the key in the future.

If that is plausible, the engine torque should satisfy the ECU, right?

tuna55 wrote: In reply to alfadriver: Perhaps we're not speaking the same language. I was advocating for a signal to convince the ECU that the transmission was in drive. I imagine this could be easily toggled with the key in the future. If that is plausible, the engine torque should satisfy the ECU, right?

Maybe, but the ECU would also know what gear it should be in via feedback from the trans. So not seeing info from the trans would cause a new set of problems. (and all the trans actuators would be in error, which probably would also require some torque truncation to make sure it stays alive.)

So from what I understand, you have two problems: The ECM isn't getting the gear sensor or VSS signals it expects. You need a device to fake these signals. First you need to find out the communications protocols of these devices and what signal they'll send under "acceptable" conditions (so what the gear sensor sends when the car is in drive, and what the VSS sends at a vehicle speed that won't cause the ECM to give trouble).

alfadriver wrote:

tuna55 wrote: In reply to alfadriver: Perhaps we're not speaking the same language. I was advocating for a signal to convince the ECU that the transmission was in drive. I imagine this could be easily toggled with the key in the future. If that is plausible, the engine torque should satisfy the ECU, right?

Maybe, but the ECU would also know what gear it should be in via feedback from the trans. So not seeing info from the trans would cause a new set of problems. (and all the trans actuators would be in error, which probably would also require some torque truncation to make sure it stays alive.)

I see, so we need a magic box to simulate that also?

I presume a ride along with a donor vehicle would tell us, but then again, if it's a complicated signal, you're not going to reverse engineer it from a trace.

I'm just a dumb mechanical engineer. You would be far more educated at guessing what form of signal we may be talking about here.

In reply to tuna55:

Well, one theoretically could do that.

Or (assuming GM is like us) a hack should be able to set the code up so that it think's its a manual. For me, it's one parameter I change to be manual, and I can choose a more elegant set up with a clutch/neutral switch. IF this is the problem, this is the path I would take.

(FWIW, I'm an ME, too. Just dabble a lot in EE, CS, CE etc- to be a powertrain guy)

So, we had a similar issue in a friends Subaru race car. We got around it by attaching the VSS to an electric motor attached to a potentiometer. We were in effect telling it the car was doing 60mph. Always.

Its CAN bus isnt it? Should be able to get in and change it to a "manual" you just have to find someone with the computer to be able to do it...

I had a whole response to these wonderful suggestions and somehow between mouse click and the forum, it disappeared. I'll try to answer everything. -There is a different rev limiter for the auto and manual trans cars. I can adjust this limit with HPT but that only controls what rpm the ECM allows it to rev to, the ECM still cuts the throttle opening to 30% regardless what the pedal is doing. -I am looking into the possibility of one of the ECM pins needing a signal to tell it that the car is in gear. I hooked up the pin for clutch and have a switch that does convince the ECM that the clutch is depressed or not depressed. -I have a friend with an oscilloscope and a wife who drives a car with similar engine to mine, he tested it and was able to replicate the signal with the Dakota Digital Box that I have -The electronic gas pedal is the one from the original car. On HP Tuner I can see what the pedal position is vs the throttle position. everything there seems normal except that the ECM cuts throttle almost immediately when I step on it -Trifecta emailed today to tell me that they have nothing for my application.

DaveEstey wrote: So, we had a similar issue in a friends Subaru race car. We got around it by attaching the VSS to an electric motor attached to a potentiometer. We were in effect telling it the car was doing 60mph. Always.

Bwahaha There was probably a more elegant way to do that with an Arduino or similar...

Really out there, but you might try contacting this guy https://www.youtube.com/watch?v=vdZKbARFgoY and see what they are doing with their engine management.

(seems like they are using the same engine in an airboat)

Sparkfun has a CAN BUS shield: https://www.sparkfun.com/products/13262

Don't know if that would help but it would be something to try.

Apexcarver wrote: Really out there, but you might try contacting this guy https://www.youtube.com/watch?v=vdZKbARFgoY and see what they are doing with their engine management. (seems like they are using the same engine in an airboat)

Yes, been in contact with them. They use Trifecta and hate them. The tune has the AFR all out of whack and they surge. Trifecta isn't willing to fix their tune and now Trifecta won't provide a tune for swapped or standalone engines.

There are people with the capability to hack the operating system of the E39A and permit full power without a vehicle speed sensor. Does anybody on GRM know one of these people?

More details will come on my build page soon but long story short, it looks like my problem got solved today. Bad News Racing has been beating the bushes to find a solution and today they sent me a tune that magically made the vehicle speed sensor work. I only took a short drive but with brakes fully applied, the engine just overpowered the rear tires and dragged the locked up front tires along the track. I'm stoked!

Congrats! Looking forward to seeing this thing in action!

loosecannon wrote: .... with brakes fully applied, the engine just overpowered the rear tires and dragged the locked up front tires along the track. I'm stoked!

Was this maneuver intentional? Sounds like a full throttle error and an attempt to survive the event.

NOHOME wrote:

loosecannon wrote: .... with brakes fully applied, the engine just overpowered the rear tires and dragged the locked up front tires along the track. I'm stoked!

Was this maneuver intentional? Sounds like a full throttle error and an attempt to survive the event.

I am quite unfamiliar with turbo cars (although I drove a turbo diesel truck for 10 years) but it seems to me that they need some kind of load to build boost. My indoor "test" track is not big and the car is very light so to add some resistance, I add some brake pressure. The car doesn't feel particularly fast compared to the NA V12 or V8 but it walks away from a kart like they are standing still. The car has 56% weight on back tires so it's no surprise that the front tires can't stop the car when the throttle is applied.

I have been looking over this thread and after looking at the schematics, in the automatic cars, even though the shifter cable mechanically adjusts the lever, a message (via monitored voltage drops)from the shifter to the bcm to tell what gear it is in when in tap shift mode. From the bcm a message is sent through the high speed data line to the tcm, ecm, and the instrument cluster. This doesn't happen in the manual trans cars. They only have a clutch apply switch and vehicle speed sensor. I checked in my parts dept. and found the all of the ecms are the same despite what motor you have, the 2.0, 2.4 or the 3.6. I would be curious if you were to use your ecm programmed back to stock and in the donor car, with a stick shift, had a dealer to reprogram your ecm as a stick shift car. I feel like there is something in an auto cars program that is blocking you or a parameter that the ecm is not seeing. I work with someone who races in the X275 drag racing series who has a friend who specializes in CTS_V tuning in the Dallas area. He has a facebook page "Calibrated by Karger". I know you are frustrated and are not wanting to spend any more money, but it appears as though you are at a road block. And maybe a fresh set of different eyes would do some good. My friend Mike has had his Camaro tuned multiple times by him with great success. Mike wanted it to lay down big power but he is very picky with how it drives, idles, and general behavior while driving. I personally don't know the "Calibrated By Karger" guy and not trying to sell you on him. I just have only heard good things about his tuning abilities. Hope some of this helps. Good luck with the build and I would be very interested to here what was road blocking you.