Tuning Problems 2004 Sunfire Hahns Stage2

[hackish]

Someone brought me a 2002 Sunfire with an ecotec 2.2 swapped in. That swap included a computer and HPT recognises it as a 2004 Jbody. (Can't remember if it says Cavalier or Sunfire). The car has a Hahns Racecract Stage2 Turbo kit installed.

The Hahns kit came with a BEGI FPR and 32lb injectors. The customer believes that the stock injectors were 24lbs. When he arrived we removed the BEGI and installed an AEM 1:1 FPR set at 43 PSI. I Installed HPT on it. The engine had a 1 bar MAP sensor so we scrounged up a replacement. A 1.7 bar sensor from a honda was used. These have proven to register to 10 PSI reliably.

I scaled the tables as per instructions on this site. Now the car will not stay running on startup. The cranking fuel is correct as adding to the cranking VE causes the cranking fire to feel very rich. As soon as it exits the cranking mode I assume it will move to the idle fuel table.

I tried changing the injector factor from 1.7 to a variety of lower values without any luck. So I disabled closed loop mode and set the target idle to a variety of spots. The car will start and run if he pumps the throttle while cranking. I have a feeling that the afterstart enrichment is wrong. I leaned out the mixture at a variety of idle points to develop a reasonable idle VE table. It now idles at from 13.8-14.5 from 1500-700 RPM.

It still dies on startup. The idle control also has too much authority on timing as it will go anywhere from 0 to 24 degrees trying to maintain a good idle. Normally I disable timing changes during fuel idle tuning. No such feature here. The idle quality is fairly poor. Generally I'd like to see +- 6-8 degrees of authority here.

Unfortunately the main fuel tables are based on an alpha-n strategy which is not good for a turbo application. There is nothing available to change this So driving I dialled the VE table in a little to provide a conservatively rich fuel strategy at low throttle positions allowing the closed loop to trim it up to stoich for fuel economy. I set PE mode to come on very agressively. As soon as he leans on it and starts making boost it drops to the desired 11.5AFR- except at higher RPMs but I haven't figured how to properly adjust that yet but it's still within safe limits.

Tuning with the system was absolute hell and has taken more than 15h of work. The VCM editor has a number of bugs and throws the occasional index out of range exception. (I will report that as soon as I can reproduce it properly). The scanner doesn't seem to support anything useful. There is no real time tuning. I played with the histogram feature for almost an hour but can't stop it from saying unsupported on every page. A few small VCM editor changes would go a long way. I'd love to see the ability to scale the concept of MAP for a larger sensor. It's only a display anyway. If only the scanner and editor could talk to each other and show me where in the map it is!

Right now if I didn't have a lineup of other cars to tune I'd start dumping out the calibration files and disassembling them myself to find the proper settings.

Can anyone offer suggestions? Maybe there is another VCM that would be better supported on the ecotec?

-Michael

[Keith@HPTuners]

Few things..

Post the version of VCM Suite you are working with.

Post the HPT file you read from the vehicle, before and after your changes.

Post some log files.

The VCM editor has a number of bugs and throws the occasional index out of range exception. (I will report that as soon as I can reproduce it properly).

Please do so.

The scanner doesn't seem to support anything useful.

Such as? Many other 4cyl tuners are using our product with great success. Our scanner software is also one of the most powerful on the market for GM vehicles.

There is no real time tuning.

This is a limit of the PCM, not our product. No GM PCM's that we support have real time tuning capabilities from the factory.

I played with the histogram feature for almost an hour but can't stop it from saying unsupported on every page.

For this you should post in the help section, its likely your histograms are not configured properly or you are not adding the required parameters to support those histograms to the table display.

A few small VCM editor changes would go a long way. I'd love to see the ability to scale the concept of MAP for a larger sensor. It's only a display anyway.

The ability to scale the MAP for a larger sensor is no small change.

If only the scanner and editor could talk to each other and show me where in the map it is!

That is what the histograms are for. If you define the histograms properly they will size automatically to the exact dimentions of your calibrations table for VCM Editor. Our scanner is the only software that does this for GM vehicles.

[imphat0260]

Keith is correct... It honestly sounds like you do not have all the proper items that need to be logged in the primary table of the scanner.... Did you start running into the problems when you switched MAP sensor out??? I along with others have had some success with keeping the 1BAR in with boost. I along with others have also had success with faking a 2BAR (just takes longer/loss of resolution)... Anything over that and your wasting your time.. Also, not trying to be an ass, but what are you used to tuning??? Are you a shop?? Have you ever tuned using HPT before?? I think its a pretty easy program to use compared to others I have used in the past....

[Russ K]

Only the idle VE table is referenced to map, so I don't see why you would want to fake a 2 bar or install a 1.7 bar map sensor.

Russ Kemp

[hackish]

Only the idle VE table is referenced to map, so I don't see why you would want to fake a 2 bar or install a 1.7 bar map sensor.

Russ Kemp

It looks to me like the PE mode references the vacuum signal with a multiplyer to achieve the target AFR and that is why I opted to use the MAP signal hack. Otherwise if the driver is WOT at 4000 RPM there is no way the computer would know if we're at 1psi or 9 psi of boost - a very significant difference in required pulsewidths. In actual fact as soon as PE mode kicks in the computer is pretty good at achieving the requested AFR of 11.5.

-Michael

[hackish]

I along with others have had some success with keeping the 1BAR in with boost. I along with others have also had success with faking a 2BAR (just takes longer/loss of resolution)... Anything over that and your wasting your time..

The idea of using a 1.8 bar map sensor is consistant with what you're saying about resolution loss. I was told by the customer that the stocker was 2 bar already but measurement with a meter showed otherwise. As a result we were stuck and used the 1.8. I don't think 0.2 bar of resolution would cause a no-start.

Also, not trying to be an ass, but what are you used to tuning??? Are you a shop?? Have you ever tuned using HPT before?? I think its a pretty easy program to use compared to others I have used in the past....

I think the technical depth of my question illustrates enough to answer this question. I helped someone a while ago with HPT but his car was apparently better supported. The options available to him resembled LT1Edit. Obviously I've tuned with that, hondata, crome, openECU, DTA, FAST, AEM, Motec, Pectel, GEMS, Hydra, SDS, CalMap, Tec2, Haltech. Been doing it for 7 years now. I've also worked on the design team for a standalone ECU and have written a number of firmware features for both it and the honda PGMFI group. I've also tuned cars from 100-700hp. Not a newbie of any sorts.

I don't think HPT is altogether crappy but out of the box it does appear to require a lot of time just to make it work. I'm in the business of tuning cars so I don't expect it to take a few hours selecting PIDs to log before a system can show me if PE is enabled or where in the fuel or spark map it's looking. I still haven't been successful in getting these to work.

I'll load the images and post them up shortly.

-Michael

[spyhunter]

I'd love to see the ability to scale the concept of MAP for a larger sensor. It's only a display anyway.

-Michael

No kidding, you guys know most of us would be using it to tune boost. What will it take for you guys to just look at the 4cyl stuff again and be able to at least tell us about the possibilities of giving, 2bar and 3bar options?

You know there is a demand for it, all this fake'ing nonsense is somewhat a shitty fix. *sarcasm, I absolutely love it that I get to lose resolution, and my somewhat fine tuning if I want to go 3bar*

god bless america.

[hackish]

The idea of faking it is all in how you look at it. Both sensors will read 0.05-4.95v or somewhere around there. The VCM sees the signal. It doesn't care if you're calling this a 5bar MAP sensor all it knows is the voltage it sees coming in.

I realise that properly supporting it would mean changing the firmware so the MAP signal would be used rather than the TPS for the fuel and ignition tables. Alpha-n v. Speed Density. Alpha-n does not work well with boost since 55% TPS could mean 0 PSI or 10 PSI of boost on a small turbo.

The lookups for the tables themselves would need to be altered so you could specify that for example the idle VE table would use the new settings for your larger MAP (not losing resolution by throwing away the now boosted rows).

I don't know what sort of hardware fixtures HPT has or uses but having worked for some well financed groups I know it's very expensive to set up a complete test rig with data acquisition and a completely emulated engine.

Patching the firmware code may be outside their capacity. Both in technical and in resources. They can answer that. I don't have the time or finances to do something like this independently. There is a lot of code in a VCM and the MCU and hardware has to be well understood before one could think of changing/adding features.

-Michael

[redhotjamedjimi]

im faking 2 bar it isnt that bad except the idle tune. the lowest 25 kpa is around 34 real. below that stock it retards the timing but no ability on a fake. my only problem. it was a pain to do the algebra to make it absolute.

[imphat0260]

Michael, I dont want you to think I was doubting your abilities, I was simply just inquiring what you are used to tuning with... Im used to EFI Live, LS1 Edit, Electromotive and some other piggyback units.... IMO out of all the stuff I have used, HPT seems to be the best choice fo me.(more options/platforms & better customer support) The Jbody platform really is limited compared to other platforms. I would honestly suggest just putting the 1BAR back in and going from there.. Im going to attach a "generic" Jbody scanner configuration. Once you hook it up to the vehicle. all the histograms should work.. You may need to change the wideband settings if you are not using the AEM UEGO.... Hope this helps you....

[hackish]

The biggest unanswered question is with a 1 bar map sensor how will the computer know the difference between 1psi and 10 psi and thus fuel and spark it correctly?

Thanks for the config file, I'll see when the custom is available to return so I can try it. I'll also fire up my tuning laptop and post the .hpt file I have. (Been stuck welding intercoolers so far this week).

-Michael

[redhotjamedjimi]

the 1 bar wont know your even at 1 psi. its the same spark from 106kpa and on thats why 2 bar the way to go for blown engine so u can advance the spark at 0 and have the correct under boost. i even have my timing killed after 9-10 psi to 0 degrees in case of wastegate malfunctioning. but at idle and when u try to accelerate out of idle wont be nearly as good. sometimes when its idling at 1000 rpms ill tap the gas and it will stumble from the inability to advance it corectly.

[runndownn]

Sounds like Its the VATS that needs to be disabled...Vehicle Anti Theft System...Its in System under the General tab i know its just a little late...