My GVR4 is the Devil.

Memphetic · Jan 24, 2015

'92 GVR4. I put 720's on it and an EPROM chip I burned. Was running great. I put a wideband in place of the narrowband set on simulation mode and hooked it to the ecu. A couple days later, it starts running lean under boost and rich at cruise. I immediately think "air leak prior to turbocharger between MAF and turbo". I check for leaks in the intake pipe... none. So, I think, must be the fuel pressure regulator being overrun (it's an Airtek E2448, aka a Walbro 284 in a $45 box), and swap it with the one from my VR4 and set the fuel pressure to 43.5psi. I test it up to 90psi, no leaks. Solves nothing.

I go to Advance and swap the pump out for a new one in their parking lot because it takes 3 tools and 10 minutes... solves nothing.
I replace the fuel filter with a brand new one... solves nothing.
I hotwire the pump to receive constant voltage and it solves nothing.
I think maybe my fuel injectors are clogged (all of them? LOL

??), so I swap them back out for stock 450's and drop the fuel pressure to 36psi. I also remove the ECU, because that may be a culprit, and replace it with a stock non-EPROM. Solves nothing.
Also tried replacing the resistor pack with a known good one, as well as the MAF sensor. Solved nothing.
Driving around on 17:1 AFR's right now, and if you hit boost or go over 30% throttle it fuel cuts immediately

Dsmkauai · Jan 24, 2015

You sure your wideband is reading right? What brand and how old is the sensor, can it be calibrated?

Nate355RS · Jan 24, 2015

Sounds like an exhaust leak before the sensor. What brand wideband and where is the sensor grounded? Have you tried putting the stock sensor back in and seeing if it runs right? If it does that'll tell you is a problem with the wideband setup.

GST with PSI · Jan 24, 2015

I'm confused on what you're using to tune the car with. You said Eprom chip, but then made it sound like you could simulate the NBO2 (like with with ECMlink). What are you actually using to tune the car with?

If you want people to take your seriously when asking for help, take the time to add a profile for your car with an updated mod list. Some of the more knowledgeable guys who could provide you with some really good feedback won't even bother posting if they have to guess what's done to your car.

Just as my first question to you could easily be answered by a quick glance at your profile, a complete mod list of what's done to the car would help greatly. You'll save yourself and others a lot of trouble by doing these simple things.

Memphetic · Jan 24, 2015

GST with PSI said:
I'm confused on what you're using to tune the car with. You said Eprom chip, but then made it sound like you could simulate the NBO2 (like with with ECMlink). What are you actually using to tune the car with?

If you want people to take your seriously when asking for help, take the time to add a profile for your car with an updated mod list. Some of the more knowledgeable guys who could provide you with some really good feedback won't even bother posting if they have to guess what's done to your car.

Just as my first question to you could easily be answered by a quick glance at your profile, a complete mod list of what's done to the car would help greatly. You'll save yourself and others a lot of trouble by doing these simple things.

An EPROM chip can do everything I want it to, since I program in assembly. I can do way more than ECMLink can dream of because I can add and remove code as I wish. In addition, all AEM UEGO widebands have a Nernst emulation mode that's turned on by rotating the calibration dial on the back to P04. You then just cut the Rear O2 (or just O2 sensor on a 1G) input and splice the analog output from the gauge to the ECU.

In addition, I've listed every mod that my car has in the first post.

I've programmed native 0-5v wideband input into my chip as well, but have it turned off and the gauge set to Nernst emulation (acts like stock). However, if you're familiar with the programming of the ECU, you'll note that nothing O2 sensor related will cause an effect in open loop (meaning under boost or acceleration), aside from if your fuel trims have already been modified by cruising in closed loop. That's why I listed my fuel trims.

The car didn't start running like this until a few days after I swapped out the wideband, and I'm certain it's reading correctly because when the gauge goes lean, the car shudders.

Thanks for the thoughts thus far, though.

Nate355RS · Jan 24, 2015

Details like what turbo, intake, are you on the stock MAF? Is your bov recirculated?

And I'll say it again, sounds like an exhaust leak either before or at O2 housing. Exhaust leaks cause all kinds of problems, inaccurate O2 readings are one of them.

If you know about the ECU code you know about all the conditions that cause fuel cut. You should be able to spot them fairly easily in a datalog. That will narrow it down.

Memphetic · Jan 24, 2015

Stock turbo, stock intake, stock MAF. If BOV was vented, it would (at worst, if it leaks) cause a post-turbo boost leak, which would cause me to run rich under boost.

The only reason I don't believe it to be an exhaust leak is because it has no bearing on acceleration... you can literally replace your narrowband with a wideband set to 0-5v and your WOT will still not be affected because you only go closed loop under the following conditions:

; engine has been running for more than 180sec
; baro sensor value is reliable
; no fault code on baro, coil, iat, ect, cas
; ectRaw and iatRaw are within acceptable range
; temperature(ectRaw) > 86degC
; temperature(iatRaw) > 0degC
; temperature(iatRaw) < 55degC
; engine is running normally (no fuel cut, etc..)
; airVolTB < $68
; airVolTB > $33
; rpm31 < 4000
; rpm31 > 2000
; all conditions for closed loop mode are met

GST with PSI · Jan 24, 2015

Memphetic said:
An EPROM chip can do everything I want it to, since I program in assembly. I can do way more than ECMLink can dream of because I can add and remove code as I wish. In addition, all AEM UEGO widebands have a Nernst emulation mode that's turned on by rotating the calibration dial on the back to P04. You then just cut the Rear O2 (or just O2 sensor on a 1G) input and splice the analog output from the gauge to the ECU.

In addition, I've listed every mod that my car has in the first post.

I've programmed native 0-5v wideband input into my chip as well, but have it turned off and the gauge set to Nernst emulation (acts like stock). However, if you're familiar with the programming of the ECU, you'll note that nothing O2 sensor related will cause an effect in open loop (meaning under boost or acceleration), aside from if your fuel trims have already been modified by cruising in closed loop. That's why I listed my fuel trims.

The car didn't start running like this until a few days after I swapped out the wideband, and I'm certain it's reading correctly because when the gauge goes lean, the car shudders.

Thanks for the thoughts thus far, though.

Well, since filling out your profile is too much trouble, I'm not going to guess about all the sh|t you may or may not have done. I doubt many others will either.

I know enough about programming to know that it doesn't take much to completely dick things up. Not saying that's what's happened in this case, but unless you find someone here that has the time and knowledge required to double check your source code, it's pretty much impossible to know if you've created an issue. Again, I'm not saying you don't know what you're doing, but there could very well be an issue there. You may be able to do a million times more than ECMlink by tweaking code and whatnot, but you lose the ability to share your tune and configuration in a form that's easily readable to others. Looking at a log file would be pretty helpful right about now, no?

I can't tell from your first post if you completely returned the car back to stock when you swapped in the non-Eprom ECU or not. If you didn't, that's what I would suggest doing based on the information at hand.

Memphetic · Jan 24, 2015

Yes. I did. That's why listing mods is pointless - there are no mods. As I said, I listed the mods I had done, and also listed the mods I've undone (which was everything). I don't know why it's so hard for you to follow that without looking at a profile. Here's my profile: Stock GVR4. It's obviously not an issue with my code, as it's running a stock non-EPROM ECU. As I said. But you did not follow.

I don't mean to be standoffish, but you're apparently not even reading my posts. The car was stock prior to me getting it. I added an EPROM with my code, 720's, and a wideband. I removed all of them, except the wideband set to Nernst emulation mode (which is verified working properly through my datalogger). Thus, it is a stock car.

CrackedDSM · Jan 24, 2015

GST with PSI said:
You may be able to do a million times more than ECMlink by tweaking code and whatnot, but you lose the ability to share your tune and configuration in a form that's easily readable to others

Not true. Any modifications you do the code can be used and adjusted in the form of a Scalar, flag, or table which is very obvious and easily accessible. Plus those who have used TunerPro, burned their own chips, or etc pretty much know where stuff is. Not at all hard to see if he's doing something screwy at all. I've got tunerpro experience and thus far, what he's saying lines up from what I've experienced and seen others do. Hard to explain but if you actually had TP experience or Ostrich experience it would make more sense. It sounds more like his comment towards ECMLink ruffled your feathers, which it honestly shouldn't. Both have their high and low points but from a raw coding standpoint the Ostrich has a significant advantage in terms of "what can I do.."

OP, sorry I can't be more help but soon as I get home in a few hours I'll chime in more. Can only do so much on the work PC...

GST with PSI · Jan 24, 2015

CrackedDSM said:
Plus those who have used TunerPro, burned their own chips, or etc pretty much know where stuff is.

Yeah, and as I said, the amount of people here who fall into that category are far from the majority. My point was ECMlink provides less hoops to jump through for most people who help with these sort of things. Get my drift?

Memphetic said:
'92 GVR4. I put 720's on it and an EPROM chip I burned. Was running great. I put a wideband in place of the narrowband set on simulation mode and hooked it to the ecu. A couple days later, it starts running lean under boost and rich at cruise. I immediately think "air leak prior to turbocharger between MAF and turbo". I check for leaks in the intake pipe... none. So, I think, must be the fuel pressure regulator being overrun (it's an Airtek E2448, aka a Walbro 284 in a $45 box), and swap it with the one from my VR4 and set the fuel pressure to 43.5psi. I test it up to 90psi, no leaks. Solves nothing.

I go to Advance and swap the pump out for a new one in their parking lot because it takes 3 tools and 10 minutes... solves nothing.
I replace the fuel filter with a brand new one... solves nothing.
I hotwire the pump to receive constant voltage and it solves nothing.
I think maybe my fuel injectors are clogged (all of them? ??), so I swap them back out for stock 450's and drop the fuel pressure to 36psi. I also remove the ECU, because that may be a culprit, and replace it with a stock non-EPROM. Solves nothing.
Also tried replacing the resistor pack with a known good one, as well as the MAF sensor. Solved nothing.
Driving around on 17:1 AFR's right now, and if you hit boost or go over 30% throttle it fuel cuts immediately

Memphetic said:
I don't mean to be standoffish, but you're apparently not even reading my posts. The car was stock prior to me getting it. I added an EPROM with my code, 720's, and a wideband. I removed all of them, except the wideband set to Nernst emulation mode (which is verified working properly through my datalogger). Thus, it is a stock car.

I'm reading your posts just fine. Clearly, conveying facts isn't your strong suit. Sorry if I'm not able to put all the pieces together from the wealth of information you're willing to provide. You mentioned adjusting your fuel pressure, so did you accomplish this with a stock regulator or what? In addition, base 1G FP is 37psi, not 36, although I doubt the 1psi in variation is your issue.

According to you, "You add a wideband in place of the narrowband set on simulation mode and hooked it to the ecu and a couple days later, it starts running lean under boost and rich at cruise." That tells me you screwed something up. Reverse what you did, and I'll bet the car starts running like it's supposed to again.

I read your previous post saying you already did reverse everything, and you've obviously missed something. Again, according to what you've said thus far, the only variable between a car that was running just fine, and a car that's now running like crap, is you. Figure out what you haven't undone, and that's your issue.

Memphetic · Jan 24, 2015

I did forget to mention that the regulator from my (3000GT) VR-4 is an Aeromotive AFPR.

I really don't understand why it's hard for you to get that outside of adjusting cruise AFR's and fuel trims, both of which only occur in closed loop, the stock O2 sensor does absolutely nothing - and that my fuel trims are very close to 0% adjustment.

GST with PSI · Jan 24, 2015

Memphetic said:
I did forget to mention that the regulator from my (3000GT) VR-4 is an Aeromotive AFPR.

My point exactly. Nobody wants to guess.

I understand fuel trims just fine, and how they relate to closed/open loop. It sounds like you've got it all figured out, so I'm not even sure why you're asking for help. I hope you figure it out, and good luck.

CrackedDSM · Jan 24, 2015

GST with PSI said:
Yeah, and as I said, the amount of people here who fall into that category are far from the majority. My point was ECMlink provides less hoops to jump through for most people who help with these sort of things. Get my drift?

I get what you're trying to say, but as you've never used it you aren't getting the point. ECMTuning being the powerhouse of advertisement and the fan base they have, they are definitely in the majority...however it's not exactly trying to decrypt Mandarin to look at a log or show your DA tables using TunerPro. Again, you're in a huff because he spoke the truth against a product you are passionate about which is fine, but if you're just going to insult the guy and swing from an unrelated products nuts I don't really understand why you are here. If he bugs you that bad, it's easy to just click the back arrow and ignore the thread.

@Memphetic honestly I'm at a loss. You've already reverted back to the stock ECU, and changed everything I would tell you to check. MAF is good, injectors aren't the issue, you've changed ECUs, fuel pressure and fuel pump both have been verified...no amount of adjustment will change the AFRs? The only thing at this point I can think of is a fuel restriction somewhere. Bent fuel line, clogged lines at the tank or rail, etc. That'd be my best guess at this point.

GST with PSI · Jan 24, 2015

Right, I missed the point...

What do you mean you're at a loss? Why don't you look at a log using TunerPro and give us ECMtuning nutswingers a hint? I'm confident I could take a look at a typical ECMlink log and have a pretty damn good idea what's wrong. So could half of the other forum members here; which was my point in case you missed it for the THIRD time. What voltage is his hacked up O2 cycling at? What are his ACTUAL fuel trims looking like? What's the MAF output look like? Hell, coolant temp and battery voltage would be nice to know as well. You'll have to feed me the answers, since ECMlink has brain washed me into abandoning all other forms of engine management. Lets not forget any other mods, eh? I'd still like to see an updated profile to clue everyone in on the other sh|t that may have failed to get mentioned.

Instead of updating his profile and answering the simple questions people have posed, he wants to give everyone the run around. Sounds to me like you two have everything figured out? You should have just PM'd each other, rather than start a thread. Don't ask for help if you're going to get butt hurt about getting asked to do the basics.

Memphetic · Jan 24, 2015

GST with PSI said:
Right, I missed the point...

What do you mean you're at a loss? Why don't you look at a log using TunerPro and give us ECMtuning nutswingers a hint? I'm confident I could take a look at a typical ECMlink log and have a pretty damn good idea what's wrong. So could half of the other forum members here; which was my point in case you missed it for the THIRD time. What voltage is his hacked up O2 cycling at? What are his ACTUAL fuel trims looking like? What's the MAF output look like? Hell, coolant temp and battery voltage would be nice to know as well. You'll have to feed me the answers, since ECMlink has brain washed me into abandoning all other forms of engine management. Lets not forget any other mods, eh? I'd still like to see an updated profile to clue everyone in on the other sh|t that may have failed to get mentioned.

Instead of updating his profile and answering the simple questions people have posed, he wants to give everyone the run around. Sounds to me like you two have everything figured out? You should have just PM'd each other, rather than start a thread. Don't ask for help if you're going to get butt hurt about getting asked to do the basics.

Aaaaand THIS is why I don't post on Tuners anymore.

First off, you don't use TunerPro to view logs. Secondly, my fuel trims are -4%, -4%, and -2% (as. I. said.) The voltage for my "hacked up O2", also known as an AEM UEGO in P04 mode, are cycling precisely as they should with corresponding less than and greater than .5v depending on whether lean or rich. Want to see a log? I'll post one, but you probably won't know how to view it.

GST with PSI · Jan 24, 2015

Memphetic said:
Want to see a log? I'll post one, but you probably won't know how to view it.

Well, I'm waiting?

My money says it probably looks like a train wreck...

Memphetic · Jan 24, 2015

What would you like me to log for your viewing pleasure...?? I'll use the datalogger hardware/firmware that I built (MMCdroid) with the software that I wrote and hopefully we can get that train wreck on here.

GST with PSI · Jan 24, 2015

Just provide a log and specify the software required to view it, and I'll take care of the rest. That's assuming it's not some proprietary one-off BS, of course.

Nate355RS · Jan 26, 2015

There's one more thing I can think of....

Did your base tune come from a 1g?

Since the replacement ECU is non-eprom I'm guessing it's from a 1g as well? If you are working with 1g code, then that could be at least be part of the reason it's running lean. A GVR4 MAF and a 1g maf are physically different and as a result a GVR4 MAF reads lower hz than a 1g. Switching from a GVR4 tune to a 1g tune usually causes fuel to be off by about 15-20% which sounds very similar to your case.

I know this because I have a 1g and #77/1000. Back when they were close to stock I tuned my 1g and then tried to put the tune on my GVR4. I was about 20% lean across the board.

The Top DSM Community on the Web

My GVR4 is the Devil.

Memphetic

Dsmkauai

Nate355RS

GST with PSI

Memphetic

Nate355RS

Memphetic

GST with PSI

Memphetic

CrackedDSM

GST with PSI

Memphetic

GST with PSI

CrackedDSM

GST with PSI

Memphetic

GST with PSI

Memphetic

GST with PSI

Nate355RS

Similar threads

Latest posts

Build Thread Updates

Vendor Updates

Latest Classifieds