Mitsubishi 3000GT & Dodge Stealth Forum banner

121 - 140 of 164 Posts

·
Registered
Joined
·
1,267 Posts
Discussion Starter #121
Maybe it was just a coincidence that it started adjusting idle fuel trim? I’m really confused too.

where I park my car there’s a divot in the gravel under my car port lol.

I’m seriously confused with this car. It makes no sense that it randomly adjusts idle fuel trims whenever it feels like it!? That’s a normal thing at idle right? How does it go from idling at 12.9 afr to 16 afr on the same drive? Fuel trim for low went from 120s to 81 in less than a minute.

electrical gremlin hell?
 

·
Registered
Joined
·
158 Posts
Sounds like you've got unstable fuel pressure primarily caused by fitting a pump that's too big for the FPR.
You'll need to double and triple check the pressure.
When they roll out of the factory they have 33 PSI gauge pressure at idle speed with the vac line still connected.

That's the magic number, not the 43 PSI with the vac line disconnected.
 

·
Registered
Joined
·
1,267 Posts
Discussion Starter #123
My fuel psi is normally 37-38 at idle. I set the afpr to 43 with the vac hose disconnected. Should I target 33 with the vac hose connected?

Im willing to try anything at this point. Considering buying a new fuel pump and 550 injectors to rule that out.

it’s atleast targeting 14.7 while driving now. it’s idle and light accel in parking lots that’s a problem.

My goal is to keep driving it. The issue with this car is it’s a second vehicle so I’m in it once every 5 to 10 days. It’s very hard to remember what I’ve talked to people about and what the car is doing in those 5-10 days. Datalogging is a priority right now. Getting that going should help tremendously.
 

·
Registered
Joined
·
158 Posts
My fuel psi is normally 37-38 at idle. I set the afpr to 43 with the vac hose disconnected.

It's because your FPR can't flow enough to keep the pressure down.
It's exactly what happens when you fit a walbro pump and retain the stock FPR.
Fitting a bigger pump than a walbro 255 makes it all worse.
Deleting the pump resistor makes it worse still.


Too much fuel pressure at idle means too rich at idle , that shifts the fuel trims so they're pegged at 88%or 81% or whatever the figure is.
Then when you go into boost it runs too lean due to the trims.
Too much pressure also messes with the injector deadtime settings.

Most aftermarket FPRs can't flow enough either.
When the FPR can't flow enough then the pressure varies with voltage.
Voltage varies with electrical load particularly at idle which means that switching all lights and fans on will alter the fuel pressure and therefore AFR compared with having all fans and lights off.


The easy cheap fix is to use 2 stock FPRs, one on each fuel rail with the inlets teed together and the outlets teed together.

TT FPRs are 3 bar, N/A FPRs are 3.5 bar so avoid using an N/A FPR.
 

·
Registered
1991 VR4
Joined
·
1,195 Posts
It's because your FPR can't flow enough to keep the pressure down.
It's exactly what happens when you fit a walbro pump and retain the stock FPR.
Fitting a bigger pump than a walbro 255 makes it all worse.
Deleting the pump resistor makes it worse still.
True, you can overrun a FPR like stock one, but he has a AFPR and is controlling fuel pressure, so very unliekly he us overrunning it with a hotwired denso.

Then when you go into boost it runs too lean due to the trims.
When you go into boost the car is going into open loop so the trims are irrelevant. It's not using the NBO2 and there are no trims.

Too much pressure also messes with the injector deadtime settings.
Not sure what this is supposed to mean; the deadtime settings are in the tune, hes running a stock ECU so he can't touch those, and while the FP can affect the actual flow from the injectors, its not affecting the deadtime settings.

Most aftermarket FPRs can't flow enough either.
When the FPR can't flow enough then the pressure varies with voltage.
Voltage varies with electrical load particularly at idle which means that switching all lights and fans on will alter the fuel pressure and therefore AFR compared with having all fans and lights off.
Not sure exactly which AFPR hes running, but people run them with much bigger pumps than he has. Indeed fuel pump output will vary with voltage but his is hotwired and if his alternator is working properly his voltage should be more or less contant regardless of what is running.

The easy cheap fix is to use 2 stock FPRs, one on each fuel rail with the inlets teed together and the outlets teed together.
TT FPRs are 3 bar, N/A FPRs are 3.5 bar so avoid using an N/A FPR.
Never heard of running 2 stock FPRs but hey, i've seen stranger things.
 

·
Registered
Joined
·
1,267 Posts
Discussion Starter #126
I have the Aem afpr.

So after the chaos today I drove home and switched the injectors out for stock and threw the stock maf back on.

Car runs great

The end.

Sandstone92vr4 went on in life to become a librarian because he now completely hates cars. He may try to put 550s in next spring because he still has hopes of having a fast car.

Ps. I think the denso 660s were fucked.
 

·
One fix at a time
Joined
·
7,943 Posts
Well, now you know you’ve got a good baseline lol. Sometimes it takes returning most of it to stock in order to determine what was wrong. If you want to go forward now you should send the injectors out to get flow tested and cleaned. There’s multiple places that will do it under $100 if you don’t want to break the bank. Even if you didn’t want to upgrade injectors for awhile it would be good to know how far off they were and at least cleaned they’ll sell relatively easily.
 

·
Registered
Joined
·
1,267 Posts
Discussion Starter #128
What’s weird is that I did send them out to get checked out. They charged me $70-80 and said they are good. I do remember the flow chart showing some differences in them though, they were some that definitely weren’t within 2 or even 5cc of each other. Being told they were fine I rolled with them.

I can swap injectors and reinstall the maft in about 1.5 hours now. The coolant was still in the 120s when I was done.

The only thing I had to adjust was the idle, I think that’s normal though because the 85mm maf and cone filter will allow a lot more air in than the stock maf. So it was idling at 500rpm with iac counts in the 140s lol, but super smooth. No burbles or pops from the exhaust even that low. With the 660s it never sounded right to me. So I think I learned some info for other people. If your exhaust is burbling and popping, you probably have injector issues.
 

·
Registered
1991 VR4
Joined
·
1,195 Posts
Did they measure the resistance of the injectors in addition to flow test? Also are they commonly ran on these cars? Just some thoughts that could have made them perform strange.

With the hotwired denso pump and a AFPR you can probably turn the boost up a bit and enjoy the car with stock MAF and injectors. The dr650s will flow way more than stock at a given boost level, but you can probably turn them up to maybe 12psi before getting fuel cut or any other issues.
 

·
Registered
Joined
·
1,267 Posts
Discussion Starter #130 (Edited)
No idea about any of that. They just told me they were good. I remember seeing differences in flow rates as well but I lost that paper and cant find the emails.

Drove the car to work today, temp outside is 32. The only complaint I have is when I started it it still wouldnt fast idle. Just started up and sat at 900rpm or so, put my foot on it for a few seconds and then it started to idle at 1200 and work its way down. I have another FIAV to try. Other than that its good, nice and smooth idle, no hiccups. I do think the hotwired supra pump is too much for the stock injectors even with the AEM AFPR which has me a little confused.

Im thinking I need to retain the stock MAF in my climate. One day it can be 80, the next it can be 40. Its a tuners nightmare between summer and winter. Something with IAT and Baro is important I think. Not sure theres any tuning devices that can be used with a stock MAF and 550's? Other than AEM or Chrome?
 

·
Registered
1991 VR4
Joined
·
1,195 Posts
I don't see how a fuel pump can be too much with a AFPR, unless the AFPR is being overrun. An AFPR at X FP should make the injectors behave the same way as OEM FPR at X FP.

And yes, there is, SAFC. I have been running SAFC II with 550s since i got the car in 2007. I got the car untuned; previous owner had a shop rebuild the engine, install SAFC II and some other parts, then i bought the car as that unfinished project. Back then i had the car dyno tuned but as ive talked about i recently switched to E85 and do my own tuning. I daily drove the car from 2007-2018 and put 100k miles on this setup so it definitely works.
 

·
One fix at a time
Joined
·
7,943 Posts
You probably need to set your base idle speed again with the BISS screw. If you have the LCDBC you can click over to the menu and see how many IAC steps the motor is using to correct the idle rpm. If you’re consistently off more than 30-40 steps, then you likely need to readjust the base setting. The ecu can only compensate so much with ignition timing and fuel. It’s possible that contributed to your cold idling issues. Check out the Wiki for the procedure. You can skip grounding out the OBD port if the LCDBC stays plugged in. https://www.3swiki.org/Engine:_Adjust_Idle_Speed

While it is preferred to have an IAT and Baro signal for fuel corrections, a Hotwire sensor like the GM and Ford MAFs don’t need these for direct fuel calculations. They are fixed. The reason being is they measure air mass. Our Karman vortex sensors measure air volume and then to compute density and air mass they use the IAT and Baro signals to perform the calcs. The Hotwire sensors already do this and don’t have to perform any extra calculations. Air density that is affected by temperature and altitude will automatically draw a different amount of heat away from the wire causing a different resistance and voltage measurement. You can read more about it from the creator of the MAFT TurboBob:


There is no need for Baro and Airtemp with the MAFT. Regardless of various theories, here's how it really works.

The stock MAF measures air VOLUME flow using the Karman Vortex principal. Since the ECU is trying to calculate how much fuel to inject, it needs to know the air mass. Internally to the ECU, the Air volume, is multiplied by the density (Baro & Temperature) to get mass. Then this value is used to calculate injector pulsewidth.

If the stock MAF is replaced with a hot-wire MAF the signal is a MASS Flow signal. So the Airtemp and Baro signals MUST BE FIXED VALUES, or the ECU will end up double-correcting.

There is no issue with altitude or weather with the MAFT, I live in Michigan and we have sub zero winters, and 90+ summers. I drive my car daily with 550 Denso's and just the MAFT for tuning. My MAFT is mounted underhood, and I never tweak it. (although extension cables are now available)


Thanks for the good questions, no apology necessary.
Thread for reference: MAF-T ?'s


You shouldn’t have huge tune swings and drifts. If you do, then you don’t have it fine tuned enough to where the ecu can do it’s job and use the fuel trims to compensate the fueling requirements. It could possibly have a greater effect on startup for the first minute before the ecu starts using the 02 sensors in closed loop, but that is only because there are more variables the ecu uses the IAT and Baro compensations for than just fuel alone. The fuel trims can adjust +\-12.5% so if you are aiming for near close to 100 trims for idle(Low) and cruise(Mid), then the ecu will do it’s job and correct for the rest. Plenty of people have not had issues with their tune drifting. Sometimes you just may need more finite tuning if you are running too large of injectors than what you’ll actually be using and that’s where having Chrome can help you really dial it in.

Also, you should not have any issues with the fuel pump overrunning the AFPR. There have been some cases with AEM specifically that there was still overrun and general fuel issues across all loads but that was due to erratic fuel pressure control. It was found that the return orfice that came with the AFPR was too small and could not bleed enough fuel off with some of the higher output fuel pumps thus causing a very unstable fuel pressure. Perhaps it is something worth checking into. You may benefit from wiring in a $20 pressure sensor into your fuel rail and the LCDBC for one of the external sensors so you can monitor and see what your fuel pressure looks like at all times even under boost.
 

·
Registered
Joined
·
158 Posts
True, you can overrun a FPR like stock one, but he has a AFPR and is controlling fuel pressure, so very unliekly he us overrunning it with a hotwired denso.


When you go into boost the car is going into open loop so the trims are irrelevant. It's not using the NBO2 and there are no trims.


Not sure what this is supposed to mean; the deadtime settings are in the tune, hes running a stock ECU so he can't touch those, and while the FP can affect the actual flow from the injectors, its not affecting the deadtime settings.


Not sure exactly which AFPR hes running, but people run them with much bigger pumps than he has. Indeed fuel pump output will vary with voltage but his is hotwired and if his alternator is working properly his voltage should be more or less contant regardless of what is running.


Never heard of running 2 stock FPRs but hey, i've seen stranger things.
Most aftermarket FPRs don't have sufficient flow rate , just being aftermarket doesn't make it any better. Fuel lab ones are good?
Fuel trims, they're there all the time across the whole map, They don't magically turn off.
Dead times, well yes they're programmed into the ECU but when fuel pressure rises injectors open slower so the programmed values end up wrong.
Alternator output, it varies with RPM, at low RPM the voltage is low, at high RPM the voltage is low, at mid RPM the voltage is at it's highest due to the alternator outputting it's max, They all do that, they're supposed to, you see it on every data log.Like this which I just pinch from another thread of 2 days ago.
data-log-png.291657 (1464×860) (3si.org)
Yes people run much bigger pumps and have all sorts of problems with AFR up and down und all over the place. It's really dumb. Specially when it's so easy to put right.
2 FPRs?, how about 3 of them with 2 pumps? Gives perfect stability of pressure at all times.
 

·
Registered
1991 VR4
Joined
·
1,195 Posts
Fuel trims, they're there all the time across the whole map, They don't magically turn off.
Dead times, well yes they're programmed into the ECU but when fuel pressure rises injectors open slower so the programmed values end up wrong.
Alternator output, it varies with RPM, at low RPM the voltage is low, at high RPM the voltage is low, at mid RPM the voltage is at it's highest due to the alternator outputting it's max, They all do that, they're supposed to, you see it on every data log.Like this which I just pinch from another thread of 2 days ago.
Fuel trims do "turn off" on these cars. The NBO2s are only used when the car is warmed up, at low/light load situations - idle, cruise, etc. Once you get into boost, the car is 100% open loop. The exactly threshold is not known, as it varies with RPM and engine load.

If hes running stock fuel pressure (he should check as mentioned above), that point doesn't really matter even if it were true.

Yeah, i know it varies with RPM. Having said that, those swings from 13.1-13.9 are more than i see. And the FPR should be able to handle the effect of those voltage swings on pump output, otherwise tunes would always be extremely inconsistent when voltage varied.

I'm running stock FPR, probably overran, with the same pump as him (denso hotwired), and i have no problems with my tune /shrug.
 

·
Registered
Joined
·
158 Posts
Open loop and fuel trims are 2 different things.. Closed loop is anywhere AFR is supposed to be 14.7 and it's locked into that by feedback from the o2 sensor. Anywhere the AFR isn't 14.7 it's in open loop. The low fuel trim is set at idle from the o2 sensor feedback and it's applied across the whole map.
Lets say it works the way YOU say.
In a condition like this it's going to go from 14.7 closed loop to open loop, it ignores fuel trims so it goes suddenly lean or suddenly rich as soon as the trims are removed?
Of course it doesn't.
 

·
Registered
1991 VR4
Joined
·
1,195 Posts
Open loop and fuel trims are 2 different things.. Closed loop is anywhere AFR is supposed to be 14.7 and it's locked into that by feedback from the o2 sensor. Anywhere the AFR isn't 14.7 it's in open loop. The low fuel trim is set at idle from the o2 sensor feedback and it's applied across the whole map.
Lets say it works the way YOU say.
In a condition like this it's going to go from 14.7 closed loop to open loop, it ignores fuel trims so it goes suddenly lean or suddenly rich as soon as the trims are removed?
Of course it doesn't.
Maybe its a matter of semantics, but I meant fuel trims as the correction applied during closed loop to get to ~14.7. So if your tune is a little rich, fuel trims are negative (or <1, pulling fuel), to get you closer to stoich. I am not aware of these stock ECUs applying fuel trims across the entire map, only applying them in real-time based on O2s cycling, to try and get to 14.7. I'll let someone more familiar with the ECUs confirm that, though. I would like to know this piece of information if anyone else can confirm, because i did not think the base fuel map in the ECU was adjusted at all based on O2 sensor feedback (i.e., no learned long-term fuel trims, just the real-time corrections from NBO2s).

And based on everything i have ever seen on my own car and in logs, the transition from closed loop to open loop is exactly what you said doesn't happen - it will go from targeting 14.7 to becoming much more rich, for example when you are a light throttle low RPM and go to WOT. This happens as you transition from using the NBO2 for fuel corrections to just running off the MAF, fuel tables, etc.
 

·
Registered
Joined
·
1,267 Posts
Discussion Starter #137
Haven’t driven the car since the day I did the trans fluid drain and fill and oil change with 5w40 t6 rotella. Last Thursday I think that was. Cel came on for engine coolant temp sensor so I got one of those on the way. Not so sure my fans are kicking on now. Coolant temp was 207 on a hot restart. Anything over 205 and I get a warning from the lcdbc. Dash guage showed fine though so hopefully it’s true that the engine temp sensor is the bad one.
 

·
Registered
Joined
·
1,267 Posts
Discussion Starter #138
What injector cleaning service do you guys recommend? I bought a 2001 montero limited like I don’t have enough Mitsubishi problems and of course it’s f’d lol. Doing plugs and wires, upper lower intake gaskets and injector seals. Getting a code for random misfires started all this. Thinking it would be smart to just send the injectors out now. Last place screwed me over lol!
 

·
Registered
Joined
·
1,267 Posts
Discussion Starter #139 (Edited)
So today I drove the car again, due to winter this car is driven once or twice a month. I just found out that if you set the start and end psi of the lcdbc with end psi higher than your duty cycle it doesn’t care how low your duty cycle is! Getting code for engine temp sensor, which just showed up. The lcdbc would show what the ecu see’s for engine temp right? There’s more then one temp sensor, one is for the gauges, then one is for the ecu as far as I know. I’m not sure why mine is bad because it’s showing up on the lcdbc with pretty normal numbers. Unless the lcdbc is getting that info from somewhere else?

@TurboSinceBirth, I saw your reply in that other thread. My car is driving ok but all fuel trims are 81 lol. At idle my wideband is showing low 16’s high 15s and driving regularly it’s around 14.4-15.3 trying to target 14.7. I do have that aem afpr and fuel psi is around 34-35 at idle. Which is in range.

I am curious what I could bump fuel psi up to in order to make stock 360s as big as they can go but I think 390 is as big as you can make them?
 

·
Registered
Joined
·
667 Posts
I just found out that if you set the start and end psi of the lcdbc with end psi higher than your duty cycle it doesn’t care how low your duty cycle is!
I'm pretty sure I warned you specifically about that, because I found it out for myself too. It makes for a pretty wild ride as the LCDBC slams the wastegate open and closed, doesn't it?
 
121 - 140 of 164 Posts
Top