Fuel System Chaulk Talk - Long Term Fuel Trim

[SUMOTOY]

Before I put further comments forth, let's identify the two types of EFI systems on the 93> 80 and 100 series trucks. Prior to 96(93>95.5), Mr. T used a Bosch type L Jetronic system. This includes the use of a Air Flow Meter (AFM) to measure airflow. With flap angle and air temp (sensor is inside the AFM), the computer can calculate the Mass of Air entering the engine.

96> trucks switched to a Bosch LH Jetronic type system. This basically replaced the AFM and ambient temps sender, with a Hot Wire venturi Mass Air Flow (MAF) sensor. This caused less restriction in the measurement of air, and it was a more accurate measure since it measures air density, the computer isn't calculating it.

I will split off my referencing these systems as L-AFM and LH-MAF, since some of the more technical issues and 'fixes' may apply to one or the other (sometimes both)

LT: I understand you feel rising rate FPR isn't the way to go, but you might also want to consider that every single factory turbo or supercharged vehicle uses a rising rate FPR. The reason for this is that it's a MAP (manifold absolute pressure) based measure of *load*, and much more accurate really than a reference tabled measure of load.

Dan: It appears you are lean. First, I'd ck to make sure that your fuel pump passes the resistance test, then I'd check your voltage at the pump (and your battery system voltage). Many times system pressure reductions are present when it's actually a voltage problem (lots of audi CISE-III experience talking). I'd also make sure that the WOT switch circuit is working properly, and that the switch passes the resistance test (IME 3ohms or more resistance in WOT switch affects WOT operation).

Second, AFM really only measures to about 3500-4000rpm, after that the flap is open, and the fuel is delivered via tables, so the AFM can't measure 'additional' air above 4000rpm (hmm sound familiar?). Put another way, all air flow over 4000rpm is the same to the AFM. The computer then only looks for 5 volt input switch from the AFM (AFM is wide open), and it goes to the rpm vs injector duration tables.

An excellent candidate for a rising rate fuel pressure regulator. Since you don't have the ability to change those tables, a fixed duty cycle vs rpm will deliver more fuel with either a larger injector, or a rising rate fuel pressure regulator. Since larger injectors have no 'tuning' ability, the BEGI RRFPR is a better choice, and you will need it whether you add injectors or not IMO/E. Get it first before you work with injectors. Ditch the methanol, you shouldn't need it.

Colin: I have several applications of Javad Shadzi's 034 running around, and worked with him on the very first CIS turbo to EFI turbo conversion (his system was originally designed for CIS non turbo to EFI turbo). The problem I have with it is no knock input. Actually that's the problem with all aftermarket EFI systems IMO. Now, I understand that someone close by you has cracked the motronic code to eliminate the MAF and convert it to speed density. That leaves all the safety features of motronic intact. A big development in EFI tuning, IMO.

Right now, my suggestion to anyone looking to add more fuel under boost, is to get the BEGI rising rate FPR before you spend a penny on any other piece of hardware or software. Even if you choose larger injectors, without FTU software tuning ability, the only way you will be able to dial in those injectors for on boost *load* is via fuel pressure. This comment applies to both L-AFM and LH-MAF toyota systems.

ST

 

[SUMOTOY]

I'm not sure exactly what you mean here. I thought WOT is open loop and therefore can't be used to calculate LTFT. Am I missing something?

WOT is open loop. However, the LTFT doesn't know or care if you are open loop, it just samples away. So you start dumping WOT fuel into open loop operation on OBDII, you can still trigger LTFT codes. Some of the later programs turn off LTFT sampling at WOT, but I don't think that makes a beans worth of difference. If you are dumping too much fuel at WOT, unburnt fuel will be in the cats a long time before cat efficiency comes back.

Why? Because OBDII cars are still restricted to that +/15% sampling window, including at WOT. It's not a bad thing really, if you are dumping more than .86 lamda for an extended period of time, you will trash the cats. LTFT pulls back the basline DC of the injectors so at WOT the fuel is trimmed back. Better done with a fuel pressure regulator and/or injector swap granted...

ST

 

[landtank]

LT: I understand you feel rising rate FPR isn't the way to go, but you might also want to consider that every single factory turbo or supercharged vehicle uses a rising rate FPR. The reason for this is that it's a MAP (manifold absolute pressure) based measure of *load*, and much more accurate really than a reference tabled measure of load.

I think you have me wrong on this. I'm not against using a rising rate FPR, I actually have a new one sitting on the floor next to me. What I am against is using it as a first strike against a lean situation where there might be alternatives available.

I got the Supra injectors for the situation that Dan is experiencing. These have 6% greater flow which might be just enough to releave his situation. Now the smaller pulley situation might be another story.

Since this is my first time through with evaluating the LC for forced induction, I'm taking it step by step so I only need to do this once and I want to be sure it comes out right when finished.

Don't take this personal. I do this with every piece of equipment I service. I question everything and everyone, even when I know the answer, to evaluate what they know. While I might disagree with you now it doesn't mean I won't agree with you later.

 

[Whipp]

WOT is open loop. However, the LTFT doesn't know or care if you are open loop, it just samples away. So you start dumping WOT fuel into open loop operation on OBDII, you can still trigger LTFT codes. Some of the later programs turn off LTFT sampling at WOT, but I don't think that makes a beans worth of difference. If you are dumping too much fuel at WOT, unburnt fuel will be in the cats a long time before cat efficiency comes back.

Why? Because OBDII cars are still restricted to that +/15% sampling window, including at WOT. It's not a bad thing really, if you are dumping more than .86 lamda for an extended period of time, you will trash the cats. LTFT pulls back the basline DC of the injectors so at WOT the fuel is trimmed back. Better done with a fuel pressure regulator and/or injector swap granted...

ST

How can LTFT sample if the ECU is in open loop? By definition, if it's in open loop, there's no feedback to the ECU. My understanding is that LTFT is used to compensate for fuel system components that have lost calibration etc. as the vehicle has aged. This is calculated while in closed loop. The trim is then applied to the fuel map table values whether in open loop or closed loop.

A better description that matches my understanding can be found here:
http://www.autoshop101.com/forms/h44.pdf

 

[SUMOTOY]

How can LTFT sample if the ECU is in open loop? By definition, if it's in open loop, there's no feedback to the ECU. My understanding is that LTFT is used to compensate for fuel system components that have lost calibration etc. as the vehicle has aged. This is calculated while in closed loop. The trim is then applied to the fuel map table values whether in open loop or closed loop.
...

Here's a quote from Toyota as to how Mr. T vaguely defines it....
"Long Term Trim is a learned value over time which changes gradually in response to conditions such as fuel oxygen content, engine wear, air leaks, variation in fuel pressure, altitude and so on. Long term trim is a component of what Toyota technical literature refers to as the "Basic Injection Duration". Basic Injection Duration data is stored in a nonvolatile RAM and is not erased even when the engine is shut down. This information is used during warm up and wide open throttle conditions."

As I said in the earlier post, WOT is part of LTFT, and I should have let it stay without further comment. Sustained WOT will affect LTFT in one of the two scenarios I posted. Either, LTFT is still active during WOT, OR LTFT is active immediately after WOT. My point being that in either case, too much fuel in WOT will cause LTFT to go negative. Even if we knew that Mr. T doesn't sample during WOT, immediately following a WOT event, LTFT will sample. In the case of sustained WOT dumping fuel, the post WOT event LTFT sample will show a massive rich condition leading to a rivised Basic Injection Duration. I see this all the time when too large an injector is used.

Why I specifically wanted to use WOT is part of LTFT is addressomg the Toyota quoted reference. As a rule Basic Injection Duration is used in the WOT fuel tables. IME, WOT calculations (fixed or Duration vs RPM) are written:
Basic Injector Duration + tabled increase (* rpm correction factor) = WOT Duration

ST

 

[SUMOTOY]

I think you have me wrong on this. I'm not against using a rising rate FPR, I actually have a new one sitting on the floor next to me. What I am against is using it as a first strike against a lean situation where there might be alternatives available.

I got the Supra injectors for the situation that Dan is experiencing. These have 6% greater flow which might be just enough to releave his situation. Now the smaller pulley situation might be another story.

Since this is my first time through with evaluating the LC for forced induction, I'm taking it step by step so I only need to do this once and I want to be sure it comes out right when finished.

Don't take this personal. I do this with every piece of equipment I service. I question everything and everyone, even when I know the answer, to evaluate what they know. While I might disagree with you now it doesn't mean I won't agree with you later.

I take nothing in fuel systems personally. It's math and physics calculated by a computer box. That said, here's my take on larger injectors under low boost...

The problem I see with this approach is the philosophy that you have no ability to recognize/compensate for engine load under boost. Engine load recongnition/compensation is key to a boosted engine. A larger injector only tries to flow more fuel accross the tables (given DC, given Fuel pressure, + larger injector = more fuel all the time). I'd venture that you'd want more fuel under boost (or near boost - see BEGI 2035), and the SAME fuel under vacuum.

That can't be accomplished with injector sizing alone. Add a FPR or FTU software code, yes. However, if you need to install the FPR anyway, do it first. If it solves the fuel problem, put the injectors back on ebay.

ST

 

[landtank]

"Long Term Trim is a learned value over time which changes gradually in response to conditions such as fuel oxygen content, engine wear, air leaks, variation in fuel pressure, altitude and so on. Long term trim is a component of what Toyota technical literature refers to as the "Basic Injection Duration". Basic Injection Duration data is stored in a nonvolatile RAM and is not erased even when the engine is shut down. This information is used during warm up and wide open throttle conditions."

I don't see where this quote indicates that LTFT is being altered during open loop. It looks to me that it indicates that it's used during warm up and WOT.

Does that mean that everything in between is closed loop?

 

[Whipp]

Even if we knew that Mr. T doesn't sample during WOT, immediately following a WOT event, LTFT will sample. In the case of sustained WOT dumping fuel, the post WOT event LTFT sample will show a massive rich condition leading to a rivised Basic Injection Duration. I see this all the time when too large an injector is used.

ST

OK, that makes sense to me. I couldn't see how the mixture during open loop could affect the trim, but since the EFI system resumes closed loop when the throttle is closed, I can see that it could see a rich scenario then and trim back.

I wasn't trying to be antagonistic, I just didn't see how LTFT could be affected in open loop.

Thanks.

 

[landtank]

I take nothing in fuel systems personally.

You really need to sharpen your comprehension skills.

 

[landtank]

OK, that makes sense to me. I couldn't see how the mixture during open loop could affect the trim, but since the EFI system resumes closed loop when the throttle is closed, I can see that it could see a rich scenario then and trim back.

If this is happening the new LTFT value would be short lived as after sustained closed loop it would just adjust it back to the previous value it had before the WOT situation.

This does however explain my situation with my truck this passed year. I had a hot running issue and drove the truck like that on 2 two hour drives. I then corrected the problem and drove it on another 2 hour drive for vacation. The very next day after that repaired 2 hour drive I had a high idle situation once warmed up that took several drive cycles to correct.

 

[Whipp]

OK, that makes sense to me. I couldn't see how the mixture during open loop could affect the trim, but since the EFI system resumes closed loop when the throttle is closed, I can see that it could see a rich scenario then and trim back.

I wasn't trying to be antagonistic, I just didn't see how LTFT could be affected in open loop.

Thanks.

Hmmmm. My work day is full of meetings today, which means I have lots of time to think about this stuff.

LTFT is usually only adjusted when the Short Term Fuel Trim is maxed out for a period of time, usually on the order of seconds. I think a system would have to be significantly rich when it returns to closed loop for the STFT to be maxed long enough. I don't think that 6% injector size changes would do this.

Further interesting reading here:
http://www.motor.com/MAGAZINE/Articles/092006_11.html

 

[SUMOTOY]

Hmmmm. My work day is full of meetings today, which means I have lots of time to think about this stuff.

LTFT is usually only adjusted when the Short Term Fuel Trim is maxed out for a period of time, usually on the order of seconds. I think a system would have to be significantly rich when it returns to closed loop for the STFT to be maxed long enough. I don't think that 6% injector size changes would do this.

Further interesting reading here:
http://www.motor.com/MAGAZINE/Articles/092006_11.html

Good article, and actually a good reference thinking about the piggy back injectors and what they may actually be doing (uneven fuel distribution in the cylinders).

Whipp: I don't believe I proposed that 6% bigger injectors would cause a LTFT OR a STFT code. I think the risk is there that it could, as the injectors are already pretty big in the 80. Again, only advocating that a RR fuel pressure regulator is a better starting point. Without it, you have summarily increased fuel delivery, which really hasn't addressed open OR closed loop operation.

I'm not sure it's really clear that LTFT is ignored in the WOT operation of the 80. Hence I chose my words carefully that LTFT is a part of WOT operation. Again, we should leave it there for now, as there are some other measuring blocks that can help identify LTFT issues.

LT: Mr. T defines Open Loop Mode just as vaguely:
Engine start, engine cold, WOT, fuel cutoff, hard acceleration (not sure the difference between that and WOT, maybe another load parameter - like the rate of rpm climb).

Remember, if you get a LTFT code, it triggers a CEL. That needs to be reset. In sustained WOT operation of dumped fuel, you absolutely can set the LTFT code, btdt. The bottom line is that too big an injector at WOT dumping fuel is not a good thing. The best hope is that closed loop will recover. The worst case is that the cats blow out the internals into the mufflers. BTST on stock normally aspirated vw, bmw and audis.

The gain in finite adjustment of a RR fuel pressure regulator eliminates a lot of the headaches of larger injectors and it adds a MAP based load fuel parameter under boost. Using LTFT to dial in the fuel pressure regulator can be done effectively.

BTW, my tests and concerns noted in other threads concur with the need for the 2035 BEGI regulator. What really causes problems in my truck is the crossover point from vacuum to boost. Specifically I identified lean condition from 0 vacuum to ~3psi boost. Bell identifies this as one of the main advantages of the 2035 RRFPR. This really confirms that larger injectors with the stock regulator will continue to have a problem identifying MAP based engine load at the vacuum/boost crossover point.

From Bell Engineering Website
"Adjusting the 2035
Three adjustment are necessary with this regulator. NA engines that are subsequently fitted with forced induction often exhibit extreme leanness in the range of 4 inches of vacuum to approximately 3 psi boost. The cause for this is that the turbo/sc can achieve atmospheric pressure in the manifold prior to the throttle being fully open. Without full opening, the throttle position sensor will not send the proper signal to the ECU, hence, a lean condition occurs. To counter this we’ve made the “onset” of fuel pressure rise adjustable. The 2035 can actually start raising the fuel pressure prior to being under boost. ..."

This reaffirms my point that money and time should be spent on the RR FPR first, then possibly look at injector sizing.

ST

 

[landtank]

From Bell Engineering Website
"Adjusting the 2035
Three adjustment are necessary with this regulator. NA engines that are subsequently fitted with forced induction often exhibit extreme leanness in the range of 4 inches of vacuum to approximately 3 psi boost. The cause for this is that the turbo/sc can achieve atmospheric pressure in the manifold prior to the throttle being fully open. Without full opening, the throttle position sensor will not send the proper signal to the ECU, hence, a lean condition occurs. To counter this we’ve made the “onset” of fuel pressure rise adjustable. The 2035 can actually start raising the fuel pressure prior to being under boost. ..."

This reaffirms my point that money and time should be spent on the RR FPR first, then possibly look at injector sizing.

ST

Speaking for myself, sales literature is always suspect for me and doesn't indicate need or value.

However I did reference this before buying mine so I could be sure that I'd have the adjustability I might need should I want to install it.

 

[sleeoffroad]

Good article, and actually a good reference thinking about the piggy back injectors and what they may actually be doing (uneven fuel distribution in the cylinders).

ST, how did you get that out of the article?

This reaffirms my point that money and time should be spent on the RR FPR first, then possibly look at injector sizing.
ST

OK, I looked at the BEGI and I am still not sure how you are going to tune the open loop situation. I know this discussion is on closed loop, but whatever is added should work for both cases. We have already done a FMU on a early model truck and could not get the open loop tuned properly.

 

[landtank]

Christo, do you generate an O2 fault prior to tuning?

 

[SUMOTOY]

ST, how did you get that out of the article?

Christo, how do you make sure that your 2 extra injectors are putting the same amount of fuel down each runner? Couldn't you (just like the article referenced) indeed have a STFT and LTFT reacting to the SUM of fuel, and yet have individual cylinders running lean and rich because the fuel flow on a dry manifold is not ideal adding to low fuel atomization from a low duration injector pulse width? Have you done any measurements to know that's NOT happening?

Christo, Colin has put forth a lot stronger opinion on the add-on injectors than I have, but I agree with his thoughts on it completely. On a properly built EFI system, it's just not a necessary component, and without individual cylinder measuring capability, has risk associated with it. That's the nicest way I can think to put it.

OK, I looked at the BEGI and I am still not sure how you are going to tune the open loop situation. I know this discussion is on closed loop, but whatever is added should work for both cases. We have already done a FMU on a early model truck and could not get the open loop tuned properly.

Christo, did you use the 2035 or the 2025? The 35 replaces the stock fuel pressure regulator which should allow a lot more adjustability. How exactly did you route the vacuum line to the FPR?

LT: Bell engineering makes good stuff, and their newer 2035 solution documentation explains well exactly what I've been measuring. Reversing the logic a bit, I think it would be more difficult to find any evidence to support larger injectors on an add-on forced induction system WITHOUT any change to the FPR. Every production Forced Induction system uses a RR FPR.

Moot point really, since you own one, you see the value. I'm advocating that it goes into a forced inducted truck with testing before you think of the injector swap. Bell has identified crossover issues from vacuum into boost on aftermarket boost systems. Since I've measured it on my truck with an aftermarket boost kit, it markets an identified need IMO. There is a lot of tuning with a 2035, not sure why anyone would want to add injectors as another variable. If, after adding the 2035 you still identify a lean condition injectors may be something to consider.

What size injector did you calculate was necessary with a RR FPR installed? If you just install larger injectors, what happens to fuel at the crossover point?

ST

 

[landtank]

I actually got the 2027, it goes after the stock regulator. I've been actively watching EBay for various components. Basically I have some time before the build and when I see something at a good price I'm picking it up. The BEGi I got was 2 years old unused for $90.00, IIRC.

I figured when the time came to fire the truck up after the install it would make things easier if potential items where waiting in the wings.

 

[sleeoffroad]

Christo, how do you make sure that your 2 extra injectors are putting the same amount of fuel down each runner? Couldn't you (just like the article referenced) indeed have a STFT and LTFT reacting to the SUM of fuel, and yet have individual cylinders running lean and rich because the fuel flow on a dry manifold is not ideal adding to low fuel atomization from a low duration injector pulse width?

I do not, however here I have to refer to AVO's experience and the number of systems they have running in the middle east that is trouble free. Again, remember we did not design the kit, we adapted it for the US trucks.

Have you done any measurements to know that's NOT happening?

I have not measured it. Agreed one could, but I would say that if the system was problematic in this way, we would have known about it from AVO.

Christo, Colin has put forth a lot stronger opinion on the add-on injectors than I have, but I agree with his thoughts on it completely. On a properly built EFI system, it's just not a necessary component, and without individual cylinder measuring capability, has risk associated with it. That's the nicest way I can think to put it.

So does adding a FPR, although probably on the drowing side.

Christo, did you use the 2035 or the 2025? The 35 replaces the stock fuel pressure regulator which should allow a lot more adjustability. How exactly did you route the vacuum line to the FPR?

It was not a BEGI. I can not recall the exact line routing. It was 4 years ago.

LT: Bell engineering makes good stuff, and their newer 2035 solution documentation explains well exactly what I've been measuring. Reversing the logic a bit, I think it would be more difficult to find any evidence to support larger injectors on an add-on forced induction system WITHOUT any change to the FPR. Every production Forced Induction system uses a RR FPR.

I have no doubt that they make good stuff. However I have seen what these trucks do on Open Loop with both the supercharger and turbo. I just know that we were able to tune it better with the extra injectors and piggy back computer vs rising rate regulator.

Moot point really, since you own one, you see the value. I'm advocating that it goes into a forced inducted truck with testing before you think of the injector swap. Bell has identified crossover issues from vacuum into boost on aftermarket boost systems. Since I've measured it on my truck with an aftermarket boost kit, it markets an identified need IMO. There is a lot of tuning with a 2035, not sure why anyone would want to add injectors as another variable. If, after adding the 2035 you still identify a lean condition injectors may be something to consider.

What size injector did you calculate was necessary with a RR FPR installed? If you just install larger injectors, what happens to fuel at the crossover point?

I am not advocating a injector swap. There are tons of trucks running around (95 and later) with the SC that run ok, does not trip CEL's and does not run lean.

I have personally bought 2 93 trucks that had SC on them for less than 100 miles with blown motors. We have also installed SC on earlier trucks and have experienced lean running etc. In fact, we do not recommend a SC install on a earlier truck. I think it is important to make this disctinction.

My whole interest in this is to see how you are going to back the fuel out at open loop under higher RPM and Load. It goes against what a FPR does. However when we tune with the piggy back and extra injectors, we do not have to add as much fuel under those conditions as under high load, lower RPM

 

[sleeoffroad]

Christo, do you generate an O2 fault prior to tuning?

Rick, we never run them for long enough to get that. After getting the trucks running, we start the tuning immediately in baby steps to ensure I do not have to buy the customer a truck.

My 100 series is running for 10k with the same setup as the 80 (different tuning) and no check engine lights. We have another 80 running with a customer that has about the same amount of miles, with no issues either.

 

[SUMOTOY]

I have no doubt that they make good stuff. However I have seen what these trucks do on Open Loop with both the supercharger and turbo. I just know that we were able to tune it better with the extra injectors and piggy back computer vs rising rate regulator.

I think the 2035 is worth it's weight, and that's my route. Remember, the RR FPR was Corky's toy for the last 30years. If you used a copy of his toy, I doubt you got close to the abilities of the 35. Fine tuning of the fuel pressure regulator will give equal fuel to all injectors. Open loop requires more attention because of how the FTU calculates it. Could it be SMALLER injectors would solve the problem? Coming clean, my spreadsheets indicate this might be a real possibility. I'm still in disbelief frankly, that Mr. T put 29lb injectors in the 80!

I have personally bought 2 93 trucks that had SC on them for less than 100 miles with blown motors. We have also installed SC on earlier trucks and have experienced lean running etc. In fact, we do not recommend a SC install on a earlier truck. I think it is important to make this disctinction.

I might advocate that TRD do a better job of warning that the AFM and computer have to relearn the new airflow on the early trucks. In my install, I got ck engine light, pinging, bucking and the worst performance I could imagine under light to moderate load (hills of Iowa). My buddy reminded me to look at pp 18 of the manual:

"Pings during acceleration > Computer has yet to adjust to supercharger > Drive serveral hundred miles in different driving modes (For example: Not at steady state highway cruising)"

I know during that time, had I decided to push thru the knock, bad things would have happened to my motor based on turbo experience with those symptoms. The fact your purchases were less than 100miles indicates to me, that maybe others didn't abide by the implied learning process of the FTU, or just figured to ignore (what I considered) the warning signs of engine failure.

My whole interest in this is to see how you are going to back the fuel out at open loop under higher RPM and Load. It goes against what a FPR does. However when we tune with the piggy back and extra injectors, we do not have to add as much fuel under those conditions as under high load, lower RPM

That's consistent with my measures as well. I identified the real problem as peak torque under less than WOT. Taking that further, with add on forced induction, we know to have 3 possible fuel trouble points in the stock programing of the FTU in J-AFM trucks.
* ~0 - 3psi crossover vacuum/boost and/or
* max AFM output event causing a fixed airflow table shift and/or
* less than WOT + boost (since the stock truck can't differentiate this isn terms of 'load')

On the LH-MAF trucks, by defintion, we have those three issues as well, excepting however, *possibly* a higher rpm threshold measure for MAF output before fixed MAF input tabling.

Christo, I see and agree that 'tuning the fuel' is the issue given the above. I don't agree that it's necessary or good to use piggy back injectors to accomplish that task. Maybe less confrontationally put, I see many advantages to addressing fuel issues with 6injectors vs 8 on a 6cylinder motor.

ST