RORERI: A JDM 1992 ST185H-BLMVZ

underscore

Well-known member
I'm not sure where the best spot to tap in is, I'm just hunting for options to add an easy way to drain bad fuel without risking a fuel leak either.
 

Roreri

Active member
Ah, understood. I picked the top of the fuel filter as one of two options suggested in a recent thread about installing a fuel pressure gauge.

If you were to get a passthru Banjo bolt then attach a male to male 1/8” NPT fitting and rubber hose to the top of it and run it down, you could probably do it that way. In my case that would not work as the fuel system does not pressurize when the key is turned to the ON position. Perhaps a JDM ECU peculiarity.

I am learning stuff every day, which is both great and a little overwhelming.

I observed that residual fuel pressure gradually drops to zero after shutting down. It immediately drops from 32-34 to 29psi, then to to 26psi in ten minutes, and after an hour, it had dropped to 12psi. Some time between an hour and three and a half hours, it drops to zero.

Is this pressure loss and rate of loss typical?

If not, or if the rate of loss is higher than normal, perhaps my fuel system is compromised in some way and the 45psi max I observe under full load is not the fault of the fuel pump but the fault of the fuel system being unable to sustain pressure.
 

grip-addict

Active member
It's normal to expect the fuel pressure to drop when the pump's not active. We have a return-style system, where the pump in the tank pressurizes the line (And with modified pump circuits, the pump always runs flat out at max flow). The fuel goes through the lines, hits the fuel filter, then the rail/injectors, then the regulator before making the return trip to the tank. The return's always open and available, and all the regulator does is just compress a diaphragm/spring to increase pressure downstream of itself. As long as you're getting your sensor feed before the regulator, that should be a good reading. And it would make sense that pressure would dissipate after the pump is no longer pressurizing the line.

It would be nice if you were able to determine how your fuel pressure compares to other gen2 3s owners. Possibly someone on mr2oc has it logged if it's not recorded here. I also wonder what your base fuel pressure is; you can see what it is by unplugging the vacuum line to the FPR while the car is idling and see what the pressure is. Or just do the fp b+ bridge when it's off. I don't think you need to do it as a troubleshooting step at this point though, I'm just curious.
Actually, down this thought process, do you have an air compressor? You could manipulate the FPR by putting varying pressures on it and seeing how the fuel pressure reacts. Maybe your fuel pressure DOES get to the target but it takes a while to get there. A steady-state amount of air pressure would help you determine that.

And just to hit one of the last convo topics - water injection just gets you free octane boost basically. It's a really neat concept but it's expensive to install. The best setups in my opinion inject water directly into each intake manifold runner, which obviously would not be fun to do on our cars :\ I miiiiight kick it around as an idea whenever my motor gets pulled for service, but I'd have to find a place to mount the tank. That, or just steal water from the windshield washer res.
 

Roreri

Active member
That's good to know about the fuel pressure slackening over time. So, nothing to worry about.

Fuel System Flowchart (Check for Confirmation):

I think the fuel system is set up like this:

Tank-->Sender to Fuel Gauge
Pump
Fuel Line From Tank-->Charcoal Canister
Fuel Filter-->Fuel Pressure Gauge
Fuel Rail
Injectors-->Cylinders
Fuel Pressure Regulator
Return Line to Tank

2Gen 3S-GTE Fuel Pressure:

Various forum discussions I've found regarding the 2Gen 3S-GTE say stuff like:

Ignition to ON position, but engine off: 33-38psi--I do observe this range when starting the car.
FP and B+ jumped Ignition to ON, but engine off: 33-38psi
Idle: 27-31psi, or 28-30psi--I observe 34psi, which is a little high. Upgraded fuel pump? Something else?
"Stock Fuel Pressure" (while under load?): 40-42.5psi
"pump running full blast and the engine off (no vacuum on the regulator)": 40-43psi


I'll check the base fuel pressure with FP +B and ignition to ON tomorrow morning.

Fuel System Limitations:

This is a good old discussion about max boost and fuel system limitations:

viewtopic.php?f=10&t=3059

Especially ChrisD's reply which seems pretty full of useful detail, provided his facts are straight.

Here is his opinion on Fuel Pressure: "It is VERY important to calculate fuel pressure by using base FP + boost pressure. Both combined contribute to pressure in the fuel rail due to the 1:1 FPR."

So, in my case: If my max fuel pressure is 42.5psi, then base of 34psi + 8.5psi boost is my limit. So...thought here...I'm only seeing 42.5 because I'm only achieving that much boost. If I achieve more, I'll see more pressure up to the fuel pump limit and the injector duty cycle. That might be DUH to the wise, but I'm just now grokking these concepts with the aid of instrumentation.

This observation from ChrisD is particularly telling because it seems to speak directly to my observations: "I think most people can run 12-14psi safely on the stock turbo, as long as everything is in good condition and they *pay attention* to how their car responds to the boost. If there is anything like hesitation, loss of power like its being "held back", stumbling, etc., you might not be able to run as much power."

So, whatever the case, it seems evident that I'm outrunning my fuel pump or injectors when I try to go for the higher boost levels. The Blitz Twin SBC can boost beyond the setup's limitations. I observe plenty rich conditions. Low 11s under full load at up to 8.25psi stock boost (sometimes it will give me .64kg/cm2--9psi).

Here is a representative full load onramp run: https://youtu.be/Sr-txGnDAgE

Full rich (11.3:1) under stock 8.25psi boost at 5000rpm. 42.5psi fuel pressure. I think my base is 34psi. That's what it sits at on the gauge at idle. Which suggests something, because all the discussions seem to center around 30psi as the base for a 2Gen 3S-GTE. Upgraded fuel pump? Different FPR?

I could perhaps adjust the "high" boost setting (the Blitz Twin SBC has a dial knob for the "low band" and "high band") down to like 12psi and see how it does.

US 93 octane E10 fuel is almost certainly sucking for the JDM ECU which is probably always asking in its silent little digital Japanese voice: "Why you cheap out on gas, asshole?"

Fuel Pressure Regulator

I found this an interesting and informative article and video from Haltech discussing the Fuel Pressure Regulator:

https://www.dragzine.com/news/under-pre ... ith-boost/

Water/Meth Injection

Look...any time I think about being able to say "This car is on meth" with a straight face it makes me giggle. There's gotta be a bumper sticker. Plus, for whatever reason I'm not attracted to an FMIC. So, water/meth injection to achieve proper octane and correct IATs seems like a fine approach to mitigate U.S. gas octane deficiencies and get the most that I want out of this car: Stable 14psi boost, which is the limit of turbo efficiency and probably coming close to the advisable limit for stock 30 year old internals.

Though to be honest I have no idea about the engine build. Could be stock. Could be forged pistons and uprated injectors. Though why would anyone do that and keep the TMIC?

The stealing water from the windshield washer idea for a water/meth injection kit is of course classic. It is quality discourse like this that I come here for. Because most videos show people installing their tanks in the rear compartment and the pump in the spare tire well and running a line up through the cabin and through the firewall and yaddayaddayadda YAWWWWWNNN. Stealing from the windshield washer bottle you can use the blue 30/70 mix stuff too. If you have no ABS, there's some room there at the front right corner, too, that could fit a tank and pump.

I agree that a direct to manifold approach would be best, but God what a PITA. F that. From what I've read the best approach without going full manifold is to inject as far from the throttle body as possible on the engine side of the TMIC--so maybe actually on the intercooler outlet. Or...just skip the TMIC and let the water/meth do the charge cooling.

I wanted to know more. So I watched this: https://youtu.be/SgzDRaa4xLc

My takeaways: Water/Meth injection lowers charge temperatures and increases octane making higher boost possible. But it riches out the air:fuel ratio overmuch, sapping power. So, a tune might be necessary, to take fuel out to compensate for that. Might be possible to spray a little water/meth and get away with not having to do a tune, though. My goals are modest: Stable 14psi boost, elimination of possibility of detonation.
 

grip-addict

Active member
Roreri":2ks666xc said:
Ignition to ON position, but engine off: 33-38psi--I do observe this range when starting the car.
FP and B+ jumped Ignition to ON, but engine off: 33-38psi
Idle: 27-31psi, or 28-30psi--I observe 34psi, which is a little high. Upgraded fuel pump? Something else?
"Stock Fuel Pressure" (while under load?): 40-42.5psi
"pump running full blast and the engine off (no vacuum on the regulator)": 40-43psi


I'll check the base fuel pressure with FP +B and ignition to ON tomorrow morning.
Are the running numbers with the vacuum hose still attached to the fpr? When at vacuum, the fuel should have less pressure than the "base" pressure. Base pressure is whatever the fuel pressure is when the pump is in high-voltage mode and when the FPR doesn't have any vacuum applied to it.

ChrisD":2ks666xc said:
Here is his opinion on Fuel Pressure: "It is VERY important to calculate fuel pressure by using base FP + boost pressure. Both combined contribute to pressure in the fuel rail due to the 1:1 FPR."

The idea is that after the intake manifold starts seeing positive pressure (something an NA car will never experience), there would start to be resistance pushing back on the injector. Adding 1psi of fuel pressure for 1 psi of boost solves that as the ratio of base pressure to manifold pressure should stay the same. It makes math calculations easy for the ECU and the injector doesn't have to stay open longer than it would normally need to (if fuel pressure didn't increase, that would be the requirement, and there is definitely a finite amount of time an injector can stay open, thus limiting power/safety/etc of an engine).
Modern ECU's that track fuel pressure can overcome deficiencies like this and keep the injector open longer than the map calls for if they see a lack of fuel pressure. Unfortunately, that wasn't a capability when our cars were manufactured.

Roreri":2ks666xc said:
Here is a representative full load onramp run: https://youtu.be/Sr-txGnDAgE

Video's private. Don't worry, I make this mistake too lol.

Roreri":2ks666xc said:
US 93 octane E10 fuel is almost certainly sucking for the JDM ECU which is probably always asking in its silent little digital Japanese voice: "Why you cheap out on gas, asshole?"

E10/93 Would be safer for the car to run than E0/93 since the ethanol keeps the burn a bit cooler. Don't beat yourself up that you're not putting in zero ethanol gas.

Edit - I watched the Haltech video. Scott does a much better job explaining fuel pressure differential than I did 8)
 

Roreri

Active member
Video is unlisted, now, for certain sure--you should be able to watch it from the link.

https://youtu.be/Sr-txGnDAgE

I have no idea whether the vacuum hose was still attached for those fuel pressure claims, but I get now what you're saying.

Great to hear about the E10/93 being just fine for daily use.

The more I learn about water/meth injection, the more I think it's a good idea to get control of intercooler temperatures eliminate the possibility of detonation and achieve max turbo efficiency, which will give just enough thrust to weight (especially as I get toward 3000lbs fuel tank empty weight with a 12lb carbon fiber hood and lightweight Shorai battery) to be fun without being crazy.

Tippo, I missed your invitation--that would be great! I don't have much longer here in the east, though. Moving at the end of June. I have a couple more projects I want to see done before I go. Carbon fiber hood, lightweight battery, get the AC compressor and receiver dryer replaced.
 

alltracman78

Active member
Roreri":2fyk69q3 said:
I took the Lorelei on a number of loops onto the expressway and off and back through the neighborhood, running on stock boost max 9psi. I gave it a number of full load onramp accelerations and some good load on the expressway. Did some video recording so I could analyze afterwards--too many gauges to look at all at once. My findings were as follows: 32psi at idle, 40psi at load, up to 45psi max.

This is lower that what grip-addict posted should be observed:

"At full vacuum, 30-32 something-ish psi fuel pressure
At 0 bar boost, 42-44ish fuel pressure (please review bgb for exactly numbers)
At 14.7 psi boost, 58ish psi fuel pressure.
1:1 is the law of the land."


Performance was generally good up through 9psi. Air-fuel ratio was properly rich under load. One time, I experienced dropout, while at full load--wasn't recording then so I don't know what exactly was going on, but I'm suspecting fuel starvation. I think I should be seeing above 50psi fuel pressure at 9psi boost.

Per BGB page FI137
Fuel pressure (FP) @ 0PSI. If your engine isn't running you need to jump FP and B+ with the key on to make the pump run. Or if the FP regulator vacuum hose is disconnected with the engine running. Or if the engine is running at idle with FP regulator vacuum hose connected and intake manifold pressure is 0 PSI. (Does that all make sense? ).
So
FP @ 0PSI 33 - 38 PSI (This is the "base pressure". The pump doesn't need to be at full voltage to achieve this, the amount of fuel needed is low enough it can supply it at lower voltage. )
FP @ idle 27 - 31 PSI
The 3SGTE FP regulator is a 1:1 rising rate regulator. Every 1 PSI extra in the FP regulator diaphragm means 1 PSI more fuel pressure.
So if your "base" is 32 PSI and you're at 10 PSI of boost you should have 42 PSI of fuel pressure. Keep in mind the regulator and gauge aren't perfect though. So if the gauge says 41 or 43 it's probably fine. If you're that concerned buy a $400 liquid filled NASA spec gauge (it is nice to have something very accurate to compare other tools to FWIW).

Keep in mind these are rough specs. And the factory regulator isn't going to go bad and give you too high a pressure unless it is completely stuck closed. So if you're a little high the system is working fine. Also, IIRC the 3SGTE FP tends to run a little higher than what the manual calls for. Again, not a bad thing.
Once again, keep in mind with any measuring device there's a tolerance. So your 34 PSI on your autometer gauge might actually be 32 PSI. Or 36. I wouldn't get hung up over 1 or 2 PSI necessarily.


Roreri":2fyk69q3 said:
If you were to get a passthru Banjo bolt then attach a male to male 1/8” NPT fitting and rubber hose to the top of it and run it down, you could probably do it that way. In my case that would not work as the fuel system does not pressurize when the key is turned to the ON position. Perhaps a JDM ECU peculiarity.

This is NOT directed at you. Or anyone on here specifically. It's not your fault if you've been fed wrong info. But this is a perfect example of information from the morons that worked on their dads 1947 Ford in their driveway so they think they know about cars. Kind of a sore spot for me because it's one of those pieces of "knowledge" that's passed around and is blatantly false and people obviously haven't actually confirmed it because if they had they would find out it's wrong.
THE ALLTRAC FUEL PUMP WILL NOT RUN UNLESS THE ENGINE IS STARTING/RUNNING OR YOU JUMP FP B+ IN THE DIAG BOX. IT WILL NOT RUN IF JUST THE KEY IS TURNED TO ON. Turning the key to on DOES NOT "prime" the pump.
This goes for pretty much any Japanese car made in the last 35ish years (It's possible there are 1 or 2 exceptions to this. There's always an exception.) That is more of an American car thing (they may not do that any more, been a while since I worked on cars).
The fuel pump only runs if it sees the engine rotating or if it's forced to (EG bridging FP and B+). Helps to prevent fuel pumping out of a leak for obvious reasons.


grip-addict":2fyk69q3 said:
Actually, down this thought process, do you have an air compressor? You could manipulate the FPR by putting varying pressures on it and seeing how the fuel pressure reacts. Maybe your fuel pressure DOES get to the target but it takes a while to get there. A steady-state amount of air pressure would help you determine that.

Good idea but you're better off using a hand held pump. :) An air compressor is going to give WAY too much pressure and airflow for that diaphragm. The other thing to keep in mind is the injectors aren't going to be running, so no fuel flow; the system will be static. So this might give you a rough idea of the condition of the FP regulator (it will show if the diaphragm is shot for example) it's not going to show how the system is behaving under working conditions.


Roreri":2fyk69q3 said:
Fuel System Flowchart (Check for Confirmation):

I think the fuel system is set up like this:

Tank-->Sender to Fuel Gauge
Pump
Fuel Line From Tank-->Charcoal Canister
Fuel Filter-->Fuel Pressure Gauge
Fuel Rail
Injectors-->Cylinders
Fuel Pressure Regulator
Return Line to Tank

2Gen 3S-GTE Fuel Pressure:

Various forum discussions I've found regarding the 2Gen 3S-GTE say stuff like:

Ignition to ON position, but engine off: 33-38psi--I do observe this range when starting the car.
FP and B+ jumped Ignition to ON, but engine off: 33-38psi
Idle: 27-31psi, or 28-30psi--I observe 34psi, which is a little high. Upgraded fuel pump? Something else?
"Stock Fuel Pressure" (while under load?): 40-42.5psi
"pump running full blast and the engine off (no vacuum on the regulator)": 40-43psi


I'll check the base fuel pressure with FP +B and ignition to ON tomorrow morning.

There is a separate air line coming off the tank for the charcoal canister. It doesn't share a line with the pump/supply system.
See my first answer for pressure info. :)
"Under load" is anywhere from just above idle pressure (load is basically airflow or air pressure [in a MAP based system] multiplied by RPM, so you can be coasting at 4K RPMs and have no load because the TB is closed) to max boost pressure. There's no set FP for this. Only a broad range.


Roreri":2fyk69q3 said:
Fuel System Limitations:
So, in my case: If my max fuel pressure is 42.5psi, then base of 34psi + 8.5psi boost is my limit. So...thought here...I'm only seeing 42.5 because I'm only achieving that much boost. If I achieve more, I'll see more pressure up to the fuel pump limit and the injector duty cycle. That might be DUH to the wise, but I'm just now grokking these concepts with the aid of instrumentation.
Heinlen fan?

Roreri":2fyk69q3 said:
This observation from ChrisD is particularly telling because it seems to speak directly to my observations: "I think most people can run 12-14psi safely on the stock turbo, as long as everything is in good condition and they *pay attention* to how their car responds to the boost. If there is anything like hesitation, loss of power like its being "held back", stumbling, etc., you might not be able to run as much power."

So, whatever the case, it seems evident that I'm outrunning my fuel pump or injectors when I try to go for the higher boost levels. The Blitz Twin SBC can boost beyond the setup's limitations. I observe plenty rich conditions. Low 11s under full load at up to 8.25psi stock boost (sometimes it will give me .64kg/cm2--9psi).
This is purely conjecture on my part, so definitely not gospel. But food for thought.
My understanding is the JDM ECUs make more power by more advanced timing and running the engine leaner than US versions.
With US versions you can make more power by running higher boost (this leans the engine out some). To really take advantage you need to bypass boost cut (old school way was fuel cut defender or zener diode) and/or run a larger more efficient turbo.
Also, if you are at max boost then max airflow is reached long before you hit max RPMs. If so, at that point there's no real accurate fuel delivery, it's basically "dump as much fuel as you can".
The JDM and USDM use the same injectors, fuel pump and AFM (with the exception of the homologation models). So in theory they both have the same stock max power potential.
BUT, if the JDM ECU dumps less fuel at max airflow than the USDM one does there is less extra fuel to use if you go above stock pressure. And AFAIK most (or all?) of the old school stock gen II power info is based on US spec ECUs, not JDM. It may not. At max load they may have the same injector duty cycle. But something to at least think about.


Roreri":2fyk69q3 said:
Full rich (11.3:1) under stock 8.25psi boost at 5000rpm. 42.5psi fuel pressure. I think my base is 34psi. That's what it sits at on the gauge at idle. Which suggests something, because all the discussions seem to center around 30psi as the base for a 2Gen 3S-GTE. Upgraded fuel pump? Different FPR?

I could perhaps adjust the "high" boost setting (the Blitz Twin SBC has a dial knob for the "low band" and "high band") down to like 12psi and see how it does.

US 93 octane E10 fuel is almost certainly sucking for the JDM ECU which is probably always asking in its silent little digital Japanese voice: "Why you cheap out on gas, asshole?"

I'm guessing your 15.6 is a little high for max boost.
You probably don't even need to go all the way down to 12. 13 Or 14 should be fine.

If there's a problem with E10 it's not so much with fresh gas. The problem is more with older gas. Either it eats away at certain parts (I don't think that's much of a problem with cars from the last 30+ years) or parts of the fuel evaporate, leaving a slightly different, less "pure" mix. I think the ethanol slightly raises the octane, which means they don't need to refine out as much of the "crap". But if the ethanol (or other more "pure" parts of the fuel) evaporate that leaves a higher percentage of "crap" left, which gives you a slightly lower octane.

Roreri":2fyk69q3 said:
Water/Meth Injection

Look...any time I think about being able to say "This car is on meth" with a straight face it makes me giggle. There's gotta be a bumper sticker. Plus, for whatever reason I'm not attracted to an FMIC. So, water/meth injection to achieve proper octane and correct IATs seems like a fine approach to mitigate U.S. gas octane deficiencies and get the most that I want out of this car: Stable 14psi boost, which is the limit of turbo efficiency and probably coming close to the advisable limit for stock 30 year old internals.

I wanted to know more. So I watched this: https://youtu.be/SgzDRaa4xLc

My takeaways: Water/Meth injection lowers charge temperatures and increases octane making higher boost possible. But it riches out the air:fuel ratio overmuch, sapping power. So, a tune might be necessary, to take fuel out to compensate for that. Might be possible to spray a little water/meth and get away with not having to do a tune, though. My goals are modest: Stable 14psi boost, elimination of possibility of detonation.

VERY interesting video. Looks like you want 50% plus methanol if you want the same/more power. I wish I had time to mess around with this stuff more. Too many hobbies.
Seems like water actually takes away power, even if it lowers temps. Maybe vaporizing the fuel does enough?
I wonder how that work on a stock or slightly modded 3sgte? Maybe the same? Maybe different?

One thing, it doesn't increase octane, it just lowers intake temps.
I wonder if you could run a larger turbo or more boost to lean it out?
More theory than anything else in your current situation IMO.
If water injection will hurt power that much manbe it isn't something you should do right now. Might be better avenues to pursue?

I wouldn't mess with the fuel pump just yet, maybe drop the boost and see what happens.
 

Roreri

Active member
Yes I do like the Heinlein, to the extent that he doesn't get too weird. I like much of his fiction.

As regards the fuel pump: With FP and B+ jumped, and the ignition to ON, it makes 38psi. Idle at 50cmHG vacuum, 19.6inHG (-9.6psi) is 32psi. 0psi boost is 38-30psi. 8-9psi boost is 42.5psi. I'll leave the fuel pump alone for now, and continue to observe. The "Low" setting achieves 11.5 to 12 psi.

I agree that 15.6psi is too high. I'll fiddle with the boost controller settings and see where the happy medium is--I agree that 13-14psi is probably it.

As regards my mistaken belief that by default the USDM fuel pump would pre-pressurize with the key in the ON position without jumping FP and +B, correction is a blessing. I was mistaken, probably having read it somewhere.

I think I agree with you on the lean tendencies of the JDM ECU. I've observed a slight tendency toward lean running conditions with mine.

More Thoughts on Water/Meth

Interesting possibilities with water/meth. The truth is how much more power can you make on water/meth? None. However, if you can avoid detonation, lower temps, and achieve higher boost then maybe you can make more power.

I found this an interesting discussion:

https://www.focusst.org/threads/running ... it.148899/

Key quotes and assertions to be checked up on:

Meth helps by: Cleans valves. Low to no corrections at all on aggressive tunes. Cooler intake temps. "Eliminates" misfires. A tune is NOT required and most don't recommend it. WHY? Because if you are tuned for it and you drive off into the distance and run out...BOOM your motor is gone. If you catch it and either always make sure you have meth on board and NEVER run out and always keep on it or switch maps before the meth runs out. But for as much as I drive I just wouldn't want the constant checking of it in fear that if it runs out, kaBOOM.

"Meth is 100% ok to run without tuning for it as long as you don't spray too much and cause rich conditions...You get a nice octane boost* with pump gas, lower charge temps, valve cleaning, knock suppression and it helps keep fuel pressure up at lower rpms."

You can run plain 'ole water and get a steam clean effect on your valve backsides. You don't have to run meth, or a meth/water blend if you don't want to.

If you do run meth or a 50/50 blend, you get the benefits of cooling, but also an octane boost*.

Do you need tuning for it? No and yes... I did that backwards on purpose.
*No tune, you can benefit from either the cooling aspect or octane boost*. But not really a gain in power.
*Tune for it, you can now increase boost and/or increase timing to take advantage of the cooling effects of water, or the higher octane boost* that meth provides."


*It should be understood that it's not an octane boost, it has the effect of an octane boost.

So, even without a tune there are benefits. An approach I'll explore in research is:

1. Bringing in injection at 9psi
2. Keeping the nozzle small for minimal injection
3. Running 49meth/51water (Boost Juice or similar), alternately Blue Windshield Wiper Fluid 30meth/70water.

There is a little space between the power steering reservoir and relay box. Might could fit a small bottle for water/meth.
 

alltracman78

Active member
Please understand I wasn't giving you a hard time at all about when the fuel pump runs. You're asking questions and trying to learn. And there's absolutely nothing wrong with making a mistake. I make plenty of them.
The problem is when people refuse to check and just keep arguing and insisting they are right and don't have the balls to admit they were wrong, so the confusion stays out there.

I know very little about water/meth injection other than the theory. If it does work in the way outlined in that video it seems pretty pointless except to clean inside your engine (there are probably better and cheaper ways to do it?). There are better ways to keep the engine safe.
But I wonder if it's as simple as it was outlined. And if running it in our engines has the same effect.


By the way, I feel like you had asked about the fuel pump relay a while ago? How to get rid of it maybe?
Just in case, it's the yellow relay in the picture.


This is the factory jumper that comes in the 5th gen 5SFE Celicas in place of the FP relay. May come in the 3SGE ones too? I don't know. It sits in the same place as the FP relay in the Alltrac. All you have to do is remove the FP relay and plug this in. The fuel pump will run on full voltage all the time. Fits perfect and looks stock. In case you're interested.
 

underscore

Well-known member
I had a port for meth injection added to my car, but I ended up not putting the kit in. My thinking was that if the cars could run fine on the crappy gas they had 30 years ago then they should be okay now. I might be wrong though. If it does fix something then it's sort of a bandaid, but if it makes the problem go away does that really matter? Especially since you're already trying to work around having a top mount ATA intercooler with an RC hood.

Personally I'd hunt down that fuel leak first, if nothing else just for safety. I'm assuming if it's leaking gas when full it may be leaking fumes at all times which can be dangerous.

Also don't forget that the stock fuel cut is 14.7psi, so if you're setting boost north of 12-13psi (factoring in tolerances of equipment) that's something to be aware of.
 

Roreri

Active member
@alltracman78: Your words were absolutely taken in the spirit given. That's why I said correction is a blessing. Truth beats ego stroking. My GT-Four is important (bordering on too important) to me and so I'm glad to the truth of the matter. And I'm grateful for the advice and information.

Thanks for the information about the 5SFE Fuel Pump relay jumper. It definitely seems to be a good idea at some point. Any time I can replace a vulnerable and aged relay with a dead simple jumper and get the same or better function I'm interested. I don't want to be lazy but is there a part number I should be looking for? I think it might be called a Sub Fan Relay? Alternately, I think I could just paperclip jump FP and +B if I wanted to be cheap about it.

@underscore: You're absolutely right about the ATA cooler and the RC hood. That's a case of "It's better to look good than to feel good."

And you're spitting truth, which is that I'm avoiding dropping the tank and getting after that leak.
 

underscore

Well-known member
You can just make a jumper out of a couple of male spade terminals and some wire. I wouldn't use a paperclip for that one because I believe it's all the power for the fuel pump running through there, not just a trigger wire. Why the fuel pump power takes such a scenic route I don't know.

I don't know the exact name for the 5SFE jumper but not all of them have it. I was going to swipe it from my old 92 GTS but it had a relay there so it might just be a 90-91 thing?
 

alltracman78

Active member
underscore":cimcgvf2 said:
Especially since you're already trying to work around having a top mount ATA intercooler with an RC hood.

Personally I'd hunt down that fuel leak first, if nothing else just for safety. I'm assuming if it's leaking gas when full it may be leaking fumes at all times which can be dangerous.

Also don't forget that the stock fuel cut is 14.7psi, so if you're setting boost north of 12-13psi (factoring in tolerances of equipment) that's something to be aware of.
Holy crap. I totally forgot he had an ATA intercooler.....

That's a whole bunch of your problem right there..... I wouldn't even run 14PSI with that. You need to get some cooling on that intercooler.
All your doing is superheating the intake air. The intercooler is heat soaking like crazy because there's no where for it to go. The hood over it acts as insulation and keeps all the engine heat in. Then your shoving hot air from the turbo through it. Which is probably cooler than the intercooler itself....

Drop the high boost to 14 PSI. Then put the boost controller on the low setting until you sort something out for the intercooler. I wouldn't waste any more time after that until you do. RC hoods are cool, but are useless if you have an ATA top mount.

As for your fuel leak, you get a puddle when it's full? I doubt it's something in the pressurized part of the system or it would leak all the time. You probably either have a very small leak on the lower portion somewhere (maybe where the filler tube meets the tank?) that only leaks when the weight of a full tank of fuel forces it out, or you the tank is pressurizing and can't vent (fuel cap?) so it's being forced out the EVAP hose. Which shouldn't leak either, it should go to the canister. But it's not as strong as the pressurized hose is. It's also possible one of the seals for the 2? access ports on top of the tank are leaking? I would start with the filler tube though because IIRC you said it was on that side?
Don't be scared to drop the tank, it's not that big a deal. 2 Straps and the fuel lines. You can even take just the straps off and lower it part way to look for starters. Your car came from Japan, they don't use much road salt and it doesn't seem to be that rusty so it shouldn't be a huge problem to get the straps loose.

If you want you can get a couple of 5 gallon fuel cans and fill the tank once it's partially lowered. Keep in mind it might change the center of gravity because the tank isn't symmetrical.
 

alltracman78

Active member
Roreri":8uvp97u9 said:
Thanks for the information about the 5SFE Fuel Pump relay jumper. It definitely seems to be a good idea at some point. Any time I can replace a vulnerable and aged relay with a dead simple jumper and get the same or better function I'm interested. I don't want to be lazy but is there a part number I should be looking for? I think it might be called a Sub Fan Relay? Alternately, I think I could just paperclip jump FP and +B if I wanted to be cheap about it.
Not a necessary mod. I don't even know if it's worth it to do it. How it works is the relay normally routes power for the fuel pump through a resistor, so the pump gets less voltage, and works less, delivers less fuel flow (pressure should still be the same because it's governed by the pressure regulator, not the pump). Less flow does deliver slightly less fuel to the injector (Flow and pressure are related, but different. Topic for another discussion. :) ). Slightly better gas mileage and less wear on the pump.
When you hit WOT the relay bypasses the resistor and sends full voltage to the pump. More flow, ect ect.
The theory is replacing the relay with a jumper eliminates that split second delay between you mashing the pedal and everything happening. Not exactly much of a difference.
I jumped mine because I was having problems with the relay. In that case it's definitely worth it, though you could also order a new relay.

No clue on part number. You'd have to look in the GT/GTS relay parts diagram. Not a fan relay though. :)

Jumping FP is a whole different thing. That's in the DIAG box. Which is in the little alcove directly behind the L strut tower. That relay is in the relay block by the alternator. Bolts to the radiator IIRC. Jumping the DIAG box bypasses almost the entire pump circuit. And it will run any time the key is on. Don't want that consistently.
 

alltracman78

Active member
underscore":3c8hevu7 said:
You can just make a jumper out of a couple of male spade terminals and some wire. I wouldn't use a paperclip for that one because I believe it's all the power for the fuel pump running through there, not just a trigger wire. Why the fuel pump power takes such a scenic route I don't know.

I don't know the exact name for the 5SFE jumper but not all of them have it. I was going to swipe it from my old 92 GTS but it had a relay there so it might just be a 90-91 thing?

FP in the DIAG box isn't a trigger wire. It's the power wire directly to the pump. And you're putting full battery voltage (B+) to it. Doing this bypasses almost the entire fuel pump circuit. 2 relays, the fuse, multiple connections. :) The normal wiring goes the senic route for safety reasons.

A paper clip should be able to handle as much (or more) current than the wire. It's the same size and it's solid, not stranded. I don't like them because there's less surface area on the connection because it's round. Connections in the wiring are flat. I use 2 pins and wires I pulled from a bad O2 sensor. The pin is the correct size for the DIAG box connections.

They probably changed the body harness for the post revision cars. I'm guessing the pre revision ones had the same front body harness, so they needed a way to jump that connection. The 93 wiring manual doesn't show any relay there for the GT/GTS (for the fuel pump).
*The 93 manual still shows a short pin (jumper) for the 2WD. So dunno.*
 

underscore

Well-known member
alltracman78":1fsd84g3 said:
underscore":1fsd84g3 said:
You can just make a jumper out of a couple of male spade terminals and some wire. I wouldn't use a paperclip for that one because I believe it's all the power for the fuel pump running through there, not just a trigger wire. Why the fuel pump power takes such a scenic route I don't know.

I don't know the exact name for the 5SFE jumper but not all of them have it. I was going to swipe it from my old 92 GTS but it had a relay there so it might just be a 90-91 thing?

FP in the DIAG box isn't a trigger wire.

That's what I said, though I'm talking about the fuel pump relay up by the alternator.

alltracman78":1fsd84g3 said:
The normal wiring goes the senic route for safety reasons.

I'm struggling to wrap my head around how having the fuel pump wiring go from the left rear corner of the engine bay, to the right front corner of the engine bay, to the left front corner of the engine bay, back to the right front corner of the engine bay, and then to the fuel pump makes it any safer. I guess if you have a hard enough crash on either front corner you'll damage the wiring and turn off the pump even more?
 

alltracman78

Active member
underscore":2bqrem7r said:
alltracman78":2bqrem7r said:
underscore":2bqrem7r said:
You can just make a jumper out of a couple of male spade terminals and some wire. I wouldn't use a paperclip for that one because I believe it's all the power for the fuel pump running through there, not just a trigger wire. Why the fuel pump power takes such a scenic route I don't know.

I don't know the exact name for the 5SFE jumper but not all of them have it. I was going to swipe it from my old 92 GTS but it had a relay there so it might just be a 90-91 thing?

FP in the DIAG box isn't a trigger wire.

That's what I said, though I'm talking about the fuel pump relay up by the alternator.
Sorry, read too fast.

underscore":2bqrem7r said:
alltracman78":2bqrem7r said:
The normal wiring goes the senic route for safety reasons.

I'm struggling to wrap my head around how having the fuel pump wiring go from the left rear corner of the engine bay, to the right front corner of the engine bay, to the left front corner of the engine bay, back to the right front corner of the engine bay, and then to the fuel pump makes it any safer. I guess if you have a hard enough crash on either front corner you'll damage the wiring and turn off the pump even more?
Didn't give that clear an answer. Too much typing for me.

The safety part is having the circuit opening relay in there. So the pump only runs if the engine is running or starting. That adds some wiring to the circuit.

As for the rest, it's kind of where they can fit relay blocks and route wiring around hot/moving parts. Wiring length isn't usually a priority in design.
Also, they need to follow the wiring harness route. They want to keep everything as bundled together as possible to protect the wiring as much as they can. They don't want random wires running in straight lines all over the car.

It's not as complicated as you think. There are 4 wires. 3 Are basically parallel and the fourth runs a different parallel course.

The first is for the fuel pump relay. A single wire (this is the power wire for the fuel pump) runs from the COR (Circuit Opening Relay) next to the ECU through the dash harness (cowl wire) to the large relay block on the LH foot well where it connects to the engine room harness (Which IIRC is a single U shaped harness that runs from the rear of one fender up to the front of the car and to the rear of the other fender).
This single wire runs along the L fender to underneath the front upper radiator support. Along under the upper radiator support to the R side where it connects to the relay box in front of the alternator and to the fuel pump relay. This is the power input for the relay. The power output for the relay is 2 separate parallel wires (One is for full voltage power, one runs through the fuel pump [FP] resistor as the partial voltage wire) running back along under the upper radiator support. Roughly just to the right of the LH headlight one wire goes down to the FP resistor by the battery then back up to the harness where they splice together to a single wire (Now that it's past the FP resistor it doesn't need to be 2 separate wires anymore), which runs back along under the L fender into the large junction block in the L foot well where it connects to the body harness and runs back to the fuel pump.

The fourth wire is the FP wire and runs from a splice in the dash harness right by the junction block in the L foot well through the dash harness to IG2 (which is one of the 2 large connectors right next to the ECU) which connects the engine harness. It obviously runs along the engine harness to the DIAG box.
If you supply power to the FP wire (B+) it runs back along that wire and bypasses most of the circuit (It bypasses the ECU and starter relay [Which both control the COR], COR, FP relay and FP resistor) and sends power directly to the fuel pump via that splice in the L foot well I just mentioned.

So you have 1 wire running 3 sides of a square, from right next to the ECU along the L fender to the R front of the engine bay. Then 2 wires back 1 side of the same square, where then join into 1 (one of the wires has the fuel pump relay in it, this is why there are 2 up to this point) and follow the second side of the square and then continue back to the fuel pump. And the 4th wire basically doubles back on itself by running on the inside of the fire wall and then along the outside.

So I guess it's more complicated in one way. But not so much zig zag back and forth multiple times.

Holy crap. It always ends up being longer than I intended. But I try to put all the info out there so whoever is interested can learn everything about it.
 

underscore

Well-known member
Wire length adds cost and complexity though. Obviously someone had their reasons and the rest of the development team agreed with it but it seems odd to not just put the fuel pump relay in the main relay box or in the left side kick panel fuse/relay box, right next to all the other things in the circuit.
 

Roreri

Active member
Alexandria VA to Fort Loudon PA and Back Road Trip Report

I'm home safe after completing a 230 mile round trip up into Pennsylvania and back this weekend.

Alexandria to Fort Loudon and Back.JPG
The Lorelei's performance was flawless at stock boost. Started with the rest of a tank of brand indeterminate 93 octane from the Exchange gas station on base, and filled up with non-Ethanol Exxon 93 octane in Mercersburg PA. There's really nothing to say. I had that fuel pressure gauge Gorilla taped to the hood and I observed nothing to be concerned about.

Way Forward

Based on what you said alltracman78, I'm going to just count my lucky stars that I haven't screwed up my ride with my ignorant experimentation. There's the probability that the occasional power dropouts I observed was the knock sensor was kicking in and saving my ass? Or the ECU was pulling timing so aggressively it felt like some other problem? Sometimes it's better to be lucky than good.

Because I already have a carbon fiber RC hood on the way, I really have no "easy" or "stock" engine compartment/intercooler temp solution.

First, no more boost beyond stock with the way it is at the moment. My gratefulness that I'm not doing an engine replacement or piston ring job will help me keep my perspective.

Second, I need to seriously consider some kind of IAT cooling solution. Either an A2W intercooler or FMIC or water/meth injection or some combination. I kinda don't like FMIC for some reason. Maybe I'm being a goober about it. My preferred solution would be to install an ST205 air to water intercooler, though I did a little research and saw that Mishimoto offers a universal water-to-air intercooler and heat exchangers that could work as long as the dimensions sort out, I can get the piping and connectors and I can find a good electric charge coolant pump.

Third, as an immediate action step, I could order up some colder spark plugs?

I don't want to blame Toyota too much, and at the risk of drawing fire on myself because after all I do have the wrong hood, but Toyota kinda screwed the pooch going with the A2A intercooler for the ST185. They went back to the A2W intercooler for the ST205 for good reason, in my opinion.

EDIT: And i just now noticed this @alltracman78. That water/meth injection video, the guy was already on E85. No wonder he saw no power gains on 50/50 water/meth and power losses when using warm temperature windshield washer fluid. With regular 91-93 octane E10 the results would have been different.
 

Roreri

Active member
I got a tip that by clocking the CT26 so that the output is directed downward, I could eliminate about 40% of the piping length of a FMIC setup.

I could see doing that. One of my main reluctances to a FMIC was the lower responsiveness. This could help that, and retain that relatively easy bolt on installation that avoids all the moving parts and the heat soak of a TMIC that you have even with an A2W intercooler.

This thread details the process: viewtopic.php?f=20&t=16034

And apparently I would need a fabbed up bracket for the wastegate actuator. And then it's a matter of rigging up a bracket for the FMIC and finding space for it what with the gearbox oil cooler (not really needed) and radiator fan and A/C receiver dryer up in front.
 
Top