RORERI: A JDM 1992 ST185H-BLMVZ

Roreri

Active member
Was out to get the Lorelei washed prior to going to a show on Sunday--she drives well enough to get there and back.

While driving, I observed that sometimes, I get proper fuel pressure and properly rich conditions--up to 40psi which is in line with what I should be seeing at 8psi boost--but then pressure would precipitously cut to like 24psi.

My guess is I have a failing fuel pump.

I also got the coolant leak locked down. A hose had come loose and I had to secure it. It was a bear figuring out which one though and it would only leak under full operating temps, so it required going for a drive and then backtracing the leak to the back of the engine compartment, which as we all know is a pain in the ass given how crowded the 3S-GTE is in the engine compartment.
 

alltracman78

Active member
Getting a new TPS won't change the fact you have to adjust it when you install it.
Forget all the resistance measurements when you do install it. I'll tell you why in a sec.
Don't forget you have a post revision. It has a throttle opener, which the pre revision don't. You need to apply vacuum to it before you check the TPS.
You don't need to worry about all the checks in the BGB, unless you want to make sure your TPS is working properly. If you just want to adjust it put vacuum to the throttle opener, back probe the IDL wire either at the TPS, or at the ECU if you're more comfortable. With the key on and TB fully closed to .5mm gap you should see 5v. With a .7mm gap at the TB opener you should see 0v at the IDL circuit. I hope that makes sense?

Your pump could be bad. FPR is quicker, cheaper and easier. Neither one is a bad guess IMO.
Tank may be full of crap too. Which would definitely stress the pump.
 

Roreri

Active member
Got it--thanks! That does make sense. At least enough for me to do something with it.

The Tri-Lakes Cruisers car show in Monument Colorado was pretty cool, though man, talk about dominated by domestic autos. Out of over two hundred cars there were two Miatas, two MR2s, two Datsun 240Zs, and two Celicas. That was it for Japanese autos.

Roreri and 1977 Celica GT.JPG
Being out in the sun all day was worth it to get to hang out and see this old warrior. The owner got it for $500, then did all the restoration and paint work, claiming that it was in pretty good shape from the get go as it was a classic case of an old guy selling his passed away mom's Celica that he knew nothing about. As he put it, the louvers by themselves were worth $500 on the market.

I was pretty jealous.

1977 Celica GT and Roreri.JPG
As I drove the Lorelei about, I became more and more sure that something was wrong with the fuel pump. I occasionally get 40psi, then it will drop. Not always. The way it goes is like this:

Idle at stoplight. 30psi.

Apply light throttle. 32psi.

Keep applying throttle gradually

Fuel pressure begins to bleed off, to about 20psi, full lean condition

--OR--

Fuel pressure increases up to 40psi depending on how wide open, properly rich

--THEN--

Fuel pressure precipitously drops to mid 20s then continues to drop to about 20psi.

--IF--

I let off throttle, fuel pressure returns to 30psi and stoich AFR.

Like I've concluded, either it's the fuel pump, or the fuel pressure regulator (maybe a worn part).

As far as the pump: Whether it's the fuel pump itself, or voltage drop, I can't say. In many cars, you can do a voltage drop test without dropping the tank. But. This is an ST185. I'm going to have to buck up and drop the tank.

Screw you Toyota for not putting a fuel pump access panel in the ST185.

As for the FPR, which way is the best way to go?
 

alltracman78

Active member
Um, you can access the FP connector from the trunk. You just can't remove the pump.

Unless my memory is screwing me here.

*LH side, there's a white plastic cover thing. It's under that.*

I'll try to confirm in the morning.
 

Roreri

Active member
Oh. That's right. I've seen that any number of times. If so, I'll do a voltage drop test via that access point.

Fuel Pump Connection Access.JPG
Another thought: Could it be the fuel pump relay?
 

underscore

Well-known member
Maybe I'm wrong, but I wouldn't think the FPR would be able to reduce the pressure below the base pressure the pump is putting out so I'd lean towards the pump?

There is a plug in that white plastic cover you can access to check voltage.

Removing the relay from the equation is easy, you can bypass it in the diagnostic box or the relay box near the alternator. I'd do it in the relay box so you can use spade terminals and a larger piece of wire.
 

Roreri

Active member
Jumping FP and B+:

I held out some hope that the problem might be solely due to the fuel pump relay. I jumped FP and B+ and the fuel pressure rose to 36psi when I turned the key to the ON position. This was a hopeful sign.

I did a check ride and performance was better. Base fuel pressure was 29-30psi at idle, 32psi at 0psi boost, and then the most I ever saw was 40psi. I still got lean conditions and low fuel pressure under boost, but it was better than under the fuel pump relay.

This is not in line with what I saw earlier this year, which was properly rich afr under boost, and sustained fuel pressure under load up to 45psi.

Direct Voltage Test While Driving:

I back probed the voltage at the pump on the blue wire with the positive lead, clipped the negative lead to the negative terminal of the battery, and did a check ride with the multimeter sitting in the center console. 11.59v with ignition key to ON, 12.18v to 12.45v while driving. I get 14.4v at the battery terminals while the motor is running, so I'm losing 2 volts at the pump? Or should the fuel pump be getting ~12v?

This thread: viewtopic.php?t=53742 indicates that: "...low voltage at the pump also affects fuel pressure. My car is 205 swapped so fuel pressure with the car running is 36psi. A fuel pump needs at least 13.5 volts to operate the car properly. with teh battery up front i was getting 13.8v at the pump so i was definitely getting my full 36 psi. But now with 12.5v at the pump a bit of math tells me im only getting about 32.6psi. So i imagine a loss of 4 psi is ok for normal driving but not sufficient for heavy acceleration."

This is essentially what I am seeing. It looks like I need to consider measures to get better voltage to the fuel pump.

As all I'm doing is jumping FP and B+ with a paperclip, maybe my loss is due to that thin piece of wire. Maybe what I need to do is get a pair of spade terminals and a nice fat bit of wire and do it at the relay as suggested. I don't know how to do that though--which sockets to jump I mean.

Maybe a new fuel pump relay might help?

Voltage Drop Test:

And now for the voltage drop test, done correctly...I was still fiddling around trying to figure out how to do it, so just did a direct voltage measurement.

From this website: https://www.delphiautoparts.com/usa/en- ... rop-test-0

Fuel Pump Voltage Drop Test.JPG
Not surprisingly, given what I had observed so far, this is what I got:

Fuel Pump Voltage Drop Test Results 29 August 2022.JPG
So, I'm looking at like 1.7 volts of drop, which could explain what is going on. According to the instructions, I should check for resistance in the power supply circuit.

Caveat: This is with FP and B+ jumped. I tried it with the fuel pump relay in and it didn't seem to want to give readings that made sense.

So...run a fatter wire for the power circuit?
 

underscore

Well-known member
I checked through my old photos and looking from the front of the car it's the top left and the bottom right terminals in the fuel pump relay socket that you jump together. As I recall the relay itself is quite pricey.
 

Roreri

Active member
Very good thanks! I’ll pick up a fat wire and a couple of spade terminals on the way home this afternoon and make a jumper to replace the FP to B+ bridge.

I replaced the thin paperclip with a pair of fatter paperclips, in case that thin wire was putting too much resistance into the system. Not likely but still…gotta experiment.
 

alltracman78

Active member
I keep posting this and it keeps getting ignored. Forget spade terminals and wires. Get the Toyota jumper. It's in the ST184s. Same spot as the relay in the ST185. Spade terminals are properly fit for the female terminals. You won't force the female terminals apart with something too wide. And you won't have arcing problems with something too thin. Nothing to cause a problem.

underscore":ui7jpibv said:
Maybe I'm wrong, but I wouldn't think the FPR would be able to reduce the pressure below the base pressure the pump is putting out so I'd lean towards the pump?

The fuel pump doesn't create any pressure at all. It just creates flow. Restriction(s) in the system (like the FPR) create the pressure. Base pressure is what the FPR produces with out any help from the diaphragm (boost pressure). If the spring is weak in the FPR it won't create or maintain enough pressure.
 

alltracman78

Active member
This thing. Comes in the ST184s in the same position as the yellow relay in the background. Factory part. You could probably order one new too. I don't have the part number.





Also, you aren't fully understanding the voltage test concept. That's why it doesn't make sense. I can't type it all out right now. Maybe I'll get to it tonight, don't know. Don't jump into anything until you fully understand it.
 

alltracman78

Active member
Let me start with apologizing for coming across as a dick. I was in a rush and didn't properly form my reply. It came off more abrupt than I intended.

I encourage you to get the factory jumper for the fuel pump relay. I tried to find a part number online but I couldn't. I think it's a US only part.
It's called a short pin and it replaces the relay in I think all US 5th gen Celicas (non ST185). The wiring manual didn't list it as being St184 only, just as FWD. It will properly jump the 2 correct pins in the relay block.

I also encourage you to make a jumper for the diag box out of an old O2 sensor. Cut off the wires with the connector, remove 2 of the pins (with the wire attached) from the connector and attach the wire ends together. Those pins fit the female pins in the diag box perfectly. No more forcing paper clips in there.
You can use it to check your CEL, fuel pump diag, ect.

You actually have a pretty good handle on the voltage drop, I'm just going to (try) to go over the circuit in detail for you.



The normal route for the fuel pump power is this.

When the key is turned on power goes to the EFI main relay in your main fuse box. Just a switch. Shouldn't have any resistance (On the switch side of the relay. The coil side will have resistance, but the coil side isn't part of the fuel pump circuit.)
From the EFI main relay it goes to the COR (Circuit Opening Relay). This attached to the outside of the ECU on the drivers side (for US spec cars, not sure on JDM). This is also just a switch. No resistance (same deal as the EFI relay though).
Power goes through the COR ONLY IF THE STARTER IS GETTING POWER OR THE ENGINE IS TURNING OVER. The relay will NOT close if the key is in acc. Relay gets power first from the small starter wire (the one that closes the contact in the starter, not the large one that actually powers the starter). Once the engine is rotating the ECU sends power to the relay.
From the COR it goes to the FP (Fuel Pump) relay. This is also just a switch (same deal as other 2 relays).
From the FP relay depending which position the relay is in it will either go through the FP resistor if the car isn't at WOT, or directly to the fuel pump if it is at WOT. The FP relay isn't on/off. It's one/or the other.
If it's going through the FP resistor this is an added load. It will use some of the available voltage, and will have some voltage drop across it. Obviously if the FP relay is bypassing the resistor it won't be part of the circuit, so no load and no (realistic) voltage drop.
After the FP relay it goes to the fuel pump and then to ground.

The only 2 places you should have resistance (and any realistic voltage drop) are the fuel pump and the pump resistor. Any thing else is a bad connection or bad wiring.
Fuel pump should be VERY ROUGHLY 5 ohms.
Pump resistor is spec'd to .73 ohms.

If you jump B+ and FP in the diag box you're bypassing the COR, FP relay, and FP resistor. Power goes from the EFI main relay directly to the pump (with the key on) and then to ground.
Pump is the only load and only voltage drop in this case.

The ground point for the pump is in the center of the rear panel. Travels through the chassis to the small ground wire at the battery (the large wire is basically just your starter ground).


Bad connections (loose, dirty) will add load and therefor voltage drop to the circuit. A corroded wire end can do this. Or a damaged wire. The relay internal "switches" can also cause this. Over time they get dirty and damaged from the on/off sparks.
Assuming it's good, I wouldn't think the wire is so small it would cause a 1.5 v drop. However, I've never measured, and it's definitely a possibility.

IF YOU MAKE THE POWER WIRE LARGER YOU NEED TO ALSO INCREASE THE GROUND SIDE OR YOU WILL FRY THE WIRE.

Even your ground side voltage drop is a tiny bit high IMO. Probably corroded wire and/or dirty connections at the battery. But that's not your real problem or your main focus.
Your power side is obviously too high. You're going to have to break the power side down in half and check the voltage drop again, and keep chopping it into smaller and smaller sections til you find where the voltage drop is coming from. It could be more than one spot. For example 2 (or 3) relays.


To stress test the FP regulator, first get a Mightly Vac pump (one that does positive pressure also, some only do vacuum).
Jump B+ and FP, turn the key on, attach the pump to the vac line on the regulator. This will allow you to give the regulator "boost" pressure and rise above base pressure. Raise the pressure and see if the FP stays where it should.
The pump should be able to handle this even if it's weak because the injectors aren't firing, so realistically no flow, only pressure. So you're mainly stress testing the regulator.

To stress test the pump disconnect the CSI (Cold Start Injector) hose from the fuel rail. You'll need an adapter to attach a long hose to the hole in the fuel rail. You can use either something that threads in, or use a banjo fitting on the existing banjo bolt. Run the hose to a large bucket, jump B+ and FP and see if the pump can keep pumping the fuel out. If you want to measure, mark lines on the side of the bucket and time the flow. You should have already checked the regulator, and assuming it passed it should have no trouble holding pressure. Now it's the pumps turn to work. Don't be surprised if pressure drops a few PSI, flowing out of an open hose is a lot more than squirting out of injectors. Just make sure flow stays consistent.
 

alltracman78

Active member
Also, I just want to reiterate that running the RC hood with an air to air isn't helping at all. It shouldn't lower your FP, but it's raising your intake temps considerably. And the ECU can't tell because it only has a temp sensor at the AFM. Nothing after the turbo.

You're honestly probably better off with just an intake hose and no IC at all vs the TMIC. You'd probably have less heat soak. That hood is holding all the engine and intake heat into the IC.



Just personal preference; I like an analog meter for stuff like this. When you measure an analog signal with a digital meter it can miss temporary changes the analog meter will pick up. Just gives a more accurate feel for the signal.


If I can get a chance I'll try to add some pictures but I don't know if I will.
 

underscore

Well-known member
alltracman78":39t1qjmc said:
I encourage you to get the factory jumper for the fuel pump relay. I tried to find a part number online but I couldn't. I think it's a US only part.
It's called a short pin and it replaces the relay in I think all US 5th gen Celicas (non ST185). The wiring manual didn't list it as being St184 only, just as FWD. It will properly jump the 2 correct pins in the relay block.

I tried to find the part number and failed too. It's not in all FWD 5th gens as my 92 GTS and 93 GTS didn't have it. I think I found mine in a junkyard pre-facelift GT.
 

alltracman78

Active member
Interesting.

I don't remember which year it came out of.

It's listed in the 90 and 93 wiring manual.

Good to know though. Maybe it's not as ubiquitous as I thought.
 

Gert

Active member
alltracman78":20l86rc9 said:
Also, I just want to reiterate that running the RC hood with an air to air isn't helping at all. It shouldn't lower your FP, but it's raising your intake temps considerably. And the ECU can't tell because it only has a temp sensor at the AFM. Nothing after the turbo.

You're honestly probably better off with just an intake hose and no IC at all vs the TMIC. You'd probably have less heat soak. That hood is holding all the engine and intake heat into the IC.
Oh my, did I miss that??? I must read and look at pictures better.
 

Roreri

Active member
I’ve known I wanted a FMIC for some time, or something better than the TMIC.

I ought to just get the stuff and do it. Is there a kit I can get that works well?
 

alltracman78

Active member
I don't know of any kits off the top of my head. I've always ran an RC or an ST205 IC.

I would suggest you figure this problem out first, just in case you cause something else while installing.
Much more of a pain if you're tracking 2 different problems and don't know it.

Side note, I misread your title when I mentioned you have a built in fall 91 car. Yours is a built in 92 car. Same effective thing. Just my bumble.
Either way, you know you can jump TE2 and E1 and put the ECU in a more sensitive diagnostic mode, right? Probably won't tell you what's up with your fuel, but handy other times. You're supposed to drive it that way so it can run all the checks.

Also, I THINK I found a factory jumper for you. There's a guy in Fl that parts out Celicas. He's got a 93 GT that looks like has the correct jumper in it. His username on here is Mike92LX. If you're interested.
 

Roreri

Active member
Understood. I'll probably end up going up to see Nick at NicSpeed in Denver and we'll work out a project for an FMIC.

Thank you for the reminder about the more sensitive diagnostic mode. I can give that a try sometime as well.

I tried to PM Mike92LX, but it came back no such user exists.

I took the Lorelei on a rather long ride yesterday from Colorado Springs Colorado to Laramie Wyoming--360 miles round trip up the 25 and across the 80 and back.

Because I have an afr and I know how to drive it, I kept it properly ratio-ed and was able to sustain 140kph (87mph) going uphill toward Laramie, and coming downhill back, afrs trended richer and higher speeds up to 160kph (99mph) were reachable for passing. On the steep climb from Denver to Colorado Springs, 120kph (75mph) was the rule--all of these are "fine" for driving, but of course any boost leaned the mixture out, and fuel pressures dropped. I had the feeling that this was a "pumping fuel uphill" issue or, more to the point, the motor wanted more flow pushing uphill than the pump wanted to deliver. In any case, it got me thinking that the fuel flow is the basis of the Lorelei's poor performance.

So, I'm out in the garage dropping the tank, having run most of the fuel out of it. I can look at the connections and see whether there's something apparent there, or replace the pump and see whether that corrects the issue. I've got the rally shield off and now I'm digging around for tips and tricks on dropping the tank and what to do with it once I've got it off.

EDIT: So far so good--all the steps are pretty straightforward in the BGB. There is no tank shield--the rally shield replaced the function of that, so I really shouldn't roll without that rally shield in place. The filler tube appears to be in good shape, but coated with a fine dusting of rust on the inside the tank section of tube. I suppose I can steel wool that and apply Ospho.

Dropping Fuel Tank Step 8 (Disconnect Fuel Lines).JPG
Allright, so here's my deal: (b) and (c) came off with the standard degree of hassle and reaching and twisting and pulling. And, happily, only a dribble of fuel. (a) is a different connector though. How do you like to disconnect that one? At the lower connection or higher up closer to the tank? Flare wrenches, I take it?

EDIT EDIT: CSAlltrac said this back in 2012--the search function is great on this board: "Pump will be pretty straight forward and self explanatory when you see it. As far as dropping the tank, get yourself some penne lube for the fuel hardline fittings and be sure to use line wrenches, 14mm and 19mm I believe."

So now to spray some PB Blaster on the connectors and let it sit for an hour or so!

EDIT EDIT EDIT: **** flare nuts. **** flare nuts right in the ***. :bangshead:

Got the wrenches on, but man that front nut sure didn't like the 14mm flare wrench I brought to the game. It didn't want to go even with PB Blaster having been working at it for a good while. I'm going to let it continue to soak for a while, because it looks seized up. Just to be certain sure, lefty loosey still applies on that 14mm nut, right? I should be trying to clench the two wrench handles together to get it loose?

14mm and 19mm Flare Wrenches on fuel hardline.JPG
 

Roreri

Active member
I got the tank off last night. That flare nut fought me every millimeter of the way. I chewed it up somewhat but the flare wrench will still go on it. I figure that if all goes well, it will not need to be taken off again for a decade or more.

D73CB7D0-53A9-4541-8A65-5FA05F98130C.jpeg
The filler pipe was in good shape, but the overfill flapper was hanging a bit loose—I don’t know whether that was by design or out of the norm. There was the standard surface rust on the section inside the tank that will clean off with steel wool.

The fuel pump assembly had similar surface rust—nothing egregious. I did note some rust at the electrical terminals on the fuel pump and considered it possible that these might be the cause of decreased voltage/voltage drop across the power circuit.

3C7369B4-8391-4818-9B4E-0DDBD294E596.jpeg
As the existing pump is 30 years old I am going to replace it with a new stock Denso unit. The existing pump was capable of providing sufficient flow—if the issue is not the fuel pressure regulator—as late as a couple of months ago, so a stock pump in good condition should solve the matter if it is in fact the fuel pump.

The fuel pump strainer was not disintegrated nor was it clogged.

I drained the three or so gallons remaining in the tank, and the fuel was not fouled. That said…I’ll put fresh fuel in the tank and run this drained stuff through my leaf blower and weed whacker, just out of general principle.

I mixed degreaser and hot water and gave the tank a rinse. The drainage had a scum of rust. After letting the tank dry (love Colorado for this) there was some rust in the tank which I banged out and poured out.

5029EC94-CE6E-4B8A-BBE6-E4F154D4CB19.jpeg
That’s just a little surface rust on the top of the tank. I’ll steel wool it, brush Ospho on it, and paint it flat black as a rust preventative.

The gaskets were in good shape all things considered. I couldn’t tell which one of the two on top were leaking fuel when the tank is very full—if I had to choose probably the bent pipe gasket. I gave them a solid rubdown with AT205 Reseal, which is a known restorer of old rubber and left them good and slathered in a Ziploc bag to steep.

If all goes well, I should have it all back together in a couple of hours.

Any thoughts? Should I get some gasket in a tube to reinforce/backstop the existing gaskets? Should I apply teflon tape to the fuel hardline flare nut threads and NOT cinch it down super tight and make a paint mark to be able to check if it’s vibrating loose?
 
Top