Hello all,
My '95, 200K mile SC300 has been cutting out and I'm pretty sure it's starving for fuel sometimes. If I give it more than a little throttle, sometimes it cuts out badly. It'll run for a second, coast for a couple of seconds, catch again, and keep doing that until I back off the throttle and then it runs totally smooth. It would start for a couple of seconds in the garage and then quit. I put a fuel filter in and it seemed to help, but Saturday I drove 20 miles and back and it was doing the same thing at moderate throttle. Runs fine unless I start trying to accelerate very much.
I've been reading a very long, pretty old thread about this topic and it's pretty confusing. A lot of the people there are running engine mods and/or engine swaps and/or turbos but my car is bone stock. Just the standard 2JE-GE with EFI and automatic trans. I'm attaching the door plate for reference. Some people say you can just bypass the ECU and join the green wire to the black/red, and you'll have no problems and live happily ever after. Some say it will run rich at idle, some say you could smoke the wires, one guy said he hydrolocked his engine and wrecked it. Some said the pump was running with the ignition off.
So, a couple of questions. Someone said a '98 doesn't have a return for excess fuel flow, is that true and is it true for a '95? Someone said you have to add an external pressure regulator and return to tank, I sure don't want to mess with that.
And more generally, will the car run normally and happily with a constant 12V to the pump? Is there any way the pump can be powered with the key turned off, using only the wires that go to the ECU? Will the main ECU get confused or upset by the pump ECU being disconnected?
I have the wires cut and stripped and ready to connect, but I'm hesitant. I'm thinking of connecting it temporarily and trying it out, but I know there are a lot of knowledgeable owners and experienced mechanics here who know this car's ins and out, and I could use some advice before I pull that trigger.
Edit: is there anyone who can test this controller, and possibly fix it if it needs it?

 

Hello,

It may be wiser to first start from the beginning.

The Fuel Pump ECU only has 2 modes of operation: High Performance, and Low Performance - the former outputs 12V straight from the battery to the Fuel Pump, it is used for Startup and Higher Engine Loads, the latter reduces the Fuel Pump Voltage to about 9V, it is mostly used for the Idle and Low Engine Load, like cruising on the Highway, when there is really no need for the Pump to run at Full Speed. There are no other modes, nothing in-between.

The Only purpose for this function is to Prolong the Life of the Pump, maybe decrease the noise in the cabin along with it, it is not used as a way to prevent damage to the Wiring or Other Components, all of them are specifically designed to handle the Max Flow of that Pump with even some room to spare. Below is a page from the Repair Manual for some clarity, it is for a UZ engine, but the idea is the same.



The thing is, there is a way to Bypass the Fuel Pump ECU without so much as accessing it, let alone cutting the wires. To do that, you will need to locate Diagnostics Connector under the Hood, then short Pins FP and B+ (here is a Pinout for it), making the Fuel Pump run at the High Performance Setting regardless of the ECU Output.

Though fair warning - it may cause the Fuel Pump to run even without the Key in the Ignition (depends on the schematic of your particular model), this is solely a Test Run to see if Bypassing the Fuel Pump ECU would even fix the issue you are having, it is not in any way a permanent solution.

To short said pins in the Diagnostics Connector, while you can use a good old Paperclip, it's not a very reliable way, it can loose contact from time to time, voiding the test results. Instead, a better alternative would be to salvage Male Terminals off of old O2 Sensors, there is plenty of them at the Junkyard, and splice two of them together with either a Crimp or a Wago Lever Nut, here is a thread on how to De-Pin a Connector. Those Terminals are designed to work with Diagnostics Connector, which will provide a much more solid connection.

[EDIT]: Forgot to mention, here is a Parts Catalogue for your car, and it still does have a Return Line that goes from the Fuel Rail back to the Tank. You can use the Return line to determine if it is the Lack of Fuel that causes the car to Stall, to do so, you will need to get a hose that fits in place of the OEM Return Line, connect one end to the Regulator, and put the other end in the Clean, and preferably Clear container, you should see Fuel going into that container when you Engage the Fuel Pump, if you see if Sputter or Stop completely, it would indicate that the Flow the Pump creates is Insufficient, the causes for it can vary and would depend on your particular case.

Hope this helps and best of luck!

 

Thanks, that is very helpful! I did find a youtube about jumping at the diagnostic connector, he didn't say anything about it running the pump with the key off. Yikes. By that time I already had the ECU accessed and the wires clipped anyway.
To be 100% clear, bypassing at the FP ECU directly does not have that risk right? It doesn't make sense that it should, if that whole circuit isn't powered. I have actually done the bypass, jumping the green and black/red temporarily as a test, and I drove around a while and it seems to be solved. I accelerated flat out 4 or 5 times and it had no stutters or anything. I don't hear the pump when the key's turned off, and nothing has drained the battery at all overnight.
I've wondered if I, or someone, could duplicate the operation of that ECU. I'm a mechanical engineer and I have tinkered with some electronics and done some primitive programming, but this is over my head. Someone with the right skills could probably build something that would emulate that, but if the bypass works I guess there's not enough of a need.
I'm going to run like this for a few days and some more miles, and if I have no more problems I'll make it permanent and put the back seat back together. Thanks again for the help!

 

It seems like in your case the B+ Terminal in the Diagnostics Connector is powered via EFI Main Relay and not directly from the Battery, so yeah, the Fuel Pump will only trigger if the Key is in the ON Position.

That's one of those things I always hate doing, hence why I always try to find ways around issues like that without cutting anything, here is a thread on one of those occasions. In your case, a better alternative would've been to De-Pin the two wires you wanted to splice together and put them in a Jumper Connector like 90980-11398, which can be found in a 2nd-gen RAV4 used for the Power Outlets, a completely reversible mod, here is a thread on the Wiring Repair.

Well, suffice to say, by bypassing the Fuel Pump ECU you are doing the same exact thing as putting a Jumper in the Diagnostics Connector, it should be safe, but now every time you turn the Key to the ON Position, you turn the Pump on too, if you sit with the key On but Engine Off for a while, the Battery will definitely feel it. Basically you are back to 50's now, though it was the Ignition Coil back then that caused the same issue.

It's a very simple circuit, it doesn't even require a controller, something like LTC6992 Voltage Controlled PWM Modulator should be perfect for the application, the only issue is I have no idea what is the Output of the FPC Wire from the Engine ECU, could be directly the Voltage the ECU wants to run the Pump at, could be PWM from 0 to 5V, could be Ground. It seems like the reason those fail a lot is due to the fact that they are running in Linear Mode, meaning that all the excess power is being dissipated by the Transistor, causing it to turn into a smoking indicator lamp with higher loads, hence the idea to replace it with a proper PWM control, that will ensure that this ECU will last a lifetime of this car.

Hope this helps and best of luck!

 

Arsenii has nailed this!!

I can only add (or reiterate from his posts) that if the desire is to bypass the factory fuel pump ECU for pure 12-13.8V operation, the right way to do it is to replicate the fuel pump power circuit from the Supra MKIV TT by running a new 10ga or 8ga wire from a new battery positive connection from the engine bay through the factory firewall wire hole high beyond the pedals and to the left in the footwell) along the door sill and up to the area where the factory fuel ECU is located.

Then that battery +12V wire should get connected to a common 30/40A relay that shuts it on and off with the ignition.

The factory fuel ECU should be what is used to trigger that relay on and off since the factory fuel ECU is directly controlled by the main ECU. This eliminates any safety issue regarding ignition shutoff or accident safety. It will make the aftermarket relay turn on and off but will not influence it to operate at anything but +12V-13.8V.

We do this commonly for turbo builds but the same thing can be done with an NA stock car if desired or if the factory fuel ecu isn't regulating power right any longer.

When I did my initial setup I put a small LittelFuse fuse box in the engine bay for the direct power wire (I built it for 30A max load just as Toyota did for the MKIV TT but the stock SC300 fuel pump draws less current) and I put the 30/40A aftermarket relay right next to the factory fuel ECU with a little bracket.


LittleFuse fuse box install in the engine bay:
https://www.clublexus.com/forums/bui...l#post10976534

+12V SC fuel pump bypass w/relay (retaining stock fuel pump ecu as relay trigger and shutoff control):
https://www.clublexus.com/forums/bui...l#post10256118

+12V fuel ecu bypass wiring diagram theory and explanation:
https://www.clublexus.com/forums/bui...l#post10245454

And:
https://www.clublexus.com/forums/bui...l#post10249619

 

[EDIT]: Found another article, here is a link, it describes Bypassing the Fuel Pump ECU that works off of the FPC Output, meaning that the Pump will only turn On when it's supposed to, but it doesn't implement the Speed Control.

Well that does sound a bit excessive in this case, at least if Delphi is to be trusted, the Fuel Pump only consumes about 5A (here is a link), and even that I am pretty sure is a Peak Current, not the current it draws on average, so keeping the OEM wire and Re-Pinning it for a separate relay would probably be sufficient.

With that, now that I think about it, I wonder if a better option, the option that would solve all the problems with minimal effort would be to replace the Factory Fuel Pump ECU with the one off a 2nd-gen GS, keeping the entire system functioning exactly as it came from the factory, with proper shut-offs and retaining speed control. Those Newer ECUs seem to be a lot more durable, and if anything, at least there is a lot more of them still out there should your SC keep devouring them for breakfast.

As best I can tell, the FPC Control Signal is very similar between SC and GS, both appear to go from about 1V to 4V to set the Pump Speed, below is the Repair Manual for a 3UZ Engine, meaning that the newer Fuel Pump ECU shouldn't have any issues connecting to the Older Engine ECU, though it would be nice if anyone could access their working ECU to confirm that the FPC Output is indeed a 0-5V Analog Signal, as I have neither a Repair Manual for SC specifically, nor the car to try it on.

Scratch that, judging from the article linked above, FPC Output is a PWM Signal, which explains the .. particularly wonky Voltages to control the Pump, but doesn't explain why it's not even so much as hinted at in the Repair Manual. Again, I don't own an SC, you've been warned..

Still, it doesn't mean that SC is not using the same PWM Signal for its own Fuel Pump ECU, it just means that I shouldn't trust manuals anymore.. The only way to find out what signal it is would still be to access the FPC Pin with any kind of Oscilloscope, even the one linked below is plenty sufficient, and check exactly what kind of signal it is.




The other issue is the difference in Connectors, newer GS ECU uses much wider Connector Terminals, below is the picture, does seem justified in this case. That said, it's pretty easy to solve by purchasing just the Terminals from Mouser of all places, turns out they have an entire assortment of OEM Yazaki Terminals for a very good price, in this case the Part Numbers are 7116-4031 and 7116-4032, the difference being the Gauge of the Wire they are designed to work with, and you will need both. Here is a thread that goes into details on the Tools required, Places to get them, and the Process of crimping the New Terminals onto the Old wires, the only difference is that you will need a different Crimp Tool due to larger Gauge of the Wires than in my case, here is a link to it, search for ECT39. I usually hate cutting the factory harness as it's always obvious that someone has been there before, but in this case, if done properly, the retrofit would be virtually indistinguishable from the OEM.



As for the difference in the Mounting Plates, I would leave that one up to you guys to decide, it's the OP who's a Mechanical Engineer here after all..

If the SC ECU would be willing to work with GS Fuel Pump ECU, that would be a perfect solution, but without owning an SC, it would be pretty much impossible for me to verify it, all that is to say that I am not in any way responsible if you end up frying your Engine ECU as a result of the tomfoolery described above.

Hope this helps and best of luck!

 

Wow, that's a lot of information! Thanks everyone! I can't respond to everything, so just a couple of points and then 'the rest of the story.'
I likewise prefer not to cut wires, but after some consideration I decided to. It's completely reversible with a little solder and some of that heatshrink with the sealant built in, which I do have. And it was a very quick and easy way to temporarily bypass and see what happens. A calculated FAFO, if you will.
If anyone's not clear on my intent, I have no need or desire to try to increase performance. I love this car just as it is, and I like to keep things as simple as possible. Any modification is always another opportunity for something else to fail. So the simplest solution that just makes her run normally and reliably, and as close as reasonable to OEM conditions, is my first choice. I especially don't like getting very involved in electricity/electronics and maybe letting the smoke out of something. I don't have the equipment to do much testing, nor do I have a helper. So I do appreciate all the suggestions in that arena, but I'm not trying any of those things. Someone else may come across this thread later and find those useful, I hope so.
Another side note- Arsenii appears to be saying that this ECU dumps 'excess' energy from its transistor, which seems very inefficient but it does explain why it has that big heatsink. I was wondering about that, so thanks Arsenii.

Now the latest: as mentioned, after I crossed my fingers and crossed the wires, I test drove it Tuesday and everything seemed to be hunky dory. Then yesterday I went out on an errand and it started to do the same thing- any time I went past about 1/4 throttle it was starving out again. I threw up my hands and took it to my local dealership (I know, but I was in the car business for a while and I have relationships- strictly platonic ones.) I haven't mentioned here that my battery light has also been on and off for a few days, and it turns out that mattered. They found that the B+ terminal on the alternator was corroded, which I suspected but hadn't addressed yet. They fixed that and the fuel issue seems to be solved! And at no charge! So I drove home, pushing it pretty hard at times with no stalling. I'm not celebrating victory yet, but I'm optimistic.
I have a run from Houston to Katy this morning, so I'm going to keep my temporary bypass as-is and see what happens. If it keeps running well, and since the ECU was apparently not the problem, I'll try reconnecting it temporarily and give that a try for a few days. If it runs well that way I'll permanently reconnect that and put the interior back together. If not, I'll make the bypass permanent.
I'll report back in a few days, I know the suspense is killing you. Thanks again to everyone for all the great info. I like it here. I have another, non-urgent question to ask after this is all resolved.

 

fuel pump ecu is a very common failure; even for stock cars. That's why many of us have had to deal with it.

The relay suggestion from Kahn above is not a performance upgrade, but instead a safety feature to help protect yourself and your car in the case of an accident. You can find deutchwerks fuel pump relay kits which include the wiring with inline fuse from your battery which make it easy; I only had to pop the plastic door sill off to gain access under my carpet to feed the wire through.

Then I used the ignition trigger for the old fuel pump ecu, cut it and wired it into the relay; grounded the relay and fed the power wire to the fuel pump with the same soldering connection you made.

But overall you should be fine with what you've done for now; and may never have an safety issue with it.

Also glad to hear about the alternator B+. From reading your other post about the battery light flickering on I figured that was the issue, as it was mine.

 

May be my preference, but I try to avoid Soldering as much as possible, the wire seems to become a bit more brittle after that. Instead I found the Non-Insulated, Closed Barrel Splice Terminals that have the Outer Diameter the same as the Wire Insulation, making it a very simple and discrete repair. Panduit makes those, here you can find ones meant for different Gauges, and the Crimp Tool great for them is the same as for the Open Barrel Terminals described above, the thread that describes the entire process is also linked above.

Hope this helps and best of luck!

 

^^ Thank you. Exactly. Preserving safety is the main goal with the relay. Using the OEM fuel ECU simply as a trigger to turn the relay on and off is just an easy way to retain the stock main engine ECU's on/off control of the fuel pump in the event of an emergency or ignition shutoff to the key's "accessory power only" position.

It's a very simple circuit to implement and is a very common type of fuel power circuit both in aftermarket applications and in many (mostly 2000's and older vintage) OEM vehicle applications. It does not have to be accompanied by any performance upgrade.

If the stock fuel ECU still works or can be replaced with a working unit then by all means go for that option first. But if that ceases to become a reliable option then this would be the next appropriate step. The idea is that even if the fuel ECU is no longer being asked to regulate the pump directly, it should at least be working enough to act as a main ECU controlled trigger for the 30/40A aftermarket relay.

Arsenii -- Your post gives a lot to look into! It sounds like you already discovered a roadblock in differences as to how the GS and SC/Soarer fuel ECU's are controlled by the main ECU.

I don't know enough about the GS fuel ECU to comment with anything useful but it is interesting investigation.

It has been looked into before as to whether or not the stock SC/Soarer fuel ECU could be repaired with new capacitors, etc. I'm not sure if anyone ever concluded either way on that yet.

 

I agree soldering has potential problems, and is widely frowned on in automotive applications or anywhere vibration or movement is expected. The solder is rigid and the abrupt transition from the solder joint to the flexible wire is definitely at risk if there's movement. I had intended to support the joint so that it wouldn't undergo any movement, but it's an awkward location to be soldering in place. I have some good insulated crimp connectors and a proper crimping tool, so I'm going to use them.
As an update, I started to leave yesterday on an errand and started having the same problems again. I turned around before I got out of my neighborhood and parked it, and went in my Tacoma instead. When I turned it off though, I heard the pump running. The ignition was off and the key was out. This was the first time I've noticed hearing it BTW. I had the bypass just temporarily connected with a wire nut, so I just undid that connection and left on my errand.
Since bypassing the ECU didn't ultimately solve the problem (nor did the alternator connection as it turns out) it's likely not the source of the problem. I'm going to put it back in. Wish me luck, maybe the 5th time is the charm.

 

I would sure like to hear if someone does. I tried to take mine apart, but it's epoxied together. I would have had to saw it open or something, and I wasn't ready to go that far. I'm no electronics expert, but I could have at least looked for any evidence of damage. (Hope I did the quote thing right, I'm not expert at this either.)

 

Not sure if you saw it, I later edited my post to include a way of achieving the same result, while using all the Factory Wiring and Signals, with no need to pull any extra wires across the car. Here is that article in question, the idea is to put a Relay in place of the Fuel Pump ECU, then remove the Old Fuel Pump ECU completely, and use an FPC Signal directly to Power said Relay, he used a Transistor to reduce Load on the ECU, and a basic Filter to smooth out the PWM Output of the ECU, making it an On/Off Signal to trigger the Relay. As a result, the Fuel Pump is controlled by the Engine ECU with all the safety features, the only difference from OEM is that it doesn't vary the Flow of the Pump, the output is fixed to 12V, and all it takes is a couple of electronic components and a new Relay.

Can't say that it is a roadblock as I still have no idea whether SC uses PWM for the FPC Output. That said, seeing those wonky Output Voltages for the FPC Pin that they listed in the Manual, and that they look suspiciously similar to the GS Voltages, I think it's pretty safe to assume that both SC and GS utilize PWM for the FPC Output, and that SC Output Signal is indeed compatible with the GS Fuel Pump ECU. Though again, the only way I know to properly verify it would be for someone with an SC to get an Oscilloscope and actually check the FPC Output Signal, as all the scraps of Repair Manuals I can dig up totally omit this information for one reason or the other.

As written earlier, my assumption is that the Fuel Pump ECU in those cars utilizes a Transistor in Linear Mode, meaning that any excess Power that doesn't make it to the Pump is Dissipated by that Transistor by turning it into Heat, pretty much as if you were to try slowing down by stepping on the Brake without releasing the Gas Pedal, that's why it requires such a massive Radiator on the back. All that is to say that if anything fails inside those ECUs, it's very likely the Transistor itself, which, in all likelihood, will be glued shut into the Aluminum Housing to maximize the Thermal Transfer, giving it at least some fighting chance to survive this ordeal.

A video of the car cutting in and out would be helpful.

Are sure it was the Fuel Pump Running..?

With the Fuel Pump ECU Bypass that you implemented, the only way for the Pump to stay running when the Key completely out of the Ignition is if the EFI Main Relay stayed On for whatever reason. The thing here is that the EFI Main Relay is Triggered by the Engine ECU, at least if the Diagram from earlier is to be trusted, this Relay is the only piece that can force the Pump to stay Running with the Key Out, so unless you mistook some other sound for the Fuel Pump, I have some grim news to you, it seems like your Engine ECU is due for a check.

There could be a very small chance of the Relay itself going haywire, aka if it is Jammed a bit, causing it to Stick and keep the Pump Running, it can also cause Bad or Intermittent Contact, which would account for the other symptoms that you have with the car cutting out periodically. Try Swapping the EFI Main Relay with a different one, see if that does anything, that said though, Toyota Relays are extremely reliable, chances of something like that are vanishingly small, so I would hold my breath for a few days before saying that the issue is fixed with the new Relay.



Hope this helps and best of luck!

 

Wow…. A lot of information to absorb and then assess….

My fuel pump ECU was failing intermittently last summer so also went down the investigation “rabbit hole”. Seemed the simplest option was to completely bypass the fuel pump ECU with a jumper.

That had many supporters and several write ups not to mention a handful of You Tube videos being offered. Easily reversible since no wires are being cut.

Was not that comfortable with any of those. Researched getting a reportedly “new” part from Bell Lexus in AZ that had the MSRP as $700 or so but discounted down to $428. Not great. Instead, went on eBay and found a salvaged part for $103; not sure what year of a vehicle this came from but the part number was a 2nd generation version.

Installed the salvaged part and the problem went away. Had one glitch about a month later but that I believe was due to the battery starting to fail (it was over due anyway) and since the new battery no other issues have surfaced so far.

If the same symptoms arise anytime in the future the information above will be considered.

 

Hi all, I reconnected the fuel pump ECU as original and everything is back to normal. I drove around for a couple of hours Saturday to test it out and there wasn't a single hiccup. That ECU was never the problem unless it was an intermittent failure of some sort that's corrected itself. The fuel filter probably didn't have anything to do with it either- it ran well after the filter was installed (a whole other nightmare I hope my brain will block out of my memory soon) but the same problem returned the next day. All evidence points to the B+ alternator connection, which doesn't really make sense to me since the battery voltage was never low. What ever. It is what it is, main thing is it's running great now. through quite a few miles and two days. Thanks to everyone for the help.

My goal has been to be able to drive this car, instead of the Tacoma, across town to meet my daughter at a small blues festival yesterday. After much tribulation I succeeded! I got to show it off and give her a ride home from the festival in it, and she loved it. I'll give it at least a week to see if this recurs, then I'll trust it enough to go visit my sister 160 miles from here.