LocostUSA.com

I have installed all the wiring and ECU on my Locost, using as much of the 1995 Miata donor system as possible. The ECU is sending power through the white/red wire to all of the components (fuel pump relay, fuel injectors, EGR solenoid valve (vacuum), EGR solenoid valve (Vent), purge solenoid valve, idle air control valve, pressure regulator control solenoid valve, camshaft position sensor, and mass air flow sensor), but the return wires going to the ECU do not seem to be going to ground inside the ECU in order to complete the circuit and make the components work. I have made sure the five ground wires from the ECU go to ground (1K black/light green, 2A black, 2B black, 2C black/light green, and 2F black/red to black/light green) and the body of the ECU is properly grounded, and all of the other grounds for all of the cars components are properly grounded. I made sure all instrument grounds were separated from other component grounds. Everything else seems to be working because I have spark and the engine will kick over when I spray fuel into the intake. I have eliminated the Diagnostic Port from the ECU system (terminating two wires from the ECU: white/green, light green/yellow) and I have eliminated other unused components from the system terminating their wires from the ECU: automatic transmission control module (blue/white), air conditioning relay (blue/black), coolant fan relay (black/green), power steering pressure sensor (blue/yellow), refrigerant pressure switch (light green/black), heater control unit (blue/orange), and the rear window defroster relay (blue/red). What have I overlooked? Why isn’t the ECU completing the circuit for the return wires coming from these components? Thanks in advance for your help. Bandana's Build

I didn't use a Miata engine but ECUs tend to be similar.

It's very very strange that you're getting spark but the fuel pump isn't being enabled...

How do you know you're getting spark, by laying the plug and wire against the block? If you're getting spark, at least some of the grounds are working. Have you gone through every single wire on the ECU connector to determine what they are and where they need to go? Are there any wires that you aren't quite sure what they are?

First off, where did you run your multimeter ground wire to? It's very important to connect it to the reference point which you're checking, in this case, the ECU ground point. Second, while you're cranking the engine, do you see the bottom of the fuel pump relay being pulled to ground? Most of the time the pump will be off if the ECU isn't seeing crank pulses.

I suspect your crank and camshaft position sensors are working fine, else the ECU wouldn't be firing the plugs.

Also, there's no reason why the ECU would control power to the cam sensor, so I think that one should be off your list.

All the EGR stuff may not get enabled until the engine's actually running, don't worry about those for now.

When you say "terminating", what does that mean? It should mean disconnecting them from everything, cutting them short and putting a bit of shrink-sleeving over the end so they can't short against anything. In EE-speak, "terminating" typically means connecting a wire to a standardized load - that isn't required here.

How about checking the obvious. With ignition on, does the fuel pump turn on if you ground the corresponding ECU pin? Check that. Does the fuel pump have its own fuse? It should, and it may be missing or blown - find out.

It would be interesting to see what happens if you manually enabled the fuel pump... does it start and run?

Hopefully you're using a known-good ECU, and one from a manual transmission car. Many automatic tranny cars have signals coming from the tranny that the ECU needs to see, but hopefully that's not the case here.

_________________
Midlana book: Build this mid-engine Locost!, http://midlana.com/stuff/book/
Kimini book: Designing mid-engine cars using FWD drivetrains
Both available from https://www.lulu.com/

You say your grounding everything, but are you grounding everything to the correct place? Miata ECU and others are very picky. I think on a Miata everything is grounded to a bolt on the left side of the head. But your right the injectors have power and fire fired by the ECU grounding them. You can buy noids to make sure the injectors are firing. I had lots of problems on my Honda engine, first the injectors were not firing, but then the actual injectors were blocked, so I solved one problem but I could not tell. I picked up a noid kit for Harbor Freight, and that told the real story, once the injectors were cleaned we were good.

Thanks, guys, for your responses. I was very careful to make sure every wire from the ECU was identified and connected to the proper component. I doubt I made a mistake but you never know. The ECU was working fine in the donor when it was shut off two years ago and yes, it was a manual shift. Sorry about using the incorrect terminology, but unused wires from the ECU were “terminated” by taping the cut end and shrink wrapping them also. I checked the spark by laying the plug against the cam cover and, as I mentioned, when I sprayed fuel in the intake, the engine kicked over so, you are right, the cam sensor must be working if that is what controls the spark timing. I do not hear the fuel pump running when I crank the engine. When I disconnect the light green return wire from the fuel pump relay going back to the ECU and then run this wire from the fuel pump relay to ground, yes, the fuel pump runs when the key is on, so there seems to be no fuse issues (I checked them all and they seem to be fine, also). The problem is, with the fuel pump running, no fuel is coming back to the tank from the return line and the pump sounds like it is dead-heading, which kind of tells me the ECU is not letting the pressure regulator control solenoid valve work either (the yellow/green feedback wire is not being grounded by the ECU) and so the fuel flow to the injectors is being stopped. I have not checked to see if the injectors are firing. I thought I would make sure they are getting fuel first before I started messing with the injectors. As you suggest, I will start by segregating all of the five ECU ground wires and the ECU body ground wire and connect them directly to the battery ground wire terminal to see if this works. I will keep you informed. Thanks for your help. It is nice to know I am not alone in this quest! Bandana’s Build

Going for dead simple here, but do you have your fuel lines hooked up properly, not reversed?

_________________
Chris


Build log http://www.locostusa.com/forums/viewtopic.php?f=35&t=12374
Completed build showcase http://locostusa.com/forums/viewtopic.php?f=33&t=16865

Are you in a position to plug in a code reader into the diagnostics plug or has that been chopped off?

Bob

_________________
Build Log viewtopic.php?f=35&t=16640&p=187700#p187700

Mancave http://s1116.photobucket.com/user/mypic ... ow/mancave

First of all I would reinstall the diagnostic connector. (If it can provide relevant diagnostics OBD?) Check for codes.

Then I would go through the wiring diagram pin by pin and verify.

If the fuel pump is not priming you need to do some testing in that area.
Pump power & ground, manually activate the relay and check rail pressure (with a gauge).

_________________
Honey anyone?

Sounds like an immobilizer, but you shouldn't have an immobilizer on a 95 model.

_________________
mobilito ergo sum
I drive therefore I am

I can explain it to you,
but I can't understand it for you.

Well..... You can show that jumpering ground to the fuel pump turns it on and the vehicle can run so all the downstream circuits are proved good.
First check the grounds, if that doesn't do it check the plug to make sure none of the pins have sunk so they aren't making a connection.
Is pin 1C connected to start signal? As I recall, that's the fuel pump during start command for the ECU and the run program takes over once engine vacuum comes up.
Actually. ... If you start it with the wire connected to the ECU and a jumper to ground then remove the jumper, does it keep running?
If it dies, you know that the ECU is saying no (for whatever reason) but if it keeps running then it doesn't know when you are giving a start command so it won't provide the fuel pressure to start on.

_________________
"There are times when a broken tool is better than a sound one, or a twisted personality more useful than a whole one.
For instance, a whole beer bottle isn't half the weapon that half a beer bottle is ..." Randall Garrett

Thanks for all of the additional suggestions, guys. I disconnected and removed the diagnostic port from the system for simplicity, but it appears I am not good enough at troubleshooting to be without it, so I will re-install the diagnostic port and read the codes if I run out of options.

I will re-trace the fuel lines and make sure they are feeding and returning from the fuel rail properly.

I never did remove the connector from the front of the ECU, just cut and re-soldered wires, so I doubt I have any damaged pins on the ECU.

Yes, the violet wire from ECU pin 1C goes to the St(art) Sign(al) Fuse in the Main Fuse Block which is supposed to be hot with the engine cranking. I will check this to make sure it is getting juice when the engine is cranking.

When I grounded the fuel pump relay return feed wire (light green, ECU pin 2T) to see if the pump would run when the ignition was on, I cut this wire from returning to the ECU for fear the grounding may damage the ECU. But, as you suggest, I will ground this lead, leaving it also connected to the ECU, and see if the fuel pump continues running when the grounding lead is removed when the key is in the cranking mode.

Give me a day or two to perform your suggested tasks and I will then get back to you with the results. Thanks, again, for your help. Bandana’s Build

Read codes and write them down for reference. May have random codes from jiggery poking.

Erase codes (disconnect battery, turn on head lights, 10 min?)

Attempt start and recheck codes. These code are real (right now) failures.

Address lowest # code first, clear codes, then retest, and recheck for codes. To verify failure has been fixed and to see any remaining issues.

If fuel pump does not prime, either bad pump/circuit or PCM power/ground issue.

_________________
Honey anyone?

SUCCESS! The engine runs! I checked to ensure the violet wire from ECU pin 1C goes to the St(art) Sign(al) Fuse in the Main Fuse Block and that it has power when the engine is cranking. I changed termination of the five ECU ground wires to spots on the engine more closely resembling where they were terminated on the original Miata donor (1K black/light green, 2B black, 2C black/light green, and 2F black/red to black/light green to the front of the engine, and 2A black to the top right of the engine). I also tied 2F black/red to black/light green to the black/light green ground for the Mass Air Flow Sensor, and tied ECU grounds 2A black and 2B black together, similar to what was done in the original Miata donor. The fuel pump then started to run as the engine was cranking. And, as Chris suggested, I checked the routing of the fuel lines and, sure enough, I had them reversed. Still, it is nice to know, at least as far as I can tell currently, that all of my wiring of the vehicle and ECU seems to be correct and functioning. Guys, I would never have solved this problem without your help. Thanks again. Now the real fun begins – final assembly. Bandana’s Build

_________________
"There are times when a broken tool is better than a sound one, or a twisted personality more useful than a whole one.
For instance, a whole beer bottle isn't half the weapon that half a beer bottle is ..." Randall Garrett