[tristessa]

So, it's been running better and better since my last post as I get things more dialed in, up until last Wednesday afternoon. Cut out on the highway, wouldn't restart, ultimately got towed home by AAA.

Verified that I'm getting spark. Verified that I'm getting fuel from the pump to the TBI unit -- flow at least, don't have a TBI pressure tester -- and back out the return. No fuel from the injector. The injector LEDs on the NO2C board blink while cranking, but no fuel from the injectors, the lack of fuel injected verified by stuffing a paper towel into the TBI beneath the injector and cranking the engine for 15-20 seconds and not having even the slightest bit of wetness on the towel. Checked wiring continuity from the ECU to the injector plug, it's fine and wiggling the wires doesn't break the connection. Swapped in the spare injector, retested and still the same thing. Borrowed a noid light set from O'reillys and the noid blinking was erratic for the first 0.5-1.0 seconds and then nothing.

Ohm checked my injectors, both read 1.2-1.3Ω which is correct for them, and on re-reading the installation guide wiki realized that Speeduino needs an external resistor for low-Z injectors. Ordered up two 8Ω 25W resistors, put one between the INJ1 output and the injector and still no joy.

So at this point I'm thinking I may have cooked the injector driver MOSFET. Beyond that, is there anything else in the circuit that might have gotten damaged that I should look into replacing? I see R1, LED1 and D1 and a connection to RN1 in the injector circuit of the NO2C schematic (https://github.com/turboedge/SpeedyBoar ... ematic.pdf) along with the MOSFET at Q1, but I'm not smart enough with electronics to read it any farther. I've got spare MOSFET's from the parts kit because I didn't bother installing the ones for VVT and boost because this engine doesn't have either.

[LPG2CV]

Post your tune and we can perhaps spot any corruption. Or you could try loading a base tune again.

When you tried the paper towel test, were you inadvertantaly opening the throtle and putting the engine into flood clear?

Flashing fuel channel LED's should equal fuel I think! Though they may have a different ground to the injector circuit. The mosfet is probably ok, so check the ground trace for the mosfet is not damaged. And check the pins in the harness connectors are not pushed back.

Are you able to vheck the FPR, as if it fails open then you will not have fuel at the injectors.

[tristessa]

I wasn't touching the pedal at all with the paper towel test, in fact the throttle linkage wasn't even connected. to the TB. I wanted to make *damn* sure nothing was keeping the butterfly from being closed. The FPR is built into the throttle body along with the injector so there's no easy way I can think of to test it, but they're $30 cheap so I'd be willing to throw another one into it. This one's only a year old, though.

I would have thought flashing LED's meant fuel as well -- schematic shows them (and the injector) grounding through the MOSFET. Connector pins at both the ECU and harness look good, connector pins at the TB look good, and like I said wiring harness continuity from the ECU to the injector tests good as well, no matter how much I wiggle the harness, and has no continuity to ground to suggest a short either. Maybe I cooked the injector itself? Also relatively cheap to replace if so, the Advance Auto up the road has a rebuilt one for $56 or I can get new Standard overnight from Amazon for $76.

[PSIG]

The FPR is built into the throttle body along with the injector so there's no easy way I can think of to test it, but they're $30 cheap so I'd be willing to throw another one into it. This one's only a year old, though.

Fuel pressure and flow is a requirement, and should be tested to confirm spec. There are multiple ways to test actual, or just indicated.

For example, you can install a pressure gauge anywhere after the pump and prior to the FPR and injector to measure pressure. Or, you can disconnect the return line and measure flow to spec.

I would have thought flashing LED's meant fuel as well -- schematic shows them (and the injector) grounding through the MOSFET. …Maybe I cooked the injector itself?

Flashing injector LED indicates the MOSFET is functioning enough to ground the powered LED. It does not ensure there is ground reaching the injector, or that the injector is good.

I would disconnect the injector and test injector ohms, where typically ±zero ohms or ±10% from spec is bad. The injector should audibly click when operated, and you can also feel it to confirm action (e.g., key-on injector prime). If fuel pressure is present, and sufficient voltage, it will squirt fuel.

If the injector has been unused for a period of time, they can often stick closed (fuel varnish) and require cleaning. Let us know what you find by testing.

[tristessa]

There is plenty of flow coming from the fuel pump -- it half-filled a 20oz Gatorade bottle within a few seconds when I disconnected the hose from the TBI and jumpered the pump on. Will get a tester at Harbor Freight later to confirm that pressure is OK later this afternoon, need to finish getting the roof of my house closed up because there's some rain in the forecast for tonight.

Injector tests at 1.3 ohms, which is in-spec. Which I also understand now is too low of impedence for Speeduino, so I've since installed an 8 ohm 25w injector in series with the injector line on the INJ1 line (single injector), but doing so didn't change anything. I cannot hear the injector click from the driver's seat, and can't easily crank the starter and feel the injector to know if it's clicking at the same time, will need to borrow a buddy or neighbor for that.

Car had been getting driven every few days for the last several week, and was doing 55-60ish at the end of an onramp when the breakdown happened so I doubt the injector is stuck closed from varnish.

[tristessa]

Solid 12.8v at the injector connector with the key on, and the wiring from the injector to the ECU -- the ground side -- is already verified good. So given that the wiring is good, if the LED blinks, the injector should be triggering, correct?

The injector needs 10-14psi according to GM specs for the cars it was used in. Fuel pressure maxed out at 8psi with the gauge teed in between the pump and TBI, and spiked the gauge well over 60PSI when I pinched the return line. Off to get a new regulator, at least it's cheaper and easier to replace than the pump.

[dazq]

There is plenty of flow coming from the fuel pump -- it half-filled a 20oz Gatorade bottle within a few seconds when I disconnected the hose from the TBI and jumpered the pump on. Will get a tester at Harbor Freight later to confirm that pressure is OK later this afternoon, need to finish getting the roof of my house closed up because there's some rain in the forecast for tonight.

Injector tests at 1.3 ohms, which is in-spec. Which I also understand now is too low of impedence for Speeduino, so I've since installed an 8 ohm 25w injector in series with the injector line on the INJ1 line (single injector), but doing so didn't change anything. I cannot hear the injector click from the driver's seat, and can't easily crank the starter and feel the injector to know if it's clicking at the same time, will need to borrow a buddy or neighbor for that.

Car had been getting driven every few days for the last several week, and was doing 55-60ish at the end of an onramp when the breakdown happened so I doubt the injector is stuck closed from varnish.

Just picking up on what you just said.

Have you been running the setup WITHOUT the additional injector resistor before you had the failure?
The issue with the injector resistance is not just about the speeduino not being ok with it being too low! Driving such a low resistance injector directly will likely cause it to FAIL due to over current(burnout the coil inside it) .
To drive these sort of injector coils directly without use of the resistor needs what is known as a peak and hold driver. This controls the current flowing through the injector preventing it overheating from excessive current flow to prevent damage.

To check the injector is being pulsed , unplug it and wire a bulb into the connector instead.
Then when the ECU pulses it you can clearly see it in operation right at the injector plug end of the loom.

[tristessa]

Have you been running the setup WITHOUT the additional injector resistor before you had the failure?

Yes.

I realize the mistake now, but what's done is done. Injector resistor is added now to rectify my mistake, and a replacement injector is easy enough to source and cheap enough that I'm just going to pick one up today to have on hand.

[PSIG]

I wouldn't stress too much about it, and it's either good or it's not, by testing. As in other posts, I have run GM TBI injectors (accidentally) with no resistors, and they ran fine for a long time before discovered and corrected with 5Ω. This is no guarantee that your injectors or board components are OK, but it wasn't an 'instant smoke' situation.

Speaking of that, two thoughts. First is that you can test (as in my previous post) in sections to prove or eliminate areas of wiring and connections. For example, you could use a heavy jump wire directly from the battery(+) to the injector resistor and test operation, in order to prove the power circuits to the injector are OK. Same for injector to ECM, and so on. Likewise power and ground directly with manual wire touching to prove the injector works and clicks at all. Simple tests for proof.

Second, the GM TBI injectors are designed to take full non-resisted power to open, then PWM to hold. I compromise by using a 5-ohm resistor, maximum, for each injector. This results in over 2A current at 14V, which seems to work well with low latency (dead time), even with a low battery.

Speeduino can handle up to 8 amps per-channel on official boards, by calculation, even though I have run them much harder. 8-ohms for 1.5A at 14V is rather high resistance (only 1A at 10V cranking), and it may work but I would test dead-time if it will run to see if latency is acceptable, as it may slow the injector operation excessively. As my own rule-of-thumb, I have not used a resistor more than 5-ohms with any injector on any ECM. That's me. Do your thing!

[tristessa]

Didn't see your post until this morning, I'm still racing to get my roof finished in the evenings after work and haven't been online.

Replaced the regulator and injector both, since I have to take the TBI off the manifold to do it. Now I've got 13PSI when I hotwire the fuel pump, and the car goes VROOM after cranking the starter so I think the wiring and ECU are probably fine. Doesn't want to idle yet, but I suspect I had low fuel pressure before and the tune & idle settings are off because the pressure is now correct.

Fun times....