Sale ends todayGet 30% off any course (excluding packages)
Ends in --- --- ---
Hi guys,
I swapped my NB miata with a Honda J35A1 from an Odyssey. I originally had it running with the stock Honda harness and ECU, but got sick of trying to troubleshoot limp mode. It did start without issue, and ran with the exact same internal setup. Now I have a Haltech 750 and a prebuilt harness, but after wiring it all up, it doesn't start.
.
Here's what I've already established:
-I had to grind the cam gear reluctors down to a single tooth as the native pattern was unsupported.
-TDC angle is correct (timing light read consistent pulses during the setting of it)
-I've supplied relayed and fused power to all necessary components, all sensors appear to be working
-My ground circuit is all accessories to - battery term, engine to -, - to chassis, adding extra grounds did not change the spark strength
-All wire harness plugs are in firmly
-Noid on injector says consistent injector pulse
-Battery voltage is good, car cranks strong
-Fuel pressure to the rail is 60psi
-Fresh spark plugs
-I have had Haltech remote connect to my computer and finish any smaller setup with the tune like 3 times now, no luck
-Cam and crank both are VR sensors, and are calibrated (seemingly) correctly for this application. Confirmed that they do not use ground reference.
.
And here's the symptoms of the issue:
-It will start to spin faster than cranking and cough and sputter, but not start. Sometimes it will run longer if I cut the fuel pump right after it catches, suggesting it only kinda combusts when EXTREMELY lean)
-It has Coil on Plugs and my COP inductive tester does NOT read a pulse even at max sensitivity. The timing light would pick it up when a spark wire was run inline though.
-Adding or reducing fuel VE has not made it run any better or worse. I found a middleground it consistently sputters but any more or less and it just cranks
-I checked that every wire out of the ECU for injectors and ignition was the correct color for the cylinder it went to.
-I pulled the plug and grounded it to chassis with jumpers, and observed that the spark only actually fires randomly maybe half the times it's supposed to. The spark looks and sounds weak too.
-In Haltech, when cranking it says "home tooth count error", but looking at the oscilloscope for CPS/CMP shows it always detects a consistent single tooth home for every 24 crank teeth. I'm not sure why the error comes up
-The waveform of cam and crank look good but I noticed RPM is dropping out randomly. It seems like it's dropping out less often than it's missing spark but may well be related!
Some of my speculations:
-I noticed that, when using a power probe, touching the 12V pin on the ignition wire with the key OFF, I see occasionally blips of 0.3-0.5v on the circuit. The wire from the relay supplying the power does not do it when the ECU is unplugged. I don't see much more than normal floating voltage with a multimeter (5-15 mV), maybe there's a short to ground somewhere and it's not getting sufficient power when cranking?
-The car was running extremely rich for the time it was on the stock honda ecu, so it fouled the plugs. I'm not sure if that maybe cooked the ignition coils? I tried swapping to spare oem coils and they looked a little stronger, but still intermittent sparking.
-Maybe I should replace the coils, but then they only have maybe 3000 miles on them.
-Maybe I am missing a ground or power in the right place? This harness didn't come with any instructions. Could also just be faulty.
.
I'm just out of ideas as to why it could be so inconsistent... Anyone have any? I'll a pic of the oscilloscope when cranking for reference, and my tune tmrw when I can get back to the car.
I seem to remember some of the Honda ignition coils charge when the ignition trigger is high and fire on the falling edge. Most ignition systems charge when low and fire on the rising edge. Here was some discussion about that: