Stumped - S52 OBD2 no start (resolved)

[arch4ngel]

I'm at my wits end trying to finish my swap.
The car cranks but doesn't sound like it's trying to fire.
I've tested everything I can think of besides a compression test.

The car is a 1991 325i with an obd2 s52 swap.
The motor was rebuilt by a reputable machine shop.
Bottom and top end were assembled by the shop.
I only installed the head on the block so as to leave little room for error.
Timing was triple checked. I had to take it apart after the first attempt was a tooth off and made sure it was spot on afterwards. Engine was rotated several times by hand.

All emissions equipment has been retained.
EWS flashed out by myself.
I wired my c101 using oem crimp pins and the chart here with a couple modifications.

Line routing verified from filter to rail.
System pressure tests good. ~45psi during prime and ~50psi when cranking.
No fuel smell in engine bay for whatever that's worth.

New fuel system components:
- Hoses and clamps
- Fuel pump
- Filter
- Fuel pressure regulator
- Rebuilt 21.5 lb injectors
- Relay

Ignition system:
- New plugs (NGKs, can't remember the part number)
- Low mileage coils and boots all tested at .8 ohms
- CPS is brand new, swapped out with a used one to rule it out.
- Cam sensor is new
- Main relay is new
- Battery is new
- Starter is new
- Plugs are dry and unburned, look like they've never fired.
- No idea how to test for spark, haven't found a good way to do so

Wiring:
- Battery+ voltage at battery and ignition pins 26 49 54 87 at ECU
- New engine ground strap from body to engine mount
- ECU ground at shock tower
- ECU itself in 318i bracket
- All relays terminal 87 are powered up
- Battery+ at injectors and coils.
- Injectors pulse, once a second or so during cranking
- ECU power was originally connected to the fusible link, moved to battery for testing
- I haven't been able to find a value for the updated m52tu cps but I traced the signal wire to the ecu and verified 12v and ground at the cps connector.
- Pulled MAF connector, no difference.
- Replaced ecu
- Check engine light comes on when key is in the run position
- No codes!





If anyone has any ideas I'm all ears.
This shouldn't be a difficult problem to solve, am I missing something obvious?
TIA

[328ijunkie]

You verify the changes you made to the ecu flash stuck? Pull a read and compare?

[arch4ngel]

No, I figured the elimination of the ews code verified that.
Good idea though, I can pull the 24k on both ecus and check tonight.

I also have a new cable on the way that will let me flash the 256k portion using the chipster.no software. This will let me flash a verified stock bin but still I feel like that might be a shot in the dark.

Assuming both ecus aren't fried and ews is in fact deleted the car should start and run correct? The only modifications I made were deleting ews and rear o2s on the first ecu and just ews on the second. The galletto software I used completely successfully both times.

[arch4ngel]

I pulled the bin from one of the ecus tonight and the ews byte was modified as expected.

Have any other ideas?

[IRON-E]

Did you have a 13 button obc?

If so, trace the harness to under the steering wheel and loop the 2 ends, unless it already has a loop which makes direct ignition.

[arch4ngel]

Sadly no, 6 button obc so no code function.

[IRON-E]

Sadly no, 6 button obc so no code function.

Well I tried sir, what about the harness, what year harness are you using?

[328ijunkie]

Did your ing wire on the X20 originally have 2 wires or 1?

[IRON-E]

Did your ing wire on the X20 originally have 2 wires or 1?

This ^^^. Pin 7 and 20 are together in some models.

[arch4ngel]

Well I tried sir, what about the harness, what year harness are you using?

All help is appreciated here :)


The harness is from a '97.
I did the wiring quite a while ago but I vaguely remember there being two ignition wires. I'm pretty sure I reconnected like stock them by wiring them to the same pin. I can check this tonight.

I see 12v at the ecu end of the ignition wire.
Any idea where the other end goes?

Thanks for the help guys!


Edit - Do you mean pin 7 and pin 21?
I'm sure I left the ews pin 7 disconnected as I wasn't sure if I would be keeping it at first.
Pin 21 I'm pretty sure had 2 leads terminating into a single pin at the x20.

[IRON-E]

Sorry, yeah you're right., one of those 2 wires is ground on the c101, so make sure you test those out.

[arch4ngel]

I traced the two-wire green pin back to pin 49 at the ecu.
Noticed a loose copper strand at the pin, decided for fun to cut the small gauge wire and run it to a fused terminal at my aux fuse box. No difference when I tried starting the car. I noticed the check engine light didn't come on when I turned the key to the on position with the green wire disconnected so it's definitely powering the ecu as I thought. Very little resistance in the wire.

I have someone bringing a spark tester by this weekend so I should have a chance to narrow down the problem a bit further. Also have a compression tester on the way.

[328ijunkie]

Does your harness use the Crank sensor adapter?

[arch4ngel]

Yes it does. The engine came with the m52tu sensor and adapter harness but I purchased a brand new harness and cps. Currently the old cps is in place (swapped for testing purposes) but I'll swap it back this weekend.

[arch4ngel]

Kgiynmgs brought over a spark tester over the weekend and we were able to confirm the car has spark. The reason the spark plugs are clean is there is no fuel making it to the combustion chamber. That means the cps is working, the dme is getting timing data and is signaling spark and fuel. However I wasn't smelling any fuel even though the injectors were supposedly pulsing.

I already verified fuel pressure so we we pulled the injector harness and tested the harness. The positive side was getting battery voltage as expected but the ground side was oddly showing 4.4v. Tested every injector and got the same reading. It turns out that several sensors and the fuel pump relay ground showed the same thing, 4.4v at the ground side of the connector. This is the case whether or not the dme relay is in place supplying power to the harness. To isolate the dme from the rest of the harness I cut the vanos ground wire close to the ecu and verified I was still seeing 4.4v.

So the dme is supplying 4.4v to several grounds it should not be supplying any voltage on. Perhaps this weirdness is preventing the injectors from firing? I tried both ecus I have and both exhibit the same behavior. I pulled one open to make sure there was no corrosion and it looks perfect.

I'm going to order another ecu tomorrow but is there some way my harness can be damaging the ecu internally? It just seems strange that two different computers have the same failure mode.