Car cranking, but not starting(Now with spark!)

[ST1G]

Maybe this engine has just gone caput.

Haha If that were the case I'd have spark ;)

[ThatOneEuroE30]

Haha If that were the case I'd have spark ;)

That is true. Time to tow it to a specialist

[spdracrm3]

Does your car have an OBC? another post with same issues http://www.r3vlimited.com/board/showthread.php?t=336715

So, I put in the new CPS, got it in there, got her all situated. No spark. put your meter on AC voltage and check across pin 47 &48 at ECU while cranking should have AC voltage being produced.see scope pattern below

Then I get to testing. I decide to put the multi meter on the coil positive (pin 15), and ground (body) and I'm getting 192Ω I get the same 192Ω when I put the meter between the negative terminal (pin 1) and ground (body) This was with the ignition on. This still doesn't make sense to me, since if the positive is grounded then the coil wouldn't get the signal to send spark. you cannot OHM test a live circuit (ie voltage present/flowing)
you can uses a test light to visually check between body ground and neg coil pin while cranking it should flash during cranking if ECU is controlling coil. understanding how a coil works will help diagnose ,for the coil to make spark current has to flow thru the coil primary windings for this to happen the voltage supplied thru the ignition switch has to go to ground this is done by the ECU the amount of time the current flows is called dwell ,with enough dwell time the coil primary is considered saturated and at the right time determined by the ECU using input from CKP sensor(correct spark timing)the ECU breaks the ground this causes the magnetic field around the iron core to collapse inducing a current in the coil secondary windings which have 100X more winding thus multiplying the voltage this voltage is sent out the coil wire and distributed by the cap to the right cylinders spark plug jumping the gap with about 10Kv thus ignition.if you put a test light on the coil neg you will see it flash as this voltage is present with no ECU grounding and drops away as ECU grounds coil thereby testing the entire circuit (voltage present ,yes ,light flashes yes ,ECU is doing something to control coil ) this cannot be tested with a DVOM as the sampling rate isnt fast enough (you need a scope instead)you might see the voltage drop down as the DVOM averages the slow samples but not a reliable test

After a few attempts to get the car to spark with nothing I checked the coil again and now I get a millisecond of continuity between pin 15 on the coil and ground and then it goes to infinite resistance (open). Same with the negative.

I then tried the coil from my running car. Same thing with the moment of continuity. No spark.


I don't even know any more.

[ThatOneEuroE30]

Car cranking, but not starting

Does your car have an OBC? another post with same issues http://www.r3vlimited.com/board/showthread.php?t=336715

He switched it out with a euro clock. I'm really thinking its this aswell.

For shits and gigs I'd switch the obc back in and see what happens.

[spdracrm3]

"He switched it out with a euro clock. I'm really thinking its this aswell.

For shits and gigs I'd switch the obc back in and see what happens."

yea BMW had the OBC alarm relay operate in several different ways thru the years/models ,on some it controls power to the coil (not his issue as he has power at coil) some its an input thru a three pin connector to the ECU (which wont start if its disconnected or properly bypassed)

i looked at a 90 ETM that show this year 325 relay output going to pin 27 of ECU which is listed as "start input" and he has checked that also and has power so , but might be worth a try reinstalling OBC like you said.

needs to check CKP sensor input at ECU also

[ST1G]

Does your car have an OBC? another post with same issues http://www.r3vlimited.com/board/showthread.php?t=336715

He switched it out with a euro clock. I'm really thinking its this as well.

For shits and gigs I'd switch the obc back in and see what happens.

I did install a euro clock, but I had a 6 button before. I spliced it in, and closed off the unused wire. I don't think I have the 6 button any more. I could find one though.

put your meter on AC voltage and check across pin 47 &48 at ECU while cranking should have AC voltage being produced.see scope pattern below

I will do this and report back.

you cannot OHM test a live circuit (ie voltage present/flowing)

Durka durka, I'm an idiot. I knew this but must have forgotten at the time.

you can uses a test light to visually check between body ground and neg coil pin while cranking it should flash during cranking if ECU is controlling coil. understanding how a coil works will help diagnose ,for the coil to make spark current has to flow thru the coil primary windings for this to happen the voltage supplied thru the ignition switch has to go to ground this is done by the ECU the amount of time the current flows is called dwell ,with enough dwell time the coil primary is considered saturated and at the right time determined by the ECU using input from CKP sensor(correct spark timing)the ECU breaks the ground this causes the magnetic field around the iron core to collapse inducing a current in the coil secondary windings which have 100X more winding thus multiplying the voltage this voltage is sent out the coil wire and distributed by the cap to the right cylinders spark plug jumping the gap with about 10Kv thus ignition.if you put a test light on the coil neg you will see it flash as this voltage is present with no ECU grounding and drops away as ECU grounds coil thereby testing the entire circuit (voltage present ,yes ,light flashes yes ,ECU is doing something to control coil ) this cannot be tested with a DVOM as the sampling rate isnt fast enough (you need a scope instead)you might see the voltage drop down as the DVOM averages the slow samples but not a reliable test

I have a light, so that's no problem. That is really good to know about coils, it's helpful to understand what to test and what the potential problem could be. I appreciate you spending the time to explain it to me. I also appreciate you actually reading the thread and providing me with helpful advice.

I will test this tonight and see what the light is doing; report back.

yea BMW had the OBC alarm relay operate in several different ways thru the years/models ,on some it controls power to the coil (not his issue as he has power at coil) some its an input thru a three pin connector to the ECU (which wont start if its disconnected or properly bypassed)

i looked at a 90 ETM that show this year 325 relay output going to pin 27 of ECU which is listed as "start input" and he has checked that also and has power so , but might be worth a try reinstalling OBC like you said.

I don't think it's the clock because the car ran fine for more than two months after the clock install. Also it wasn't the 13 button so I don't think I had the lock.

I am getting power to main relay, and fuel pump relay, as well as the ECU.

needs to check CKP sensor input at ECU also

What do you mean by this?

[Farbin Kaiber]

You know the CPS works at the connector to harness, but he is asking you to verify the signal at the DME connector.

[ST1G]

You know the CPS works at the connector to harness, but he is asking you to verify the signal at the DME connector.

By doing this correct?

put your meter on AC voltage and check across pin 47 &48 at ECU while cranking should have AC voltage being produced.see scope pattern below

[spdracrm3]

yes (CKP sensor same as CPS) ,by checking for AC signal at pin 47 &48 of the ECU ,or using a scope to check pattern

pretty much eliminates the OBC /clock being an issue unless a related wire is causing an issue . just for info how many ring end terminals are on each coil stud and what color wires are they ? want to eliminate some aftermarket device screwing up our ECU control signal



Farbin
how does he know the CPS is working at the harness connector ? all ive seen are OHM's tests done at sensor and ECU , havent seen any post about using a scope or checking for AC voltage output from sensor ? want to make sure i didnt miss something....

[solo318]

The problem with checking resistance is that it is a static test. If something is causing noise on the signal or the signal is too week then it still won't start. Check a/c voltage at the DME while cranking the engine. Also check DC voltage at the DME power wires while cranking. Something as simple as a bad Ign switch could be at fault. Post up your findings and we'll go from there.

[Farbin Kaiber]

Farbin
how does he know the CPS is working at the harness connector ? all ive seen are OHM's tests done at sensor and ECU , havent seen any post about using a scope or checking for AC voltage output from sensor ? want to make sure i didnt miss something....

You are right. My bad.

[ST1G]

Hmm. Went to go trouble shoot tonight and the battery is dead. I fully charged it a week ago and the car has been sitting with the ECU and coil unplugged. I wonder what is leaching power.

[ST1G]

pretty much eliminates the OBC /clock being an issue unless a related wire is causing an issue . just for info how many ring end terminals are on each coil stud and what color wires are they ? want to eliminate some aftermarket device screwing up our ECU control signal

The coils are both correct bosch coils.

As a note, I did replace my cluster with a yard one, however I started the car while the cluster was out to move it, so I don't think this is the issue. I've also replaced my dash in the last 6 months.

yes (CKP sensor same as CPS) ,by checking for AC signal at pin 47 &48 of the ECU ,or using a scope to check pattern

Testing completed. Not sure what I should have, but here is what I do have.

AC voltage at pins 47 & 48 while cranking is 2.78V. This is with the meter set to '20' AC voltage.

Resistance at pins 47 & 48 with key in 'off' position is 528Ω (I get 530Ω at the pins on the cps when unplugged from the harness).

Test light acting like it's getting less power with cranking, but it seems like a drop as a whole not a flickering.



I didn't test the new CPS with a ferrous object prior to install. I'm thinking I might pull it out and test it with a bit of iron.

EDIT: Did some additional testing.

Old CPS plugged into harness with or out ferrous object (cps touching a iron plate weight) 508Ω and 0v at pins 46 & 47 with power off and on respectively. So, no change with introduction of ferrous object. I think this means the old cps was bad.

[ThatOneEuroE30]

I'm so f'ing lost haha. Why won't this start

[spdracrm3]

The coils are both correct bosch coils.

As a note, I did replace my cluster with a yard one, however I started the car while the cluster was out to move it, so I don't think this is the issue. I've also replaced my dash in the last 6 months.



Testing completed. Not sure what I should have, but here is what I do have.

AC voltage at pins 47 & 48 while cranking is 2.78V. This is with the meter set to '20' AC voltage. well this indicates the ECU is getting a signal from the crank sensor so we know the circuit is good and we have a signal.

Resistance at pins 47 & 48 with key in 'off' position is 528Ω (I get 530Ω at the pins on the cps when unplugged from the harness). its working as proven above

Test light acting like it's getting less power with cranking, but it seems like a drop as a whole not a flickering. it will dim slightly due to voltage dropping slightly due to load , so if your testing on the right (- NEG ) coil terminal the video below shows no ECU control therefore no spark . check on both terminals just because...



I didn't test the new CPS with a ferrous object prior to install. I'm thinking I might pull it out and test it with a bit of iron.

EDIT: Did some additional testing.

Old CPS plugged into harness with or out ferrous object (cps touching a iron plate weight) 508Ω and 0v at pins 46 & 47 with power off and on respectively. So, no change with introduction of ferrous object. I think this means the old cps was bad. the resistance wont change with the CPS near a ferrous object the windings will still measure the same (thats not how the sensor works) . if you move the ferrous object quickly past the end of the sensor it in induce a current measured as AC voltage at the correct pins on connector or at ECU as you did above . move a bunch of ferrous objects rapidly past the sensor ie a tone ring (such as whats on your crank pulley near to the sensor )you will get a pattern as can be seen on a scope if the pattern is correct the ECU will use the data and start operating

so now why is the ECU not controlling the coil ,i think its time to check every pin at the ECU for correct voltage/ground /signal as something is keeping the ECU's from working ,do you have a manual that has an ECU pinout ? (ie what should be present at every pin ?)

also worth doing a very good inspection of the tone ring for any damage/excessive wobble anything that could potentially cause a bad pattern