Hi Gents,

I'm most of the way done wiring up an M Unit in my 1994 K75S. I used Zach Ross's wiring layout (found here https://www.k100-forum.com/t11686-m-unit-wiring-bmw-k)

I'm not getting any spark at the plugs.


Please ignore the messy wiring. I want to get this running and then go back and tidy things up a bit and have made a real mess of things chasing this gremlin...

The starter engages and engine turns over
Fuel pump turns on
Injectors spray (I haven't seen it but there is fuel on the plugs when I pull them out)
There's power at the coils
The plugs are new Bosch X5DCs with new spark plug wires from Euro Moto Electrics. I tried swapping out the wires and going back to the factory wires with no luck.

I've been through Bert's troubleshooting diagram and run up and down the wiring, testing grounds, etc.

Found two interesting things:
1) Pin #3 on the female connection to the ICU - when testing for continuity with ground at the battery measures pretty high. I'm getting 1kOhm
2) When I test continuity on Pin #3 of the female ICU connector while pressing the start button I get some weird results on my multimeter. It cycles between a reading and Open Line.  Video uploaded to Youtube below. 

Is this expected? Bad Hall sensor? Do I need to go back to the troubleshooting drawing board?

Your gremlin sits in a little black box with the letters M, U, N, I & T written on.

I'll never understand why so many people are spending so much money in a useless crap that will ruin the OEM reliable electrical equipment.
Fashionistas' sheepism?

jbt wrote:I'll never understand why so many people are spending so much money in a useless crap that will ruin the OEM reliable electrical equipment.

Because Gucci doesn't make accessories for Bricks yet, unless it owns a piece of Motogadget already.

lethaljd wrote:1)Pin #3 on the female connection to the ICU - when testing for continuity with ground at the battery measures pretty high. I'm getting 1kOhm
2)When I test continuity on Pin #3 of the female ICU connector while pressing the start button I get some weird results on my multimeter.

1) Measures pretty high? That's ≈1kΩ too high. Something is wrong there.
2) You already tested for continuity in Step 1) and Pin #3 failed the test. Testing Hall sensors while running the starter is not part of the testing protocol in the Troubleshooting the EFI guide here nor is it a protocol in BMW's LE Jetronic testing manual, reasonable as it might seem. Consider reviewing the testing procedures.

Thank you for the helpful reply, Laitch! I’ll review again and get to the bottom of this super high reading.

For what it’s worth @jbt I bought this project bike from someone that had already stripped the bike and taken it too far to recover. An M-Unit was included in the sale. I don’t need any fancy gadgetry but it was there and seemed silly to buy all of the pieces needed to bring it back to stock. The M Unit is not the problem in this case. It is doing its job of supplying power where it needs to go.

I sorted out the ground impedance issue and I believe I have the problem isolated.

Pin #3 on the ICU female lead shows open line - In Crazy Frog's troubleshooting diagram this is supposed to be 0 Ohms, Ground from Hall Sensor to battery.

I've tested the connection between the ICU female connector and the male plug to the Hall Sensor and everything is good there. No break in the wire, 0 ohms. 

I tested the Hall Sensor female plug and it's open line. I swapped for another Hall Sensor/wire and same result.

My question - how does the Hall Sensor ground wire complete the circuit to return current? Is it just the metal plate it's on making contact with the engine block that completes the ground circuit?

Previous owner painted the engine block and got some paint on the area where the hall sensor nestles in. Wondering if I need to remove more of the paint here to ensure good contact...unless I'm missing something about how it's grounded?

Thank you!

For some reason BMW threw their toys out of the pram when it came to the wire colour coding for the HES. On the Haynes wiring diagram (which, AFAICS, is the original BMW wiring diagram) it shows

- red is +12 volt
- black is ground
- brown is signal 1
- orange is signal 2

THIS IS WRONG. Clymer have redrawn the original diagrams in colour and do not mark positive or negative feeds, but it still implies the above sequence. When you get to the HES plate itself and compare the wire colours to the HES sensor specs, the colours change:

orange becomes black (ground)
red becomes green (signal 1)
brown becomes green (signal 2)
black becomes red (+12 volts)

See https://www.k100-forum.com/t18118-crazy-hes-wiring for a pic. It's off a K75 but that makes no difference apart from the position of the sensors.

The braided cover in the cable run is earthed at the top (ignition control unit) and floats free down in the HES housing. It's there to reduce electromagnetic interference (EMI). The HES plate itself is not earthed, except as a consequence of being in the engine block. I've had them dangling from the end of the lead, using a feeler gauge as a trigger when testing a recalcitrant ignition system.

lethaljd wrote:My question - how does the Hall Sensor ground wire complete the circuit to return current?

What is governing operating current are magnetic fields created by sensor #1 and #2 on the HES plate responding to the position of the windows of the rotor during crankshaft rotation, not ground connections. The current fluctuates from 5V monitoring current to 11V operating current.

Thanks @dai and @Laitch

I’m struggling to understand the ICU testing procedure outlined in Crazy Frog’s write up. Pin 3 of the ICU is supposed to test at 0 ohms when testing connection with ground at the battery.

When I pull back the rubber boot on the female ICU plug at pin #3 I see a black wire leading to the hall sensor male connector. I’m assuming I complete this test with the hall sensor plugged in and by placing one end of the multimeter on pin 3 contact of the female icu connector, other probe on the battery negative terminal. When I do that I have open line. This is on 2 different hall sensors and I’ve checked the wire from the female icu connecter to the the male hall connector.

Am I testing this incorrectly?

Is the pin #3 socket you're testing located in the same place as the one in the plug illustrated below?

@Laitch - Yes, that’s the one. I attached a pic for confirmation and the corresponding hall sensor female plug it connects to

Let's try this. If there's a kill switch in your control array, verify it isn't killing current. These instructions describe the OEM kill switch.
Ignition switch On, kill switch lever in vertical position, multimeter meter set to Volts DC
Black lead on pin #1
Red lead on pin #4
Volts =

Ignition switch On, kill switch lever in vertical position, multimeter meter set to Volts DC
Black lead on pin #1
Red lead on pin #3
Volts =

Ignition switch On, kill switch lever in vertical position, multimeter meter set to Volts DC
Black lead on pin #1
Red lead on pin #2
Volts =

Laitch wrote:Let's try this. If there's a kill switch in your control array, verify it isn't killing current. These instructions describe the OEM kill switch.
Ignition switch On, kill switch lever in vertical position, multimeter meter set to Volts DC
Black lead on pin #1
Red lead on pin #4
Volts = 0

Ignition switch On, kill switch lever in vertical position, multimeter meter set to Volts DC
Black lead on pin #1
Red lead on pin #3
Volts = 0

Ignition switch On, kill switch lever in vertical position, multimeter meter set to Volts DC
Black lead on pin #1
Red lead on pin #2
Volts = 0

When I press the starter, the bike turns over, fuel pump pumps, injectors spray, etc. but no spark.

Stoopid question: do the sparkplugs have bobbins/nuts/screwtops on them? If they don't, it's quite possible that they are not making correct contact with the plug caps and won't spark.

Pin #4 inconclusive.
Pin #3 inconclusive.
Pin #2 indicates no power to the Hall sensors—no surprise there.

Ignition on, kill switch lever vertical, multimeter set to VDC
Black lead on pin #1
Red lead on pin #10
Volts = battery voltage
If 0 Volts, jump pin 1 to bare metal frame ground.
Volts =

Laitch wrote:Pin #4 inconclusive.
Pin #3 inconclusive.
Pin #2 indicates no power to the Hall sensors—no surprise there.

Ignition on, kill switch lever vertical, multimeter set to VDC
Black lead on pin #1
Red lead on pin #10
Volts = battery voltage
If 0 Volts, jump pin 1 to bare metal frame ground.
Volts =

Got battery voltage with this test 13.1V. No ground jumper needed.

Thank you for helping me troubleshoot this, Laitch. I really appreciate it.

Unit power input at pin #10 and unit ground at pin #1 verified. The Hall sensor plug should have been connected during tests of pins #4, #3 and #2. If it was connected, disconnect it now, repeat test of pin #2 only and determine voltage.

I had the Hall sensor disconnected during the initial test. Re-ran the test with Hall Sensor plug connected and all tests of pin 2,3,4 still register 0V DC.

Disconnected Hall Sensor and tested DC Voltage between pins 1 and 2 of the female ICU plug. 0V DC.

Voltage is being input into the ICU via pin #10. No voltage is being output from unit pin #2 regardless of HES plug disconnection. Maybe there's a short to ground in the #2 wire running to the HES plug. Check it out. I'm leaning toward the root cause being a faulty ICU. This could be the too in your description of the PO's Brick that "someone… had taken it too far to recover. Find a working ICU to swap for this one to verify, but don't just run out and buy one until other members here with deeper experience have chimed in.

Have you ever witnessed this engine running?

Thanks, I’ll check for a short. I did buy another ICU and  have the same results (no spark). Dumb question time from me - how does this point to a faulty ICU if we’re testing with the ICU  female plug disconnected? Are we saying it’s getting the desired inputs, I.e. current at pin 10 and ground at pin 1, therefore the ICU must not be doing its job once connected?

And no, I never saw it running. Bought it from a very open, honest and trustworthy fellow that had it running, got into the modification process and then I think his friend had an accident that scared him and he decided to be done with it completely. It was running fine at the time of the rebuild starting and has only 9k miles on it.

lethaljd wrote:Are we saying it’s getting the desired inputs, I.e. current at pin 10 and ground at pin 1, therefore the ICU must not be doing its job once connected?

I'm just capturing a few inferences that have been eluding me during your work with the multimeter. What you might consider doing is the Hall sensor test in the troubleshooting guide that uses a 12V LED (LED with built-in resistor) to determine if the sensors are triggered during slow movement of the rotor windows across the Hall sensors with the ICU plugged in. Have you viewed that test procedure? That could be the next step.

Thanks. I just tested the Hall sensor with an LED as described in the guide and it works as expected.

I was also reading some other forums and saw someone had a similar problem of no spark and it was a bad/weak battery. I didn’t think that was the case here, but just to rule it out I hooked up my car battery and still the same result - no spark.

Did some more testing today and had some interesting findings.

Tested the K75 coils - all of them read 0.5-0.7 ohms between terminals 1 and 15 on the coils. Seems to me they are good

Then I tested the spark plug wires. I have 2 sets - the OEM set from BMW and a new Enduralast set from EuroMotoElectrics. The OEM set is dead. Open line/no measurement when testing with multimeter. Those haven't been on the bike aside from one set of testing and I've been doing most of my testing with the EME set.

EME set measured 6.5kOhm or so which seems to be reasonable.

Now this is where it got interesting. I have a DC bench tester and decided to try to do a coil/wire/plug test. Not sure if I had the test set up correctly, but I had the DC tester set to 12 V DC constant current. Hooked the probes up to coil inputs 1 and 15. Hooked a ground wire up to the spark plug threads and let her rip. I was expecting to get a spark, but got nothing.

The DC bench tester showed the coil was pulling 10A and about 100w, but I wasn't getting any spark. Does that seem right? Was this a dumb test? Did I kill the coil? Post-test, the coil still measured 0.7 ohms between pin 1 and 15.

Dai wrote:Stoopid question: do the sparkplugs have bobbins/nuts/screwtops on them? If they don't, it's quite possible that they are not making correct contact with the plug caps and won't spark.

Dai I have some of those Bosch spark plugs, they dont need the bobbins, the tops are made to the bobbin shape.

Yep, it has the bobbin integrated. 

I also tested for continuity using the multimeter with the spark plug inserted. One end of the probe on the spark plug wire where it would plug into the coil, the other end on the tip of the spark plug. I had continuity and approx the same ohm reading as the just the wire itself (6kOhm-ish) so I know it had good contact.

Your HT leads look like RAM leads - Enduralast probably bought them from either RAM or BSK Speedworks.

You wouldn't get a spark with that test as the spark is generated when the EMF field collapses in the primary winding in the coil as the HES is triggered. That generates a very large EMF in the secondary coil which fires the spark. From this I think you can see that a static test isn't going to work.

I don't trust spark plugs. They have a 5% failure rate off the line and those are old plugs. Do you have another bike to temporarily attach the plug to the HT lead and check for a spark? If you do, check all four plugs. Don't screw the plugs in because if the nose is too long for the engine you stand a good (bad?) chance of holing the piston.

When the bearings went on an Evinrude outboard of mine and the pistons were traveling too far I don't know what it did to the pistons but it did close the gap on the plugs.

Dai wrote:Your HT leads look like RAM leads - Enduralast probably bought them from either RAM or BSK Speedworks.

You wouldn't get a spark with that test as the spark is generated when the EMF field collapses in the primary winding in the coil as the HES is triggered. That generates a very large EMF in the secondary coil which fires the spark. From this I think you can see that a static test isn't going to work.

I don't trust spark plugs. They have a 5% failure rate off the line and those are old plugs. Do you have another bike to temporarily attach the plug to the HT lead and check for a spark? If you do, check all four plugs. Don't screw the plugs in because if the nose is too long for the engine you stand a good (bad?) chance of holing the piston.

I don’t have another bike, unfortunately.

Decided to do a few things:
1) ordered 3 new Bosch plugs (the ones I’ve been testing are brand new but whatever)
2) ordered a spark plug tester on Amazon
3) ordered a set of working k75 coils and wires on eBay

At this point I’ll have swapped everything but the:
MAF (tested to spec with Crazy Frog’s guide)
Coolant temp sensor (tested to spec with crazy frogs guide)
Jetronic - I’m getting spray from the injectors and it shouldn’t cause no spark, right?

Will report back next week when all of the goods arrive. Worst case I sell the extras or keep as spares.

I think you're saying it was running well until the m unit was installed, so I spose the hall sensors, coils etc. are ok. Maybe it's worth checking the install of the m unit.

Well, after 20 hours or so of troubleshooting I finally found the culprit. It was a loose terminal connection supplying power to the coils.

I had tested it multiple times and it read battery voltage of 12+V, but when I was measuring it I was bending the wire and connection in a way that solidified the connection. When it was laying in place as it would be when running, the connection was loose and I was getting voltage loss with variable values at the coil plug end ranging from 3-10V. 

To test it out I direct wired a coil to battery positive and it sparked as expected.

Re-wired the coil power supply connection this morning and hope to get it all buttoned up this weekend.

What a nightmare! Thanks for all of the help and suggestions. I guess I now have spare parts for the entire ignition and fuel injection system

for your patience!