I am installing an S50B32 engine from a 1998 Z3 M roadster into my 1999 E36 M3 and I have some issues with the swap that I need to resolve. I have heard that this swap has been done before and I am using the "search engine" to source material, but if someone has first hand knowledge on solving the electronic issues could you please chime in.

My car is running with the new engine but it is not 100%.

The engine that I purchased came from a rhd car but has a lhd wiring harness. It seems weird but that is what I was told by the engine importer. The wiring harnes is about a foot longer than an E36 wiring harness but there was not any issues with rhd components so I have to believe the seller.

Thanks, John

what do you mean by not 100%? what are symptoms?

It is misfiring and the timing retards under hard acceleration. We went out on a drive with a portable diagnostic machine and it shows the vanos system is working but when pushed the timing retards significantly and the engine just won't rip!

We are getting ecu faults, coil faults, and fuel relay faults, and it is running lean on three cylinders. We thought that the ecu was toast so I borrowed one from BSH and it did not change anything. The thinking is that there might be a pin or two on the Z3M roadster's wiring harness that are different than the e36 M3's. We know for sure that the fuel pump wiring was different so that was changed and my mechanic says we can ignore that fault.

sounds like you need to sort out fuel issue first, if lean detonation occurs, it will pull timing, maybe fuel pump is toast, ot maf not reading correctly etc

That is why I am searching for someone who did the swap from a Z3 M. It seems like it might have something to do with the fuel pump relay. They were wired differently in these two cars and maybe my mechanic has not set it up correctly. Any information you have is greatly appreciated.

put a fuel pressure guage along line somewhere between fuel pump and regulator, edeally in a way you can see it when driving, it should be 5 bar on S50B32

my engine is from a 98 Z3M with 5 speed.. tell me if i can somehow be of assistance

Member 'gobuffs' knows quite a bit about the difference in LHD, RHD, and MZ3 harnesses. Maybe he can help..

I have a pair of these engines that landed yesterday in Chicago, but won't make it to my warehouse until next week. (Sub 50K!)

I'm pretty sure based on conversations with Ken Arturian, (he's in BFC too on occasion) that the MZ3 harness handles the fuel pump relay differently than on the coupes or sedans.

How, I don't recall exactly, but Ken does.

I was with John for a few hours at the shop yesterday. He has a SMG ECU that was swapped out for my spare non-SMG ECU so any potential compatibility issues between the SMG ECU and the 6-speed manual have been eliminated. As for the harness, the length between the DME plug and the long black box that attaches to the firewall is nearly 2 feet long (much longer than that of E36 harness) and is consistent with the MZ3 harness (http://www.da-motorsport.com/projeler/mz3/mz3_en.htm). The fuel pump relay fault was pulled from both DME's so that's consistent with the fact that the relay is not found on the MZ3 harness.

Next step: Compare the ETM's of the Z3M and E36 for any other differences.

I have no clue what AKG means by the harness being LHD despite being from a RHD MZ3. :confused

Here are 2 pics to compare your harness with:

http://img.photobucket.com/albums/v420/trailerparkboy/Z3M_rhd_a.jpg
http://img.photobucket.com/albums/v420/trailerparkboy/Z3M_rhd_b.jpg

Didn't realize there was the new forum....subscribed!

Dreamer99 has emailed me.
1. His mechanic ground the replay seperately from the DME. May not be 100% of his problem, but can be an issue. The way the DME turns on the fuel pump is by sneding ground to the relay thereby sending +12v to the pump.
2. Need to ensure the speed signal from the diff to the cluster to the DME is intact. He may be experiencing a limp home mode problem.
3. Ther are 2 differences in the MZ3 harness versus and M3 harness. #1 the engine harness on the MZ3 doesn't have a fuel pump relay. It is after the X20 connector and is on the car's main wiring harness. #2 there is another wire that is moved on the X20 connector of the MZ3 harness. I forget what itwas but I HAD to bypass the X20 connector (there was no pin on the car side of the X20 connector that I could splice into)- I had to splice in to the right after the X20 connector on the engine side of the harness. I think it might be the speed signal thatmight be causing limp mode. Unfortunately my car is in 2 hours away so I can't just go look at it. I don't know if Ihave the notes anywhere. I might have time to run by there tomorrow when i go pick up the new project car.
4. Since he mentions runnign lean on 3 cylinders (not sure how he figured it was just 3 cylinders), he needs to ensure the fuel system is working at 5 bar.

Deleted....just realized you got the motor from Alex.

Deleted....just realized you got the motor from Alex.

I did not get the motor from Alex I got it from AKG, though Alex has been a great help.

You might need a different fuelpump.

A S50B30 car runs a 3 bar fuel system, with the same fuelpump as the rest of the E36's.

A S50B32 car runs a 5 bar fuel system with different injectors, and also has a different fuel pump.

hmm i put the S50B32, with a new stock E30 fuel pump and it ran fine.. AFRs before redline are ~12.5:1, and it has been a whole year(daily driver) with no leanness problems at all...
im sure a stock E36 M3 pump can handle it too..

Question: Do MZ3 harnesses have a VERY long plug for the HFM? I think I may have one...

I don't recall it being very long but it might be longer. The longer part of the MZ3 harness is to the DME which makes it slightly better (IMO) for the swap if you can work around the fuel pump relay issue.

I don't recall it being very long but it might be longer. The longer part of the MZ3 harness is to the DME which makes it slightly better (IMO) for the swap if you can work around the fuel pump relay issue.

I can't even begin to tell you guys how helpful this thread has been to me!

I asked because my harness is missing two things:
1. First gear "Nanny" sensor - this would make sense since MZ3's had the 5spd and therefore no plug associated only with the 6spd's on M3 evo's.

2. x20 connector(round black that clips to side of fusebox)- on the M3's there are two, my harness only has one, just like the roadster in the pic above yet nothing appears to be cut on mine. After looking at the pins on my only x20, #13 has been cut completely and #21 has been spliced/modified to a separate relay that was definitely added by someone(see photo below). I think thats the pin for the fuel pump?

My car is an OBD2 US-spec car, so it has 3 of these black plugs(one of which has to do with ASC-T I beleive.) My motor/harness came out of a 95 LTW, which only has 1 of these X20 plugs. Euro M3 Evo's had two from all the pics ive seen on the web. Hopefully these pics of mine may help you too John..

mine...
http://www.indexofimages.com/uploads/2c3ebc670c.jpg

X20 connector. Here you can see the 'aftermarket' relay on top connected to the red wire that gets power from the terminal lead in the fusebox, and the lengthy HFM plug
http://www.indexofimages.com/uploads/d42155134f.jpg

For reference, here is a photo of Frederick's bone stock Evo. It is RHD, but you can still see there are TWO black X20 plugs next to the intake. The HFM plug routing is also shorter, as the MZ3 looks like it wraps around the back of the firewall and along the fender, the Evo's does not.
http://i45.photobucket.com/albums/f55/b1mm3r/enginebimm.jpg

And lastly a US-spec 95 M3. Only 1 X20 plug
http://www.indexofimages.com/uploads/8f1ce87d47.jpg


I have everything hooked up in my own car and it would not fire. It would turn over but just spin. I don't know why so I gave up and instead of trying to match all the corresponding pins, I bought a 'new-to-me' LHD harness from a '96 M3 Evo 6-speed off ebay.de to start fresh. I have both the US and Euro ETM for my car and figured the proper harness would be alot easier than troubleshooting.

You can't have 2 X20 connectors. One is an X20, the other is something else.

That looks like about the same length as my MAF sensor wire.

You can't have 2 X20 connectors. One is an X20, the other is something else.

That looks like about the same length as my MAF sensor wire.

Im sorry, youre right... I only said that for description. 'Technically' it IS something else, but you got the idea! :D

I did try quite a bit on the phone last week.....

(Hello Bruce, via the forum.....Hard to believe after meeting up in person at Putnam last week that I we never got to talking about BFC's forum!)

BTW, within our little euro community in the US, it was pretty fun to actually hook up and talk to Bruce in person... we've talked via the internet and phone for years and never met in person.

Don't tick him off, he's a lot bigger than I realized! Bruce helped me trouble shoot a brake light issue at the track.... it was kind of fun in a weird sort of way. Anyway, missed the deleted post he had here.....was it important?

I will soon have a pair of the MZ3 engines here as as I said. One is pretty much sold already to a forum member, the second I hope to have a VERY close look at, and try to identify the differences mentioned here, as well as any others I can identify on them.

I would try to identify FOR sure if the 3 cylinders are on one header, if so, it is very possible its matching 02 sensor is bad. If the 3 cylinders are interspersed between headers, well, not likely. Mixing up the 02 sensors is a no-no, and the engine will NOT run properly.

I did try quite a bit on the phone last week.....

(Hello Bruce, via the forum.....Hard to believe after meeting up in person at Putnam last week that I we never got to talking about BFC's forum!)

BTW, within our little euro community in the US, it was pretty fun to actually hook up and talk to Bruce in person... we've talked via the internet and phone for years and never met in person.

Don't tick him off, he's a lot bigger than I realized! Bruce helped me trouble shoot a brake light issue at the track.... it was kind of fun in a weird sort of way. Anyway, missed the deleted post he had here.....was it important?

I will soon have a pair of the MZ3 engines here as as I said. One is pretty much sold already to a forum member, the second I hope to have a VERY close look at, and try to identify the differences mentioned here, as well as any others I can identify on them.

I would try to identify FOR sure if the 3 cylinders are on one header, if so, it is very possible its matching 02 sensor is bad. If the 3 cylinders are interspersed between headers, well, not likely. Mixing up the 02 sensors is a no-no, and the engine will NOT run properly.


They were not all on the same header

You can't mix up the O2 sensors unlss you put them in the wrong header. The plugs are different (as is juast about every other sensor lead has different plugs to prevent you from doing something stupid).

Question: Do MZ3 harnesses have a VERY long plug for the HFM? I think I may have one...I can't answer your question, but if you have slack in the harness (between the black box and the DME plug) once the DME is installed in the firewall then I'm confident you've got a MZ3 harness. As you can see in the pics above, the DME is installed vertically in the black housing mounted on the passenger fender apron of the Z3.

Dreamer99 is also using his US-spec cluster so that may also explain the timing being pulled above 5k RPM. He needs to get his hands on a euro cluster.

Dreamer99 is also using his US-spec cluster so that may also explain the timing being pulled above 5k RPM. He needs to get his hands on a euro cluster.

I'm not sure that makes a difference. I'd be more concerned trying to use an SMG DME than using a US cluster. The only thing you need out of the cluster is the speed signal and I don't think the signal from the US cluster is any different than the Euro one.

what exactly happens if u dont do the speed signal ?

limp mode...I understand it, it is BMW's way of easily getting around the 250km/h speed limiter. Theoretically you could cut the wire to the diff and then the DME wouldn't know how fast you were going and could exceed the 250 km/h limit.

I believe that any 6 cylinder cluster will work fine for giving the proper signal to the DME (from the diff) to avoid the limp mode.

I also believe that the limp mode was designed to help foil people with leased cars from disconnecting the diff to avoid putting miles on their car.

Alex.

I'm not sure that makes a difference. I'd be more concerned trying to use an SMG DME than using a US cluster. The only thing you need out of the cluster is the speed signal and I don't think the signal from the US cluster is any different than the Euro one.Ah, ok. At this time there is a non-SMG DME installed so there shouldn't be any compatibility issues there.

I just sent dreamer99 the ETM pdf files. I hope they're helpful.

I'm not sure that makes a difference. I'd be more concerned trying to use an SMG DME than using a US cluster. The only thing you need out of the cluster is the speed signal and I don't think the signal from the US cluster is any different than the Euro one.

I don't think it does. James@RRT was using a US cluster in his '95/S50B32 and did not report any of the above problems.

Dreamer99 says that it is running lean on 3 cylinders? How long did the motor sit? Are the injectors clogged?

Dreamer99 says that it is running lean on 3 cylinders? How long did the motor sit? Are the injectors clogged?It's possible they're clogged, but when I suggested that they be sent out for cleaning/balancing/blueprinting his mechanic was against the idea. He doesn't "trust" it and would rather swap them out for new injectors.

It's possible they're clogged, but when I suggested that they be sent out for cleaning/balancing/blueprinting his mechanic was against the idea. He doesn't "trust" it and would rather swap them out for new injectors.

Interesting....cleaning injectors cost less than $20/each. I would at least check that the injectors sound like they are working at idle (easy to do).

Interesting....cleaning injectors cost less than $20/each. I would at least check that the injectors sound like they are working at idle (easy to do).

How do you do it? I will suggest it to my mechanic. From what I understand my motor has been sitting for 2 years.

If it haas been sitting for 2 years, I would definitely get them cleaned. Fuel sitting in them for that long will gum them up.

Cleaning them is pretty easy. Take em out and send em to http://cruzinperformance.com/fuelinj.html Looks like the prices have gone up. When I started using them they were like $10/injector. Last time i used them (about 18 months ago they were either 11 or 12). Still cheaper than RC. With return shipping you should still be under $20/injector.

I got an update from CG, the fuel pressure is at 5 bar. They are now removing the wiring harness and checking it out pin by pin.

I got an update from CG, the fuel pressure is at 5 bar. They are now removing the wiring harness and checking it out pin by pin.


Would be interesting if it was a harness issue...my MZ3 harness has a problem with continuity on the crank position sensor wire....had to splice in another wire from a dead harness to get it to work....still need to send it in to get it permanently fixed. Will do that next winter.

I got an update from CG, the fuel pressure is at 5 bar. They are now removing the wiring harness and checking it out pin by pin.I compared ETMs yesterday. The Z3 harness goes to pin 13 of the X20 connector before going to the fuel pump relay, whereas the E36 bypasses the X20 connector.

I then realized that we need to look at the ETM for the US-spec 99 M3 to see how the connection(s) should be made for dreamer99's car. I still need to do that, though.

The Z3 harness goes to pin 13 of the X20 connector before going to the fuel pump relay, whereas the E36 bypasses the X20 connector.

That would explain why my pin 13 on the X20 plug of the harness was cut?

Check pin 21 Benji...

I jsut remembered...I have a RHD engine harness somewhere in the garage.

How frustrating for the MZ3 engines....

And to think that each and every one of them was here in the country before being shipped to ROW!!!!!!!!!!!!!!!

Its amazing that with the Z's being built in Spartanburg that some didn't ...well....you know....get out.

Alex.

Its amazing that with the Z's being built in Spartanburg that some didn't ...well....you know....get out.

Like the Nav computers? :-)

I got an update today. Some of the connections on the wiring harness were a little corroded so they got cleaned up. The engine will be rewired tonight. Hopefully this will solve the problems.

I have the US cluster in mine, still. My car is a 95 build (EWS-2) with a late 97/98 build S50B32 (cracked big end on the con-rod) from an E36 M3.

I notice a bit of hesitation in 1st gear running up through 4-6k RPM most of the time, possibly since the gear position switch is hooked up to nothing. I have the stock 5-speed in mine.

When I cleared the codes last week (gear selector position switch and #4 fuel injector open circuit intermittent) I noticed the hesitation in 1st was gone, so I need to figure out a way to get the switch wired/fooled correctly. What is the exact workaround for the gear position switch?

Anyway, let us know how the wiring harness cleanup works.

That would explain why my pin 13 on the X20 plug of the harness was cut?

Check pin 21 Benji...I think the modification to your X20 plug was for installation into Jared's LTW M3. So you'll have to look at the ETM for a 95 build US M3 for more insight.

After looking at the pins on my only x20...

#13 has been cut completely

That's because x20 pin13 is bypassed in the circuit to the fuel pump relay.

#21 has been spliced/modified to a separate relay that was definitely added by someone(see photo below). I think thats the pin for the fuel pump?

No, x20 pin21 is for ignition. Your aftermarket relay is necessary because the Z3 circuit has a "K6315 Unloader relay terminal 15" found between x20 and the ignition coils. This relay is absent on the E36 S50 B32 and 95 US M3 circuits so power is sent directly from x20 pin21 to the ignition coils. So one branch of your spliced pin21 goes to the relay and the other branch goes to the (input) pin56 of the DME, which is consistent for the MZ3, S50 B32 and 95 M3 circuits.

There should be a diagram on your aftermarket relay. Send me the diagram and I'll compare it with the MZ3 ETM.

The fuel injectors were pulled and tested and three of them had a poor spray pattern so they are being sent out for ultrasound and balancing. Maybe this is the problem.

Any update, Dreamer99?

Any update, Dreamer99?

They are supposed to get the injectors back this morning. I will post the results, hopefully today.

Right on. I look forward to hearing some good news.

BTW - did you ask Chris to swap DMEs? I might swing by there today and pick up my DME if it's out of the car.

Right on. I look forward to hearing some good news.

BTW - did you ask Chris to swap DMEs? I might swing by there today and pick up my DME if it's out of the car.

I spoke with Rudy earlier today and asked him to test it with my DME in the car since there was no improvement with yours so it should be there for you to pick up.

I spoke with Rudy earlier today and asked him to test it with my DME in the car since there was no improvement with yours so it should be there for you to pick up.Alright, I'll give them a call.

The cleaned injectors are in and it is still cutting out around 5500 rpm on cylinders 1-3. They are going to swap 1-3 and 4-6 electrical components tomorrow. THIS IS KILLING ME!!!!

How are they going to do that?

Have you reset the stored codes then reread them?

Here is what I received from my mechanic.

I notice a bit of hesitation in 1st gear running up through 4-6k RPM most of the time, possibly since the gear position switch is hooked up to nothing. I have the stock 5-speed in mine.

When I cleared the codes last week (gear selector position switch and #4 fuel injector open circuit intermittent) I noticed the hesitation in 1st was gone, so I need to figure out a way to get the switch wired/fooled correctly. What is the exact workaround for the gear position switch?Here's some valuable info from a guy in the UK who remaps the Siemens MSS50:

The siemans ecu cuts spark in first gear on the evo.

It cuts the spark at 4150rpm. I'll post a piccy of the map to show you.

http://e30gumball.users.btopenworld.com/1stgearmap.JPG

It doesnt cut the rpm and hold it, it just creates a flat spot of power loss as the power drops from this point but it will still rev all the way round as in any other gear.

The ecu knows that the car is in first gear by air pressure/force. This is greater in 1st gear than in any other gear. Also the rpm/speed ratio differs and the ecu takes a reading from the clocks to notice the gear.

I dont think (this feature) can be removed as it is not part of the flash memory on the main processor.

The 3 litre map doesnt have the same rpm cut as the 3.2

My apologies for the OT but I thought I'd reply to James' question. There is an "in-gear" switch that has been "bridged" in order to prevent the ECU from knowing it's in gear but the outcome has been confusing. Some have felt no difference and others have felt the flat spot disappear initially only to come back.

I can post the ETM diagrams of the "in-gear switchr" but I'll do that in another thread. (I have a hunch that bypassing this in-gear switch makes no difference.)

To steer this back on topic, Dreamer99 prefers to use his Siemens MSS0 5WK9 026 DME (for the SMG M3 and MZ3, according to the ETK) over the Siemens MSS50 5WK9 025 (found in the M3 3.2 6-speed) because he thinks the 1st gear "nanny" might be eliminated from the software. He could be right as MZ3 owners in europe swear they don't have this nanny. Maybe the 026 DME has this feature removed?

I'd like to do a back-to-back comparison (once his motor is running properly) of his 026 DME vs my 025 DME to determine whether or not the "nanny" has been eliminated from the program of the 026 DME.

Here's some valuable info from a guy in the UK who remaps the Siemens MSS50:



My apologies for the OT but I thought I'd reply to James' question. There is an "in-gear" switch that has been "bridged" in order to prevent the ECU from knowing it's in gear but the outcome has been confusing. Some have felt no difference and others have felt the flat spot disappear initially only to come back.

I can post the ETM diagrams of the "in-gear switchr" but I'll do that in another thread. (I have a hunch that bypassing this in-gear switch makes no difference.)

To steer this back on topic, Dreamer99 prefers to use his Siemens MSS0 5WK9 026 DME (for the SMG M3 and MZ3, according to the ETK) over the Siemens MSS50 5WK9 025 (found in the M3 3.2 6-speed) because he thinks the 1st gear "nanny" might be eliminated from the software. He could be right as MZ3 owners in europe swear they don't have this nanny. Maybe the 026 DME has this feature removed?

I'd like to do a back-to-back comparison (once his motor is running properly) of his 026 DME vs my 025 DME to determine whether or not the "nanny" has been eliminated from the program of the 026 DME.

The guys at CG still think that it has something to do with my DME. What do you think? Should I borrow yours again?

Check out this thread:

http://forums.bimmerforums.com/forum/showthread.php?t=693870&highlight=faulty+knock+sensor

I got the results back from the GT machine at BMW. There were as follows:

ignition coil #1 fault, occurred at 5500 rpm
ignition coil # 2 fault, occurred at 5500 rpm
ignition coil #3 fault, occurred at 5500 rpm
oxygen sensor 2 fault, control limit exceeded, out of adjustment range.

Does anyone have some ideas for this?

the first time i took my car after the conversion, the ignition coils started to act up after the car warmed up and the car ran really rough with no firing in a couple of cylinders. 6 new coils and 500 bucks later, it ran perfect. I can only assume your car definitely needs new coils and an 02 sensor. Im having 2 0xygen sensors installed this friday.

the first time i took my car after the conversion, the ignition coils started to act up after the car warmed up and the car ran really rough with no firing in a couple of cylinders. 6 new coils and 500 bucks later, it ran perfect. I can only assume your car definitely needs new coils and an 02 sensor. Im having 2 0xygen sensors installed this friday.

I have been told that they tried moving the coils around without any change in the faults.

Is oxygen sensor 2 for cylinders 1-3? The O2 sensor should be replaced and knock/ping sensor for cylinders 1-3 should be inspected and/or replaced.

Is oxygen sensor 2 for cylinders 1-3? The O2 sensor should be replaced and knock/ping sensor for cylinders 1-3 should be inspected and/or replaced.

The mechanic does not think the O2 sensor is the problem because it runs good until 5500rpm. He believes it is a running issue triggering the sensor.

I can catagorically state that flipping the 02 sensors causes hickuping and misfires and rough running at higher prolonged rpm.

It was a disaster on my car, and it took a while to figure out. Compounding that was an issue we believe of a questionable 02 sensor.

I strongly suggest flipping them on the headers from one to the other.

There are VERY few things that would cause bank one to differ from bank two other than the 02 sensor which reads each bank separately.... think about it.

How likely is it that all three coils randomly went on one bank? That there is a problem with those 3 injectors?

Remotely possible that some driver in your dme was damaged that runs those 3 coils, but still, even then, for it to have been damaged implies that those three were asked to do something separately from the other 3...again, its a bank of 3 joined by a common 02 sensor...

Check the airpump cover on the head to be blocked properly.

Alex.

How likely is it that all three coils randomly went on one bank? That there is a problem with those 3 injectors?

Remotely possible that some driver in your dme was damaged that runs those 3 coils, but still, even then, for it to have been damaged implies that those three were asked to do something separately from the other 3...again, its a bank of 3 joined by a common 02 sensor...The fault code for injectors 1,2,3 showed up on both DMEs. The injectors were pulled, cleaned and reinstalled. I highly doubt they were installed in the same order that they came out so I don't think it's an injector issue either. The ignition coils were rearranged but still a 1,2,3 ignition coil fault.

The mechanic needs to check the O2 sensors, knock sensors and for any issues with the wiring harness.

Any suggestions in addition to this would be great.

Check the airpump cover on the head to be blocked properly.The mechanic said the air pump system was installed with the motor and is functional.

I had the pleasure of driving Dreamer99's "new and improved" S50 B32 powered 99 M3. It appears that his mechanic may have resolved all the issues that plagued this motor earlier. The idle is smooth and it pulls nicely all the way to redline (...and, yes, I let the oil heat up, John ;)).

I was especially interested in sensing whether or not the first gear "nanny" was present. At both part- and full-throttle 1st gear accelerations I did not detect any flat spots or hesitation at 4100 RPM. This motor and DME is from a Z3 (with the Siemens MSS50 5WK9 026 DME) and the general consensus is that the 5-speed, S50 B32 powered Z3's did not have this "nanny".

I would like to swap my 5WK9 025 DME from a 6-speed, S50 B32 E36 M3 back in his "healthy" M3 and test whether it is present with my DME. I have a hunch the two DME's differ with regards to the 1st gear map...but I may be wrong. :dunno

So what was the issue? Or was there a bunch of things done and they couldn't say it was any one of the changes?

Or was there a bunch of things done and they couldn't say it was any one of the changes?That's exactly what he said...he had the injectors re-cleaned, swapped the coils around, cleaned up some of the connections on the harness, installed a new fuel pump (?) and swapped out the faulty HFM for a new one...all around the same time.

He did say the new HFM made a big difference.

I had the pleasure of driving Dreamer99's "new and improved" S50 B32 powered 99 M3. It appears that his mechanic may have resolved all the issues that plagued this motor earlier. The idle is smooth and it pulls nicely all the way to redline (...and, yes, I let the oil heat up, John ;)).

I was especially interested in sensing whether or not the first gear "nanny" was present. At both part- and full-throttle 1st gear accelerations I did not detect any flat spots or hesitation at 4100 RPM. This motor and DME is from a Z3 (with the Siemens MSS50 5WK9 026 DME) and the general consensus is that the 5-speed, S50 B32 powered Z3's did not have this "nanny".

I would like to swap my 5WK9 025 DME from a 6-speed, S50 B32 E36 M3 back in his "healthy" M3 and test whether it is present with my DME. I have a hunch the two DME's differ with regards to the 1st gear map...but I may be wrong. :dunno

Hey BSH, I am back from holidays and picked up my car today. It still runs rough when it is cold but it is much improved once warmed up. It does not go through first gear like I thought it would, especially with the 3:64 gears. I wound it to the redline in third and fourth gear on my way home and there was no hesitation. It pulled pretty good. Did it feel like 321hp to you?

Remember a few weeks back when we were listening to the rough idle of my car. It still seems to idle a little rough. Did you notice this? Maybe it has some big cams in it. What do you think? If you listen to the engine idling with the hood up it seems (sounds) like there is an exhaust leak from that small filter on the right side of the engine. I know it is not the exhaust but did you hear it? Any ideas? What is that little filter anyway?

So my wife and kids went to bed early tonight and I thought I would take my S50B32 over the Malahat (local mountain pass) several times. I had my right foot pinned to the firewall and rowed through the gears a bunch of times to get a good feel for the engine and I must say it really rocks. It goes from 6500rpm to 7400 in a flash. It takes some getting used to. The North American engine seemed to taper off after 5500rpm without this massive pull to the redline.

For the first and second 0-150kmh sprints, during 2nd and 3rd gear there was a hesitation/misfire/hiccup at 5500, then it recovered, pulled hard to 6500, it did it again, recovered and screamed to 7400. All the full throttle sessions after that it ran exceptionally well right through to the redline. Very weird. I will take it out tomorrow and see if it does that again.

Does this type of processor have a learning curve/adaptability? Maybe sitting for two years did not help it.

On a side note, I had the diff rebuilt with 3:64 gears. Should I be taking it easy for a few miles to break it in? It has been hard to do since I have had these motor problems.

i believe they do have adaptability.

breaking in e46 m3 diff i know is to do figure 8's slow in a parking lot to make sure the clutches break in.

i hope you fix your problem soon Dreamer!

By the way, i have a 98 Z3M engine with 5-speed in my car, and there is no nanny whatsoever..

What a difference a day makes. After some smooth (and fun) running last night it was the complete opposite this morning. My car would barely idle once it started, it seemed to get the "three coil fault" rough idle again. I drove it for about 20 minutes and it almost or did stall at every light. Squeezing the gas pedal seemd to make it get the three coil fault, it would just stutter until I let off the gas.

After I stopped at a store for a few minutes it ran much better without getting any stuttering. When I pushed it to the max it would get the big stutter/misfire/hesitation at 5500, then grab again and rip to 7400rpm. I just can't figure this thing out. My mechanic figures it has to be a faulty control unit since they have checked out everything else.

It still runs rough when it is cold but it is much improved once warmed up. It does not go through first gear like I thought it would, especially with the 3:64 gears. I wound it to the redline in third and fourth gear on my way home and there was no hesitation. It pulled pretty good. Did it feel like 321hp to you?When I was there the idle was fairly smooth. It wasn't much better than it was a few weeks back. When I upgraded to 3.73 from 3.15 I expected a HUGE difference, but it wasn't nearly as extreme as I expected. 1st gear is still very useful and each gear just feels more "lively". I really noticed the change when I drove my bro's 97 M3 (with stock 3.23 gearing) after zipping around town with the 3.73 for a few weeks; his gearing felt so long...it took some getting used to.

Remember a few weeks back when we were listening to the rough idle of my car. It still seems to idle a little rough. Did you notice this? Maybe it has some big cams in it. What do you think? If you listen to the engine idling with the hood up it seems (sounds) like there is an exhaust leak from that small filter on the right side of the engine. I know it is not the exhaust but did you hear it? Any ideas? What is that little filter anyway?When I heard the engine at idle I didn't notice any major issues. Keep in mind that the euro motor is louder and more mechanical sounding than the US motor.

That little filter is for the air pump. It functions only at cold starts so you probably hear it sucking in air when your motor is cold.

What a difference a day makes. After some smooth (and fun) running last night it was the complete opposite this morning. My car would barely idle once it started, it seemed to get the "three coil fault" rough idle again. I drove it for about 20 minutes and it almost or did stall at every light. Squeezing the gas pedal seemd to make it get the three coil fault, it would just stutter until I let off the gas.

After I stopped at a store for a few minutes it ran much better without getting any stuttering. When I pushed it to the max it would get the big stutter/misfire/hesitation at 5500, then grab again and rip to 7400rpm. I just can't figure this thing out. My mechanic figures it has to be a faulty control unit since they have checked out everything else.Hmm...Chris did mention that it would occasionally stall after cold starts, but it seems the underlying issues are still not resolved. :(

Give the DME a few days to adapt (hopefully, some of the issues will disappear) and then have it scanned for fault codes.

Did it feel like 321hp to you?I only drove it around the industrial area surrounding the shop so I didn't get beyond 3rd gear. It was rush hour and the 99 was bumper-to-bumper so I didn't get it on the highway.

I don't trust my "butt dyno" so I suggest you put it on a dyno and find out. Be sure to invite me when you go (if it's on the mainland). :)

good to see its getting closer to being sorted.

The 3.2's definetly have some sort of "Nanny" in first gear which is prevelant when you are running at low rpm then nail it. it doesnt seem to restrict when using big throttle from a standing start. It dont seem to be able to be tuned out with a DME reflash. U can get temporary releif form this by bridging the "in gear sensor", but over time it seems the DME learns what u are doing and restores it, having to go thou the whole proceedure again which is painful. IM getting it tuned in AlphaN shortly and hopefuly we can establish an AlphaN base tune for the 3.2, which may make these engine changovers a little less painful to get running well (no MAF, less dependancy on temprmental sensors ect).

I started it up this morning, and like every morning it runs on three cylinders when cold. Once I drive it for 10-15 minutes, turn it off for a couple, it seems to reset itself and then runs ok except for the hesitation (but not always) at 5500 rpm under full throttle. I thought that maybe the cats were causing some problems so I jacked her up and banged on them with a rubber mallet but they sounded clean. Damn.

Anyone have any more good ideas?

Hey John,

My apologies for not returning your call yet. What you described in your message is consistent with a bad knock sensor as described in the thread I linked to above.

I don't know what's going on at cold start, though.

I thought I would try my Peake tool on this engine and here is what I got:
14
19
17
18
35

So if I read the instructions correctly I should be going to table 14 (1996 and later cars) to find codes for 17, 18, 19, 35. There is not table 14. Which tables should I read?

I thought I would try my Peake tool on this engine and here is what I got:
14
19
17
18
35

So if I read the instructions correctly I should be going to table 14 (1996 and later cars) to find codes for 17, 18, 19, 35. There is not table 14. Which tables should I read?

I email Peake and they sent me table 14.
17,18,19 are ignition output transistor (2,3,1)
35 relay electric fan

So this ignition output transistor, would that just be my coils? I have already replaced them so something is triggering these faults.

So there is a table 14? Codes 17,18,19 are consistent with the fault codes pulled at CG and the $tealer:

I got the results back from the GT machine at BMW. There were as follows:

ignition coil #1 fault, occurred at 5500 rpm
ignition coil # 2 fault, occurred at 5500 rpm
ignition coil #3 fault, occurred at 5500 rpm
oxygen sensor 2 fault, control limit exceeded, out of adjustment range.

Does anyone have some ideas for this?Does that mean the Peake tool works on our motors? Hmm.

35 should be an easy one. Does your electric fan work?

Since you've already eliminated the coils as the culprit, you should look at the ETM carefully for other components "upstream" that could be the problem.

I sent John the 3 knock sensors I removed from nmy motor when I installed them. Didn't figure there was anything wrong when I got the motor, but it easier to replace them now than later. He should have them by the end of next week.

So there is a table 14? Codes 17,18,19 are consistent with the fault codes pulled at CG and the $tealer:

Does that mean the Peake tool works on our motors? Hmm.

35 should be an easy one. Does your electric fan work?

Since you've already eliminated the coils as the culprit, you should look at the ETM carefully for other components "upstream" that could be the problem.

Yes, the electric fan works when the a/c is on. There is something triggering a fault on the first bank coils.

I sent John the 3 knock sensors I removed from nmy motor when I installed them. Didn't figure there was anything wrong when I got the motor, but it easier to replace them now than later. He should have them by the end of next week.Given that all the 3 coils "fail" at the same 5500 RPM and the sensation is consistent with that of a faulty knock sensor (like lifting off the throttle momentarily) I'm betting $1 it's a bad knock sensor in bank 1.

Who wants to take this bet? :lol

Given that all the 3 coils "fail" at the same 5500 RPM and the sensation is consistent with that of a faulty knock sensor (like lifting off the throttle momentarily) I'm betting $1 it's a bad knock sensor in bank 1.

Who wants to take this bet? :lol
What about those three coils faulting when it is cold. Once warmed up, turned off, then started again it only faults at high rpm under full load? I will give you a buck if that is all it is. How hard is it to change the knock sensors?

I'm looking at the wiring drawings you sent me Benji and knock sensor 1 is for cylinders 1 and 2. Knock sensor 2 is for cylinders 3 and 4. I do not think new knock sensors are going to resolve my problems.

There are not that hard....easier on a Euro motor than a US motor. Just have to remove the plenum.

I'm looking at the wiring drawings you sent me Benji and knock sensor 1 is for cylinders 1 and 2. Knock sensor 2 is for cylinders 3 and 4. I do not think new knock sensors are going to resolve my problems.

The only thing I see on the schematics that links cylinders 1-3 is the oxygen sensor 1.

Have you replaced the O2 sensor?

Hae you swapped in a known good MAF?

Then you should buy a new O2 sensor before you visit the mechanic in Seattle and have him install it for you.

Have you replaced the O2 sensor?

Hae you swapped in a known good MAF?

I believe a good MAF was put in but I have not changed the O2 sensors. Anyone have some good used sensors for sale?

I believe a good MAF was put in but I have not changed the O2 sensors. Anyone have some good used sensors for sale?
Used O2 sensors are kind of a waste of time. They are easily damaged internally if dropped, and there's no really good way to tell how well they're working unless you can compare side by side to a known good sensor.

...My apologies for the OT but I thought I'd reply to James' question. There is an "in-gear" switch that has been "bridged" in order to prevent the ECU from knowing it's in gear but the outcome has been confusing. Some have felt no difference and others have felt the flat spot disappear initially only to come back.

I can post the ETM diagrams of the "in-gear switchr" but I'll do that in another thread. (I have a hunch that bypassing this in-gear switch makes no difference.)

To steer this back on topic, Dreamer99 prefers to use his Siemens MSS0 5WK9 026 DME (for the SMG M3 and MZ3, according to the ETK) over the Siemens MSS50 5WK9 025 (found in the M3 3.2 6-speed) because he thinks the 1st gear "nanny" might be eliminated from the software. He could be right as MZ3 owners in europe swear they don't have this nanny. Maybe the 026 DME has this feature removed?

I'd like to do a back-to-back comparison (once his motor is running properly) of his 026 DME vs my 025 DME to determine whether or not the "nanny" has been eliminated from the program of the 026 DME.
Another jaunt back off-topic (Sorry, Dreamer99, this is some fascinating stuff!)...

That's exactly what I feel - the dip right in that rev range.

I cleared all the codes using an Autologic, and the dip was gone, but only for a couple of days/starts. I guess it relearned its "1st gear" program and reinstated the nanny. Damn.

If Dreamer99's 026 DME really does have this feature disabled, that would be an interesting swap. Probably not worth it, since you are never in 1st gear on a track. And if you're a drag-racer, you can play with launches that avoid the major dip in torque.

Another jaunt back off-topic (Sorry, Dreamer99, this is some fascinating stuff!)...

That's exactly what I feel - the dip right in that rev range.

I cleared all the codes using an Autologic, and the dip was gone, but only for a couple of days/starts. I guess it relearned its "1st gear" program and reinstated the nanny. Damn.

If Dreamer99's 026 DME really does have this feature disabled, that would be an interesting swap. Probably not worth it, since you are never in 1st gear on a track. And if you're a drag-racer, you can play with launches that avoid the major dip in torque.That's very interesting. So the nanny can be disabled for a limited time.

I simply could not detect the nanny in Dreamer's 5WK9 026 controlled DME. :dunno

That's very interesting. So the nanny can be disabled for a limited time.

I simply could not detect the nanny in Dreamer's 5WK9 026 controlled DME. :dunno

So then the question would be, can we use a 026 DME with a 6 speed? It works with a 5 speed Z.

Then you should buy a new O2 sensor before you visit the mechanic in Seattle and have him install it for you.

There will be an S50B32 M3 in the garage where I am taking my car and we plan on swapping sensors off his engine to try on mine. Hopefully this will solve all of my problems.

Well boys, the underlying problem has been discovered and fixed. I can finally say "THIS ENGINE ROCKS!".

I went down to Pat's Autosport in Seattle where they have an S50B32 M3 that they race. The idea was we could swap parts for testing if required. JR put his diagnostic machine on my car and the same faults showed up. First three coils and O2 bank 2 sensor.

JR decides to look at the coils first. Pops the cover and look at that, the front three coils are different than the rear three. It turns out that the front ones (part# 1 748 017) are from a 540/740/M5. Hmmm, could this be the problem? He then asks me if the spark plugs were swapped. I had my mechanic swap in new ones during the engine install so it would have a fresh tune-up.

He pops out plug one and guess what? A Bosch Y5DDC plug from a 3 liter. He has seen this before and knows for a fact that the plugs cause problems in the 3.2. As he is pulling all of the plugs out plug 6 was not tightened up in it's socket!

JR grabs three coils and six plugs from the race car to install in my engine. Fire her up, reset the faults and he takes it for a rip. I could hear him going down the street and it sounded awesome. My turn next, 7500 in first, 7500 in second, 7500 in third, YeeHaw!! No misfiring and it feels like I just gained 100hp. Pull her back into the shop, test for faults and the only thing that showed up was the O2 sensor on bank 2. No more coil faults. I am going to order two new O2 sensors from BMW and get them installed this week.

I started my car up this morning to see if it would run cold and it ran perfect. It seems like the issues have been resolved, and they were all caused by the wrong spark plugs being installed. The plugs probably faulted the coils and caused other issues, but when diagnosing the problems the mechanic never went back to the basics. I am going to ask for some money back. He charged me extra for trying to diagnos the problems that he created and could not find.

On a side note, for the last couple of weeks my car was consuming gas like a crazy. Heading to Seattle I started the consumption monitor and I got 26.0 mpg on the highway. On the way back I was running the crap out of my car and got 31.4 mpg. What a difference six plugs and three coils make!

Ugh. Im so jealous.. yours is done and mine still wont run. :(

The MPG sounds right, but you may need to recalibrate your OBC? The Euro has different sized injectors.

Ugh. Im so jealous.. yours is done and mine still wont run. :(

The MPG sounds right, but you may need to recalibrate your OBC? The Euro has different sized injectors.

What is an OBC and how do I recalibrate it?

orsome that its running well. often we overlook the basics and look for more complex soultions.

THE OBC can be recalibrated. for memory there is a proccedure on this on understeer.com. U press the 1000 and 10 button for 5sec for memory to get to all the settings. there is lots of other hidden functions to play with also (thats if u have the OBC working).

Hey John,

I just got back from Vegas and heard your message. That's great news!

It appears that CG really dropped the ball on this one. You should definitely be compensated, IMO.

Hey John,

I just got back from Vegas and heard your message. That's great news!

It appears that CG really dropped the ball on this one. You should definitely be compensated, IMO.

Just before I went to Seattle Chris and I had a little falling out. I was getting extremely frustrated with the fact that he was unable to get my engine to run properly and had basically given up on my car so I let him know how I felt about it. He took it personally and asked that I take my business elsewhere. I am getting reimbursed for all of the extra diagnostic time and the cost of the spark plugs.

I have to find someone to do my valve adjustment now. Chris had originally said there were some tight ones and they needed to order some shims. How do you know what thickness to get? Or do you do the check and then order the shims? Can I reuse my valve cover gasket while I wait for new shims and then put in the new gasket?

By the way, there is no nanny!

Congrats on running well.

This must have been extremely frustrating.

The real question... (and I'm not sure if it was mentioned) is this....

Did the 540 coils run the proper plugs properly, or, did they continue to cause faults with proper plugs?

If they do throw faults...even with proper plugs... well.... its dissapointing (but it does happen) that the original trouble shooter didn't notice the difference between part numbers.... particularly when it was 3 vs 3 and suspect to begin with.

Glad its healed.

Alex.

Congrats on running well.

This must have been extremely frustrating.

The real question... (and I'm not sure if it was mentioned) is this....

Did the 540 coils run the proper plugs properly, or, did they continue to cause faults with proper plugs?

If they do throw faults...even with proper plugs... well.... its dissapointing (but it does happen) that the original trouble shooter didn't notice the difference between part numbers.... particularly when it was 3 vs 3 and suspect to begin with.

Glad its healed.

Alex.

We changed the plugs and coils so I do not know if the coils would have worked properly. I figure the plugs were not firing correctly and faulted the coils, and everything down the line did not run well.

I have my car in at a new shop today to finish my valve adjustments. The original mechanic said they were all within specs except for cylinder 6 exhaust valves. This other mechanic found that none of the intake valves were within specs and only half of the exhaust valves were. The guy who did my swap and original work must have been drunk! Could they get all out of wack in 1000 miles?

This new guy also found my left side motor mount was broken. Would there have been signs of weakness when they installed my engine?

My 3:64 diff seems a little noisy when I decelerate at highway speeds so I am wondering if they got that wrong as well. I am not sure I want my car back in his shop to check it out. Should I just spend the extra money and take it else where?

Man am I frustrated!

Sorry to hear John! If anything, your woes have made me feel a little bit better about doing all of my own work myself :P

Hope it's nothing major. How many miles are on your motor?

Sorry to hear John! If anything, your woes have made me feel a little bit better about doing all of my own work myself :P

Hope it's nothing major. How many miles are on your motor?


I wish it was not may daily driver, then I could have just made a project out of it.

My engine had 37,000 miles when I bought it and I have put 2000 on it. I love the 7500 redline. It really screams.

Hm, well assuming the valves should be checked every 15-20k yours should have have been done at least once so maybe you were in fact overdue for it if the previous owner did not do it around the 30k mi mark.

Projects suck man - don't let anybody fool you because it's RELIEF they feel, not necessarily satisfaction!

Yours did get 'done' in a very short amount of time though. I've had the luxury of time to be especially thorough on every step of the job. (Except installing the fuel lines :icon28XX)

Problem with those valve clearances is that a lot of shops don't check them and tools to adjust them are hard to come by it seems. Well at least over here even if your are next to germany :confused

Good one on the sparkplus though, have to remember that.