Andrew's S54 Swap

[Andrew325is]

Yeah its like that all the time. We tried hooking it up to DIS but it wouldn't communicate. I've used a generic cen-tech obd2 reader before that worked. This is how I have the obd2 port wired:
1 GN/WS 0.75 -> Ignition switched live (C101 #6)
4 BR 0.5 -> Ground
5 BR/SW 0.5 -> Ground
7 ws/vio/ge 0.35 Diagnostic line engine (Signal Data link TXD) -> X60004 pin 32
9 SW 0.35 Speed -> X60004 pin 17 and C101 #9
16 RT/SW 0.5 -> Continuous live

It showed the following generic codes:
P0418 - Secondary Air Injection System Relay 'A' Circuit Malfunction -> I'm assuming this comes up because I don't have the air pump installed.
P0500 - Vehicle Speed Sensor 'A' -> I figured this one had to do with DSC (speed sensor at one of the wheels?) but that is disabled. It couldn't be one of the camshaft sensors or the crankshaft sensor could it?
P0506 - Idle Control System RPM Lower Than Expected -> I don't know what this means. Vacuum leak?

We noticed the throttle kind of flutters so maybe the electronic throttle motor isn't getting enough current. I extended the wires to it and a few sensors to make the harness longer so it would reach inside the car. I used the same wires from another S54 harness but maybe one of the joints isn't good. The wire's are pretty thick and my solder iron is only 50 watts.

[u3b3rg33k]

That's what my M10 feels like when I open the secondary at too low an engine speed.

[LJ851]

Holy Torque Hole Batman!

[nando]

hook that bad boy up to a diagnostics interface. looks like vanos isn't working to me as well.

[nucci]

Thanks John. :) I hope you don't mind if I post the dyno from today. My car is in red. A stock 2002 M3 is in blue. A little problem to work out below 4k, but then it took off like a rocket. I shoulda took a vid with my iphone. :D

Little problem? That's what no VANOS looks like. Do you know what caused it? Does it do that all the time?

[Andrew325is]

Here's a dyno from today. My car is in red. A stock 2002 M3 is in blue. It has an issue to work out below 4k. The upper range is good. :)

[Andrew325is]

More cruise control research:

E85/E86 Z4 has a retrofittable stubby cruise stalk which has the FGRS output. TIS says it is compatible with all the different DMEs for that car. The FGRS signal from the E85 stalk, and the E46 MFL steering wheel control, connect to the same pin at the MS45 DME. The E46 MFL is identical whether the MS45 DME or the MSS54 DME is equipped.

Additionally, you can see that when the stubby E85 stalk is installed, there is still no gauge cluster indicator lamp for cruise control, whereas the E46 has one. This explains the K-Bus signal from the E46 MFL to the cluster and no additional signals on the E85 stalk.

Worth trying...

Interesting, although I don't see how the output would be different from the Z3 FGR unit. hmm... :)

[John T]

Got the pleasure of checking out this car today, super clean swap. Seamless!

[gianluca1505]

hello guys
i'm Gianluca , from Italy and I made the same work on my bmw z3. I have only one problem : i don't understand the electric cable of my s50b32 engine. do you understand the maps of circular connettor of that engine ? do you know the number of PIN what are? please can you help me ? ciao gianluca

[nando]

I probably won't bother with cruise control but it would be nice to have the option. :)

[nucci]

More cruise control research:

E85/E86 Z4 has a retrofittable stubby cruise stalk which has the FGRS output. TIS says it is compatible with all the different DMEs for that car. The FGRS signal from the E85 stalk, and the E46 MFL steering wheel control, connect to the same pin at the MS45 DME. The E46 MFL is identical whether the MS45 DME or the MSS54 DME is equipped.

Additionally, you can see that when the stubby E85 stalk is installed, there is still no gauge cluster indicator lamp for cruise control, whereas the E46 has one. This explains the K-Bus signal from the E46 MFL to the cluster and no additional signals on the E85 stalk.

Worth trying...

[nando]

good read, I've been talking to that guy so maybe I'll pick his brain a bit.

one possiblity is that the E46 does in fact use 10 pulses per rotation, and if that's true, I believe it works in our favor. I'll have to do the math to see how a 9 pulse signal works on a K-factor based on a 10 pulse signal.

anyway, getting the speedo accurate was never the issue to me, you can adjust the displayed MPH to your hearts content - getting an accurate odometer will be much harder.

[nucci]

I dunno - using 9 pulses per rotation, the K factors in the cluster were well under 1% off my E30's tire size (235/50-15). they are stored in KMs, I think it was only 10 or maybe 20 pulses off an entire mile? out of ~7500 or somewhere in that ballpark.

the reason I believe you cannot change the odometer's K-factor without getting a checksum error is because it compares the calculated mileage vs the stored mileage (pulses vs miles or KM). I think you'd have to zero out the odometer, change the K factor, and then change the odometer back up to what you wanted it to read. but, that's just a guess of what I've tried. maybe somebody else has modified the odometer's K-factor. I know for certain you can use whatever you want for the speedo..

Good stuff worth reading here.

Matt

[nando]

I dunno - using 9 pulses per rotation, the K factors in the cluster were well under 1% off my E30's tire size (235/50-15). they are stored in KMs, I think it was only 10 or maybe 20 pulses off an entire mile? out of ~7500 or somewhere in that ballpark.

the reason I believe you cannot change the odometer's K-factor without getting a checksum error is because it compares the calculated mileage vs the stored mileage (pulses vs miles or KM). I think you'd have to zero out the odometer, change the K factor, and then change the odometer back up to what you wanted it to read. but, that's just a guess of what I've tried. maybe somebody else has modified the odometer's K-factor. I know for certain you can use whatever you want for the speedo..

a simpler way would be to intercept the signal and modify it a bit..

[nucci]

this is really simple - you'd just run the VSS signal from an E36 or E30 cluster to the DME. since it all seems to be based off a 9 pulse per rotation signal, it doesn't matter what cluster it came from. the signal on the E30 is definitely cruder, but for what the DME uses it for, it doesn't really make much difference.

That was exactly what I was gathering.

if you don't use the LCM, the cluster doesn't really add a lot of drama - it really is just a display device when you run it stand-alone. If your transplant car has higher mileage than the cluster, you can basically run the odometer on a bench setup to get it to your desired mileage. certainly easier than removing the chip and all that. how long would it take to add 100,000 miles at 180mph? :p

Math it out - a little over 23 days!

although, there is an issue with recoding the K-factor without changing the odometer's checksum. basically, you're stuck with the K-factor's built in to the cluster for the odometer, even though you can code new ones, it will throw a checksum error because I believe the E46 cluster stores pulses, not kms or miles. You can change the speedometer to whatever you wish, which is nice. fortunately, the built in odometer K-factors are extremely close to what I'd need, close enough that messing with the odometer reading and checksum isn't worth it. And there are two to choose from which at least gives a little bit of wiggle room.

Cluster stores miles - I've got a couple of late E46 M35080 (odo chip) dumps. The first 16 8bit addresses are all summed up and that is the mileage. If you are talking about the same K-factor that I've read about, your only two options are equivalent to a rear tire of about 25.7" diameter, and one a tiny bit shorter. I'm running 245/40-17 at about 24.7" so it is not even close. I cannot imagine you've got 25.7" diameter tires tucked in an E30, do you?

what I really want to find out is, what does the DME actually do without the DSC/ABS system & emissions junk? if it's just going to throw an error light - who gives a shit? at the price people are charging for flashes, it hardly seems worth the trouble. I'll find out eventually but I need to finish other things first. :)

99% certain that the DSC tells the DME to use torque reduction strategies when it has certain error situations - what those are, I guess we will find out.