E28 M5 swap restoration

tn535i · Post by **tn535i** » Jan 14, 2013 2:15 PM

If you pinched the IACV hose and it ran better it sounds like the ICV stays too open always. Does the idle change if you unplug the TPS? I know you say you checked the operation and all but humor me. It might indicate the ICV is just staying in the neutral or home position. Maybe just a flakey TPS that checks OK sometimes but really has cracked solder joints. I've had that on 2 cars now.

I guess I am grasping for some reason the idle control is not regulating the ICV like it should. You know the ICV is humming/buzzing regardless of the TPS position. With a scope meter or freq scale on a DMM you can 'see' if the ICV really changes when the TPS sends it into idle mode.

I know at one point you had ICV and TPS swapped but I think it should not have harmed the ECU in any way since each is designed to see little or no resistance. If the TPS pins read 5v on the outer ends and the switch pulls them to 0v like it should all is still well.

Also, my S38 will idle (quite nicely actually) with the AFM unplugged but only that. It's as if the minimum fueling is enought to keep it running, but only. Crack the throttle and it dies. This is learned from forgetting to plug it back in.

savannah996 · Post by **savannah996** » Jan 14, 2013 2:44 PM

You are right on all counts as far as diagnosing. My icv does not buzz. If I shut it by applying 12v the idle is normal. unplugging it has no effect.

savannah996 · Post by **savannah996** » Jan 14, 2013 3:46 PM

EDIT:
I have also checked at the dme that pin 2 is grounded when no throttle is applied and that #3 is not. This should be a solid indicator the tps is not causing a high idle.
Does the dme get a signal when the ac is running ( or ac button pushed in)? I didnt see one in the wiring diagrams.

Well im down to thinking that:
A. Coolant sensor telling dme engine is always cold. (and requesting 1200 rpm idle)
B. Dme got fried when the tps and iac connectors were switched. (indicated by the dme is not buzzing the iac when the key on engine off). And basically not signaling the iac to close at all.

tn535i · Post by **tn535i** » Jan 14, 2013 5:49 PM

Unfortunately it sounds like the DME is not regulating the ICV. You should be able to feel it vibrating/humming/buzzing or whatever you like to call it. Maybe the TPS was just too low of resistance for the driver on the ICV and it burned up. It would be a dead short at idle and WOT rather than the 4-8 ohms range I think a ICV solenoid should be. Dang, that would be bad or at least explain the problem.

Can you measure votage on either leg of the ICV and see if either side gets some voltage during key on? I can't think exactly how the ICV pins compare to the TPS pins but maybe only one driver is 'burned'.

I looked at the schematic and 1 on the ICV isground while the DME (34) retract also goes to 1. So if you put a dead short on retract is probably did kill the MOSFET or transistor that drives this. That would probably cause it to go wide open with key on but then not be able to regulate downward.

Someone can surely figure out what it would take to replace just the board level component that does this if you know the true problem. And that should be a $4 part with three dabs of solder. Having the ECM rebuilt could be much more costly but might be an easier fix if you can tell someone just what happened and what the problem was.

Sorry about this.

tn535i · Post by **tn535i** » Jan 14, 2013 6:05 PM

Wait a minute. I don't have that exactly right. According to the ETM the center pin on each harness is not always the common so-to-speak.
The ICV gets 12v on pin 2 and the ECM takes that to ground for ICV retract on pin 1 and pin 3 to gnd for ICV open. So there is no voltage applied to the ICV by the ECM only bringing it down to zero potential with 12v coming from the fuel relay.
However, pin 1 on the TPS is ground then pin 2 idle with pin 3 WOT and pins 2 and 3 having computer +5V.

With it crossed up the TPS signal input to the Motronic was in danger I think.
ECM >> actual
TPS1 >> ICV retract no problem, nothing
TPS2 >> ICV +12v probably not good 12v to ECM +5 ??
TPS3 >> ICV Gnd no problem, same as WOT signal

With it crossed the ICV control would see nothing that could harm it I don't think.
ECM >> actual
ICV1 >> TPS ground, nothing should happen
ICV2 >> TPS Idle input, nothing should happen
ICV3 >> TPS WOT input, nothing should happen

You need to see if you still have +5v on TPS pin 2 when the throttle is open. If not then the +5 from the ECM is lost and there can be no idle on I think. Again this should be a transitor or something that turns on when it's grounded. If burned it can't turn on and do whatever it does.

Just a theory or two to test out but I'm no EE.

I'm really a bit confused at this point.

and more

So confusing because your earlier symptoms sounded just like a idle plug to put the ECM in open loop with idle control.

Only thing I am sure of, do not attempt to recreate the conditions that got you here.

I'll be mulling it over and maybe have more crazy ideas.

savannah996 · Post by **savannah996** » Jan 14, 2013 6:58 PM

EDIT:
Man that is confusing.
Im looking at what the sensor did to the wires when I try to figure our what happened.
I have 5v at the tps plug, pins 2-3. (was I supposed to measure back at the dme?)

What im afraid happened was the tps shorted together (with no resistance) iac connector pins 1-2 when I plugged it into the tps, (I never gave it full throttle so pins 2-3 were not contacted).

The iac coil has 21ohms each, so pin 34 on DME is not ever supposed to see that much voltage if im correct. How much voltage is left after 12v goes through a 20 ohm iac motor?.

So 12v was applied to pin 34 on dme.

tn535i · Post by **tn535i** » Jan 14, 2013 7:01 PM

tn535i wrote:With it crossed the ICV control would see nothing that could harm it I don't think.
ECM >> actual
ICV1 >> TPS ground, nothing should happen
ICV2 >> TPS Idle input, nothing should happen
ICV3 >> TPS WOT input, nothing should happen
....
I'll be mulling it over and maybe have more crazy ideas.

Wrong after thinking more. Since idle input takes TPS 1 to 2 then a full 12V with no resistance was supplied to the driver at ICV pin 1 from ICV pin 2 meaning retract control at ECM 34 was seeing full 12V with no resistance every time the throttle was closed. Going WOT would take 3 to 1 and then nothing would be applied.

So I am back at the first conclusion that the driver for the ICV retract is toast. Sorry.

But you should still be sure that you have 5v on pin out 2 from the Motronic and that is goes low at closed throttle and high off throttle. I am hoping application of 12v to computer 5v did not do any damage.

tn535i · Post by **tn535i** » Jan 14, 2013 7:04 PM

I guess I was typing again whilst you were too. We've come to the same conclusion I think.

savannah996 · Post by **savannah996** » Jan 14, 2013 7:07 PM

Yeah I edited my post.

savannah996 · Post by **savannah996** » Jan 14, 2013 7:08 PM

I could measure the voltage of pin 34 on dme while its running.

savannah996 · Post by **savannah996** » Jan 14, 2013 7:44 PM

Pin 34 has 13v while running. So I dont think it matters if the tps shorted 12v to pin 34. Kind if suspicious though if you ask me. Does anyone know if I can use another model dme in place of a 0 261 200 079?

mooseheadm5 · Post by **mooseheadm5** » Jan 14, 2013 7:53 PM

savannah996 wrote:Does anyone know if I can use another model dme in place of a 0 261 200 079?

No, you cannot. I believe that an M88 ECU may run the car, but it is not tuned for it at all. Would probably be far more expensive, too.

savannah996 · Post by **savannah996** » Jan 14, 2013 8:27 PM

Dam BMW, why did they make this so hard! (dme swap issue).
Im gonna look into converting this to motronic 1.3, IF I cant fix this without spending $350 on a rebuilt DME. Im sure some has a tune for a S38 on a motronic 1.3 chip. I cant be the first to have thought of it. Hell I read somewhere a guy tried M3.1

I may just leave the icv unplugged. I drove 22 miles to work and she ran fine.

mooseheadm5 · Post by **mooseheadm5** » Jan 14, 2013 8:41 PM

savannah996 wrote:Dam BMW, why did they make this so hard! (dme swap issue).

Because it was a limited production car.

tn535i · Post by **tn535i** » Jan 14, 2013 8:42 PM

OK to check that and both 33 and 34 have 12v but I'm not sure that will help diagnose anything. There should always be voltage at either location since the computer completes the circuit and pulls it down to ground internally to actuate the ICV.

I was thinking about it the other way around earlier but now I hope I have it straight. 12v always goes to the ICV and then the computer pulls either retract or extend (no jokes please) to ground to energize one or the other constantly on a time based cycle. More open means more time with 34 to ground and more closed means more 33. It is always either opening or closing but it happens fast enough the solenoid just sorta hovers in between based on the time based difference.

You can check voltage but I'm going to guess it will be 12v on each. I think you will need to check current on the close side to see if there is any at any point including key on. If not it cannot close. On the other hand you should get open current if even briefly at key on.

I think you ought to seek the S38 troublshooting kit to see if a known good motronic makes it do right.

tn535i · Post by **tn535i** » Jan 14, 2013 8:47 PM

Missed there was another page of posts...

So why not for the time being put some form of restriction in in the IACV or hose or even just clamp the hose whilst you think of options. I am wondering if anyone has a schematic for the motronic. Heck, I have a bad e28 one and since the pinouts are the same basically I'll bet I could figure out what needs to be replaced. Like I said earlier it's probably a $4 part or you might just salvage on from another (non S38) computer that uses the sam IACV. How hard would that be?

savannah996 · Post by **savannah996** » Jan 14, 2013 9:17 PM

Before I go further I want to rule out the possibility of a ac malfunction causing the idle to be high. Anyone have info on this?

Also I take it this is the temp /time sw? (big brass one)=

tn535i · Post by **tn535i** » Jan 14, 2013 9:25 PM

IF I'm right each of the IAC drivers first passes through what looks like a jumper but might just be a fusable link on the main board. I have circled the retract with two yellows and the open with one. The retract is next to what appear to be an IACV capacitor (big red thing) and the ICV appears to be controlled by the bank of 5 transistors plus some other stuff. They use a bank I'd guess to controll it in a way that it can only be opening or closing and never both but it really goes so fast back and forth it hovers in between.

Anyway it might be worth a peak to see if that link is burned. Hopefully it is there intentionally to protect the controls from a short since the circuit probably relies on the resistance of the solenoid to limit the current. Your closed TPS took it to a dead short unfortunately.

savannah996 · Post by **savannah996** » Jan 14, 2013 9:25 PM

To answer your question about rigging the icv, you must have missed my earlier post. It currently has the ivc unplugged. Then I applied power to pin 2 and grounded pin 3 which retracted the valve all the way. Its idling nice. I just want to fix the real problem before I install my Dinan chip.

tn535i · Post by **tn535i** » Jan 14, 2013 9:28 PM

Why did my picture not stay in the rotation I had it. And since the M30 and S38 management is so very similar I am betting the internals of the std e28 Motronic look just like an S38. I'm not going to go open up my other car though. I've already gone above and beyond IMHO.

tn535i · Post by **tn535i** » Jan 14, 2013 9:29 PM

savannah996 wrote:To answer your question about rigging the icv, you must have missed my earlier post. It currently has the ivc unplugged. Then I applied power to pin 2 and grounded pin 3 which retracted the valve all the way. Its idling nice. I just want to fix the real problem before I install my Dinan chip.

Yes but it might not stay there. Of course you could periodically apply 12 v to the retract side to hold it there. Like maybe hook it up to the brake lights...

tn535i · Post by **tn535i** » Jan 14, 2013 9:34 PM

It was rotated to the left last I looked but now it's like I wanted it

mooseheadm5 · Post by **mooseheadm5** » Jan 14, 2013 9:37 PM

tn535i wrote:Why did my picture not stay in the rotation I had it. And since the M30 and S38 management is so very similar I am betting the internals of the std e28 Motronic look just like an S38. I'm not going to go open up my other car though. I've already gone above and beyond IMHO.

It is sort of halfway between the M1.0 basic used on the 533 and the M1.0 Adaptive used on the 535, so it is sort of different from both.

savannah996 · Post by **savannah996** » Jan 14, 2013 9:41 PM

tn535i wrote:IF I'm right each of the IAC drivers first passes through what looks like a jumper but might just be a fusable link on the main board. I have circled the retract with two yellows and the open with one. The retract is next to what appear to be an IACV capacitor (big red thing) and the ICV appears to be controlled by the bank of 5 transistors plus some other stuff. They use a bank I'd guess to controll it in a way that it can only be opening or closing and never both but it really goes so fast back and forth it hovers in between.

Anyway it might be worth a peak to see if that link is burned. Hopefully it is there intentionally to protect the controls from a short since the circuit probably relies on the resistance of the solenoid to limit the current. Your closed TPS took it to a dead short unfortunately.

Ill pull it open tomorrow.

tn535i · Post by **tn535i** » Jan 14, 2013 9:42 PM

mooseheadm5 wrote:
tn535i wrote:Why did my picture not stay in the rotation I had it. And since the M30 and S38 management is so very similar I am betting the internals of the std e28 Motronic look just like an S38. I'm not going to go open up my other car though. I've already gone above and beyond IMHO.
It is sort of halfway between the M1.0 basic used on the 533 and the M1.0 Adaptive used on the 535, so it is sort of different from both.

I would be surprised if they didn't use the same architecture. In which case it's possible there is protection built in. You just need to hold the board up to the light to figure out where pin 33 and 34 go first.

savannah996 · Post by **savannah996** » Jan 14, 2013 9:44 PM

Why is there a thermo time switch? On the eta it was used for the cold start injector?

Also, why cant I find the wiring diagrams anywhere?

tn535i · Post by **tn535i** » Jan 14, 2013 9:45 PM

You know what else. I can't find that temp switch on the ETM schematics, the big brass one you ask about. But the one behind it in blue is the CTS.

And I think you already said you did it, but I'd leave the ICV unplugged until you figure this out.

tn535i · Post by **tn535i** » Jan 14, 2013 9:49 PM

search wedophones bmw schematics or anything close to that.

There is not a TTS like the one in an eta for a CSI because the S38 doesn't use one. There is however (in an e engine) the TTS, the CTS, a coolant temp switch, an air temp switch and of course the temp sensor on the AFM. Really BMW, you needed that much information to know it got cold. In an eta the CTSwitch closes above 45C. Lets it know it's warm enough to run closed loop I think and maybe the one on the S38 does the same.

savannah996 · Post by **savannah996** » Jan 14, 2013 10:01 PM

tn535i wrote:You know what else. I can't find that temp switch on the ETM schematics, the big brass one you ask about. But the one behind it in blue is the CTS.

And I think you already said you did it, but I'd leave the ICV unplugged until you figure this out.

Oh my bad the bigger one in the fore ground is not a time switch. It looks like one though. Still is part mystery though. It has 2 wires, one (br/wt) is for dash temp gauge (verified). The other one is br/vi and idk, While unplugged it does not goto ground. With key on it has like <1 volt. Not on schematics from what I can tell, even on 528e and 535i. And yeah I have all the wedophones pdf's.

The little one in the back is for the etc. It has 2 wires and is a variable resistor. This part I have tested.

tn535i · Post by **tn535i** » Jan 14, 2013 10:12 PM

The one in the back is what most call the CTS Coolant Temp Sensor for the fuel injection input.

The one up front should have a part ending in 137 if it is like mine. According to BMWfans it is a coolant temp warning switch and will close above 120C. Maybe we have a high temp dash light? IDK, it's not in the etm I have. My car also has a coolant temp switch on the radiator that will disable the A/C if the radiator gets too hot that I have not seen in the etm.

One more thing while we are at it. There was some discussion about the uselessness of a barometric pressure sensor. I can't find one on my S38. Either that or it is extremely well hidden. I know it's NOT tied into any part of the AFM wire harness I can see like it is shown in the ETM.

MyE28.com

MyE28.com

E28 M5 swap restoration