can you measure the resistance of the s/c clutch coil and post back what it is?

still looks like K40 to me... sounds like a relay on the k40 board wont stay energised or is intermittently failing. what you can do is potentialy borrow a k40 from a known working SLK230 and test it. failing that you can buy one new (they arent too much)... others might chime in but my money is on k40...

i suppose you can give the relays a clean with electrical contact cleaner and re-lube them with say a spray of WD-40 and see how you go?

 

Thanks Subby - the s/c clutch coil resistance was 3.7 ohms from memory. I tested all the relays for operation off the car - all ok and dragged a piece of paper between the contacts - all clean. I didn't lube them - not a big fan of spraying dry circuit/relay boards with something containing mineral oil.

I take your point about it still being a dodgy relay on the board - it sounded to me as if the system was working perfectly for that short time period then the k40 relay was being told to shut off by a defective sensor somewhere - kinda like "limp mode". The engine runs fairly well but has to be "pushed" harder than I would expect from this kind of engine, indicating the charger isn't working (kinda quiet too!) - like I say no engine management light coming on so things aren't to far out of their perameters??

Should the bypass valve be fully open with just the ignition switched on (it goes through it's start up "click" routine) and also when the charger is operating, albeit briefly?

Could the MAF cause this if it was out of it's working perameters? I can't find and resistance figures for the hot film anywhere :dunno:

 

Ok, here's some more info for anyone interested.....................

The wire which resets the s/c clutch after deing disconnected and reconnected is the red/green wire in the "B" multiplug, B3. This is an "out" wire from the relay, in other words, when the relay is energised the connection is made from the thick red wire next to it on #1 pin (#2 is blank). I back fed a voltmeter into the 3 red wires while running the engine, they all show battery voltage (the red/green wire shows zero volts with the ignition turned off).

None of the wires shows a signal failure when the s/c clutch turns off after 7 seconds (the red/green wire increases 1 volt but only because the s/c clutch has disengaged and raised the battery charging voltage slightly)

The brown (earth/ground) wire has around 2 ohms resistance when checked from the wire to earth - pretty much to be expected - HOWEVER - when the engine is running, there's a 25 ohm resistance showing. I ground the cable directly to earth but nothing changed. Any thoughts on what could be giving the 25 ohm resistance or could it be spurious because the system is powered up and running????

What is bugging me is that the relay is holding in as soon as the ignition is switched on and is staying held on, even when the clutch disengages!. I've tried to make sense of the K40 board diagram here http://www.benzworld.org/forums/att...tachments/r170-slk-class/85305d1147897584-k40-relay-circuit-associated-k40-.jpg but the relays and A - E plugs don't correspond with my relay board.

Here's some pics, apologies for the long winded post but it's now bugging me.................thanks in advance

# 3 pin on block B is the culprit.

Red/green wire is #3 on block B (second block from the bottom, ignore the in line fuse holder, I've disconnected it from the loom)

The middle relay is the one fed from the red/green wire on block B

 

Are you still getting the MAF fault code?

On our USA models the supercharger is used temporarily on cold start up as an AIR pump for exhaust emissions, so it runs right away for that purpose. Not sure about your country. But after that I expect the supercharger would stop running as no boost is needed at idle. So it would be normal I think for it to shut off. I don't know why it would even come on at start up if not for emissions. The supercharger clutch is run by the ECU and the ECU apparently doesn't like something.

 

DickB is right some countries use the S/C upon startup for emissions control.

but to confirm its working

1) s/c clutch should not be enagaged at idle (it may be enaged upon startup depending on the country)

2) when car is in D and taking off the approx engagement RPM is about ~900rpm

3) when car is in N or P the clutch should enage around 1900-2200rpm. get someone to go inside the car while car is in N or P, and slowly rev it to that rev range. watch the clutch carefully to see if it enages and starts spinning as one unit clutch/pulley. if it does its working OK

 

@ DickB - the MAF P0100 code still shows.

@subby - the clutch only engages in N or P at the range you mention BUT only if the red/green wire is pulled from the B connector block and refitted (with the ignition off) The relay stays energised (I ran it with the plastic cover off) even when the clutch disengages.

I've stared at the wiring diagrams I found in "search" til I'm cross eyed - difficult to get fine detail, even magnified but it's obvious the red/green on block B feeds the fuel pump. I just can't understand why disconnecting/reconnecting that single wire would kick the clutch into life :dunno:
If I manually close the middle relay which closes the circuit to feed the fuel pump, the throttle bodies both click into operation.

Can anyone tell me exactly what sensors operate/deactivate the clutch?

I've read that pin F21 at the ECU earths/grounds the s/c clutch but at the same time another info table mentions that F21 is "Magnetic supercharger clutch activation output" and that 12 is measured at pin A2 (wherever that is) :surrender:

Thanks for your help - some fantastic info on here. Think I'll have to give up my day job to read it all

 

from WIS, how the system works

 

Good stuff Subby - thanks.

I reckon I've now established that the problem lies within the earth switching side. With a voltmeter back probed into the live side of the cylindrical connector at the s/c clutch, battery voltage is present during the short time the s/c is active and also during the same time that the s/c cuts out - so, it's back to the earthing side (although I still can't work out why disconnecting/reconnecting plug "B" on the K40 should temporarily activate the s/c :dunno

I checked for continuity/resistance between the earthing side of the s/c clutch (blue/green wire) and pin F21 on the multiplug connecting to the ECU. No problems there

Could someone explain to me what "Engine at idle, measure to pin A2 12vdc" is in the instructions below at F21? I'm getting into "frightened to blow the ECU" territory now

 

Ok, further developments (or not!)

As well as having no s/c working (except when disconnecting/reconnecting the wires mentioned in previous posts) is seems the air recirculation valve is always open. It goes through it's start up checks, moving a few degrees in the process but is always fully open. The feed and ground are good but get only around 1.5 volts from the green/blue wire at ant time. I also checked with the wheels off the ground and the traction control switched off.

The continuity from the air recirc. valve and plug E3 at the ECM is good. I temporarily earthed the blue/green wire at E21 on the ECu with the engine running (probably a big mistake!) and the s/c clutch came on - problem now is that with the earth wire off, the s/c clutch now doesn't turn off! I guess I may have blown the MOSFET in the ECM - any thoughts??

Is there a boost sensor in the pre face lift 230s? I'm guessing now that the s/c clutch etc. could stay off if the ECM had a spurious signal telling it there was too much boost????

Has anyone changed the MOSFET in the ECM box? I've just read 14 pages of info about fitting an override switch back in 2007/8 and filling my head with more information.........................

 

I haven't tackled the ECM yet but I will soon when I take my car off the road for the winter next week. My problem is similar except my s/c clutch is engaged all the time (whenever the car is running or the ignition is on). I suspect F21 is being grounded by the ECM all the time due to a bad power transistor since they often fail "shorted".

I've been trying to aquire a schematic before I tackle this but haven't had any luck thus far. I'll keep you posted when I rip into it.

Rob

 

Thank you Rob, I'll keep posting here with anything I find. I found today that the inlet cam variator isn't functioning - power going to it, earth grounding with the ignition off but not grounding when running - sounds like the same earthing component in the ECM is affecting both??

There's also a rythmic light "clicking" (not loud enough for "tapping") from the cam cover around the front of the engine whether the variator connection is on or off I'm assuming the variator is controlled by oil pressure and the magnetised unit??

Although I've thoroughly checked the K40 relay board and resoldered all the points I'm still tracking down a spare board to try it out to eliminate the K40.

PS - interesting handle Barchetta - I looked at a couple of them before buying the SLK.................

 

Hi dandywarhol,

I pulled my ECM (ME-SFI) apart to take a look and see if I could trace pin F21 back to a transistor (MOSFET). It's a multi-layer board so many of the traces are sandwiched in between so I didn't get anywhere. :crybaby2:

I need a schematic to go any further. The car still ran after I buttoned it all back up.

P.S. - Barchetta was the name of a boat I used to have. BTW my Dad was born in Cambuslang.

Rob

 

Thanks for that - I might do the same and have a look for that particular MOSFET. I got quoted £60 for a full ECM test and £360 for a rebuild today. think I'll do a few checks first.

Are the MOSFETs identifyable by number? From what I've gleaned on here MOSFET # BUK101-50DL has been known to cause problems - do you think it is identifyable?

Does anyone know if replacing an ECM with a like for like part number requires the ignition key/unit to be reprogrammed on a '99 car?

PS Barchetta means "little boat" in Italian................

 

Just reading this with interest.
Of course 2002 C coupe doesn't use a clutch.

But just to point you in the right direction.
MOSFET is most likely to be one of the items I circled IMHO.
Used to be an electronic tech long ago.

 

Many thanks for that. Does anybody have an equivalent mosfet for BUK101 - 50DL?
I've done some hard searching but can't find anything close - I just don't know which units are the most important to get close to the original. Only thing near it is VNP 10N07. Has anyone tried the Chinese ones on EvilBay?

 

In the photo with the circled MOSFET's, the 2 on the left and the 1 on the bottom right are the BUK101-50DL's so that narrows it down a bit to which ones I need to desolder and test.

I did find a cross-reference list to this obsolete device and went down the list until I came to one that Mouser stocked...STP55NF06L.

The spec sheets are similar and it's a whopping $1.48. Worst case is it doesn't work....which it doesn't now anyways.

Sure beats spending $500 to send out the ECU to get "refurbished". I'm waiting until I have some downtime with the car....maybe this weekend. If you tackle it first, let me know how you make out.

Rob

 

In the photo with the circled MOSFET's, the 2 on the left and the 1 on the bottom right are the BUL101-50DL's so that narrows it down a bit to which ones I need to desolder and test. I did find a cross-reference list to this obsolete device and went down the list until I came to one that Mouser stocked...STP55NF06L. The spec sheets are similar and it's a whopping $1.48. Worst case is it doesn't work....which it doesn't now anyways. Sure beats spending $500 to send out the ECU to get "refurbished". I'm waiting until I have some downtime with the car....maybe this weekend. If you tackle it first, let me know how you make out.

Rob

Rob, you said BUL101.......did you mistype? BUK101 seems to be the ones folks talk about.
Alan

 

Ooops....I meant BUK....I'll go back and edit my post.

Good catch!

 

I changed my mind on the replacement transistor after additional research and will be trying to source a VNP14N04 which has overload protection circuitry similar to the BUK101-50DL.

The problem is finding a retailer who will sell you just a handful rather than a min. of 400 or more. Let the fun begin....

 

I'm not convinced (nor is an electronics lecturer colleague) that the 14 will be heavy enough for the job. There's so much overload protection it can only take 14 amps.

Sent from my GT-I9100 using AutoGuide.Com Free App

 

You are right and I had some concerns as well. The last thing I want to do is potentially damage the ECM by using a component that wasn't part of the circuit design.....a costly $$ mistake.

So earlier in the week I found a source for the BUK101-50DL (Ali Express).....sounds kind of hokey, I know. I ordered 5 ($14.00 including shipping) and they are currently on a slow boat "from" China. (chinaman)

They may not ever show up but I figured a cheap risk to take when the alternative is $450 to get the ECM refurbished.

Barchetta

 

Just took delivery of 3 BUK101 - 50GL locally - identical to the DL but short circuit load protection max. is 20v, the DL is 35v but the switching rate is faster.

Electronics guru says the differences are minimal and should do the job. The "G" denotes a higher standard of component (read more expensive) and reckoned the manufacturers of the ECM use cost effective components to at least see it through the warranty period.

I'm going to take the car off the road during the holiday period and start fitting (not me, the components!). It's my first time doing this - wish me luck

 

Thanks for the update.....glad you were able to procure some.

My car is parked until spring so I'm ready to go (just waiting for some really cold weather ). I'm still waiting for my BUK101-50DL's to get over here from China....if they ever do.

Good Luck with your replacements! Maybe I'll wait and see how you make out before I tackle it! If you are successful and it fixes our problem I'll be doing this.....:bowdown: More than likely I'll dive into it as soon as my BUK101's arrive.

I was going to remove each of the 3 transistors one at a time and test them with a multimeter as follows (they are NPN components):

The base-emitter (BE) junction should behave like a diode and conduct one way only.
The base-collector (BC) junction should behave like a diode and conduct one way only.
The collector-emitter (CE) should not conduct either way.

*** They usually will fail shorted from collector-emitter which explains why the return from the coil (F21) is always grounded.

Barchetta

 

Yep, we're working along the same lines

I'm guessing that there should be continuity from F21 to one of the pins on the suspect mosfet to identify which one it is???

Anything to watch out for when dismantling the ECM when you had a look in yours?

http://www.nxp.com/documents/data_sheet/BUK101-50GL.pdf

Alan

 

You could try that approach but since F21 is going to ground, you may find it has continuity to pins on several of the components.....I may be wrong though.

Disassembling the ECU is simple enough. The only pain is having to remove a bunch of spring loaded clip that secure the transisters to the aluminum housing of the ECU (for heat dissipation) before you can remove the board. You may not have to do that if you can identify the bad component and are able to replace it without seperating the pcb from the housing.

Also, there is a thin white strip attached to the back of each transistors to aid in thermal transfer to the housing so don't forget to swap it over to your new component.

I would also use an anti-static mat with a wrist strap as a precaution....unless you are working in an environment where the relative humidity is above 40%. The a/s mat w/wrist strap is cheaper than buying an %RH probe!

Hope this helps.

 

I replaced the mosfet today - and......................................the s/c clutch is still on all the time! The mosfet had blown with by backearthing attempts but something else must also be earthing the clutch.

The single mosfet on the right side is the one connected to F21 (thanks c230 sport coup). It's connected to the centre pin (collector/drain?) I didn't replace the other 2 BUK 101s on the left side - I don't know what they control......................

How's the slow boat Barchetta?

 

I got the correct mosfets from China and they are actually new components. I half suspected they would have been desoldered from used boards or something. I've been traveling for the past few weeks (currently in Copenhagen) so it may be a bit before I can give it a go.

Sorry to hear this didn't resolve your problem. Like you said, there may be other damage to the circuit. So you were able to confirm by testing that this mosfet was indeed bad?

I'll keep you posted!

 

Yes the mosfet was bad (I tested the new one for a comparison). The original mosfet(s) was manufactured in the Philippines for - not that far from China

The scratch marks were caused by clumsiness when removing the spring clips - I'll use circlip pliers next time