Richard Kilpatrick’s SLK decides that wet weather, rubbish tyres and motorways are far too great a risk, and becomes a little lethargic…
Not long after getting the SLK, and throwing back into work despite a barely functional leg, a quick blast up the M1 on one of the few sunny days during this ‘summers’ constant rain seemed like just the ticket. Getting to experience the little car’s performance was a useful bonding exercise, but it seemed a little unpredictable.
Overtaking which had been a joy, became worrying. And when it felt like the car needed all the throttle just to keep up with motorway traffic, it was clear something was not right. Given that Leicestershire had clearly decided an hour of sun was too much, and opened the heavens on the little yellow car, I assumed water must be the problem. But solving it would take, on and off, most of four months. Not, I hasten to add, due to any inherent difficulty, but prioritising costs and trying to balance repairs with the inevitability that the car was a toy, an optional thing, and the Chrysler was doing all the hard work.
Research began with checking how the supercharger works. On this era of SLK, the blower is engaged with an electromagnetic clutch, and then boost is controlled with a bypass flap – so two elements can conspire here. The engine management uses the usual mix of sensors and computers, with the critical elements on the SLK being the crank sensor, cam sensor and cam adjuster solenoid.
Given the same job to do again, I’d have it sorted in no time, but that’s no fun. So here’s where the route took me:
First, there was no evidence of the supercharger engaging. Except when it did, of course, which seemed mostly random – sometimes, it was working when the car was warm. Sometimes it wasn’t. Open the bonnet and move the throttle cable, and it should cut in around 2,000rpm with a distinctive noise and of course, visual confirmation that the pulley has engaged.
Most forums point you at a substantial relay box in the ECU compartment, the “K40” relay. It suffers from dry joints and can be a source of woe for many users – if nothing else, relays for things like the fuel injection system live in here, and are hidden from view when doing a normal “Blown Relay” check. Mine was okay, but whilst investigating I found that a small blower fan had burnt out, so that needed replacing.
The blower is £133 from Mercedes, for the module. The motor is £7 from eBay. Obviously, the motor was replaced, a 10 minute job made into 40 due to the bulky set of computers and looms that need to be moved.
Most research implicated the camshaft magnet as the likely culprit. Sure enough, when I checked my car I found that the solenoid was leaking oil down the front of the head, and more critically out through the electrical connector, onto the pins and into the loom. Early cars have the connector pointing up and it’s less of an issue than it is for the later models, for which Mercedes-Benz has produced an additional loom to protect the wiring from oil contamination.
In the worst cases, the oil in the loom can damage the ECU. Again, my early car was merely a little grubby, but enough for me to be confident that replacing this adjuster would fix the issue of no boost – if not, then at least it wouldn’t have a leak.
At £64 this seemed reasonable, so I bought that, and the correct black sealant, and the plastic part of the wiring connector (a pin bushing in MB parlance) to repair a broken clip. The new part fitted, the car was tested, and the supercharger worked.
With the images taken, it’s probably easier to flick through the gallery to see the oil contamination on the plugs, and repairs.
Except, the next time I drove it… it stopped after a while.
At this point, I did what everyone running a modern car should do, and bought a code reader. I found an Autel MaxiScan for £40 or so on Amazon, but the range of OBD-II scanners is quite wide and prices range from £20 to £2000 – the Autel has engine codes for Mercedes, and with a 38-pin adaptor I was on my way. For more sophisticated and body functions, the readers are far more expensive – these are just for engine codes on most cars.
A few codes remained from past repairs – the MAF in particular caught my eye. Upon removing the pipework, I found an unmarked third-party clone MAF, rather than the Bosch or Mercedes part, and these are generally a cause of poor running. However, at £180 (or £230 for the genuine one) there was no rush to replace it – the inlet and breather pipes were given a clean and the car checked over a few days, having cleared all the codes.
One came back – P0341, camshaft sensor. This can be thrown by a faulty adjuster magnet, as well as wiring, so after checking the loom condition I took the plunge and bought a Febi-Bilstein part from eBay.
The old sensor had a date code of 1998, so was clearly the original or at least, a secondhand one fitted during the car’s previous ownership, but at £133 for a new one the Febi unit was tempting for only £39. When it arrived, it looked like an exact copy of the Mercedes one, except the logo had been ground off.
In fact, it was so exact, even the moulding lines and texture of plastic matched.
And the old sensor has a Febi logo…
Turns out that various OEMs supplying German manufacturers quite often grind off the genuine part number and manufacturer logos. Whether there are quality control or other factors, I don’t know, but what I got in the box was to all intents and purposes the same item I’d have paid almost £100 more for from a dealer.
The good news is that with that part replaced, the car works exactly as it should. Lively, smoother and generally ‘better’. Following the vague techniques of ‘suck it and see’ without the code reader, I’d probably have changed the sensor anyway, but would also have ended up going down all manner of dead ends to get there.
The code reader isn’t an absolute indication, of course. It narrows down the process. For the cost, it proves to be a very useful tool regardless. And the SLK was repaired for under £200, when a couple of trips to a specialist or dealer would probably have resulted in that cost on diagnostics alone…