I have a strange starting problem in a 4,5 yoears old MB 220CDI (model year 2000). Every now and then it happens that when I turn the key the glow plug light goes on normally for a second or so but then when I turn the key to start the engine, nothing happens. No noise, no cranking, nothing at all.

Fist time this happened over a year ago. Mostly this has happened in cold weather and therefore I thought that the cause may be a weak battery. I let check the battery and it turned out to be OK, but an internal break was suspected. So I bought a new battery and the car was behaving OK for a year but now a couple of weeks ago it happened again. The outside temperature was -7 deg C and I let the auxiliary heater heat the car about 15 mins before starting. Then the car did not start, nothing happened even when I repeated the attempt. All the other electrical equipment were working normally. I thought any case that the battery has gone low again (which sounded odd because I had driven over two hours the previous day) and connected a small battery charger to it. Immediately after connecting the charger I tried to start again and the car started normally. Ther were perhaps 4-5 minutes between the first start attempt and the second successful one.

I have been reading several of these MB forums and found out that there are many kinds of starting problems, but I have found only one (on the German forum) where the behaviour was exactly similar but they were not able to explain it there either.

Now I would like to know if anybody has had similar experiences. Also if anybody could explain how the starting sequence in this car goes. What happens in which order and how could I narrow down more the cause of the problem. One theory, which I have, is tha it is somehow related to the key identification system which occasionally picks up a noise or something and the temperature is only a factor which increases the propability of this to happen. But I don't know enough about the system to be able to judge that.

Any opinion which could help is appreciated.

 

diesel engine needs more higher ampere current from the battery compared to gasoline engine.

it is advised to replace the battery every year.

 

Well, sleepwalker, with all respect I have to disagree here. What I admit is that a diesel engine draws higher current than a gasoline engine when starting, but my opinion is that no car (especially a quality car) cannot be designed in such a way that the battery has to be replaced more often than the normal service interval. I drive about 20 000 km per year and the Assyst system lets me to drive about 25000-27000 km between services. Thus I would have to replace the battery before the service interval has gone. I have had also diesel cars before and in those the battery has lasted several years even in our cold climate.

There are a couple of facts, which ( I think) speak for the assumption that this starting problem is not solely dependent on the battery:

1) When this problem occurs the car does nor even try to start. The starter does not turn at all and no sound, not even a click, is heard. Also all the other electrical equipment work normaly at the moment of that start attempt.

2) After a few minutes wait the car starts quite normally. The starter cranks the motor at a normal speed and the car starts immediately. No sign of weak battery.

I have not yet been able to measure the battery voltage on the moment of the unsuccessful start attempt, but next time when it happens, I'll try to do that.

It is of course true, that the battery voltage might be an indirect cause for this phenomenon. My current theory is that there is some computer or electronic circuit which is disturbed at the moment of e.g. inserting the key in such a way that it prevents the starting. It might be so that lower temperature in the electronic circuitry combined with the lower battery voltage makes the system more sensitive for disturbances. Being electronic engineer myself I know that e.g. capacitance values of certain type of filtering capacitors decrease considerably in low temperatures thus making the circuitry more sensitive for disturbances.

Of course this is very difficult to prove and very difficult to investigate without knowing exactly what happens when the problem occurs. Maybe there is some trace in the fault memory about what has happened.

Any case I try to do some more investigation to be able to narrow down the cause of this problem. I have now measured the battery voltage every morning before starting to find out if the battery charge level is going down. So far it has been varying between 12,3 and 12,6 volts depending on how much the car has been driven on the previous day. So it seems to be holding quite constant.

The investigations continue.

 

Just heard interesting story yesterday. Last Saturday one of my wifes workmates was going to visit their relatives for the weekend. But the car did not start. Nothing happened when he turned the key. He tried several times without success. Well, they took the family's other car and drove away. On Monday morning he thought that he tries once more before calling the repair shop. He sat into the car, turned the key and it started normally. No problem whatsoever.

And this car is a Volkswagen less than a year old.

I just thought that maybe Volkswagen uses the same components, motor control etc. as Mercedes. Both are German cars.

It looks like the horror story about the cars behaving more and more like computers is coming true. They just hang or crash without any obvious reason and the you have to boot them again from beginning. At least I hope that my next car would have a RESET button.

 

mark, you seem to got some electrical/signal problem.
only by reading the fault codefrom the memory that we could trace the culprit.

cars that uses fuel injections got engine control units.
then they integrate the anti-theft with it. no authorization... no starter or fuel output.

a few years from now, you can step inside your car and say "take me to Starbucks"... and you can take a quick nap.

our great grandchildren will ask us questions such as: "how does it feel to drive the car with steering wheel?"

 

Thanks for the answer sleepwalker. That's what I also thought that the fault memory could be the key, if there is information stored about this behaviour. I have to contact the local repair shop to check if they can read and interpret the fault codes.

Regarding the future I agree that the development is going on but I believe that the pace is slower than we think. The problem is that in computer technology the systems are so complicated and the number of affecting variables is so vast that the systems cannot be 100% tested. Therefore certain level of faults is commonly accepted in systems. We talk about the stability of the system. Now traditionally we have been thinking that a car is a product which should be faultless, especially what comes to the security behaviour. If the car is becoming more like a computer system and less a mechanical device, we come also into a situation where we have to accept and tolerate certain level of instability also in cars. Basically it is a question of cost. The stability can be increased by including more checks and double systems as in aircrafts. But how much we are willing to pay for it and use time for the checks. Therefore I think that the pace of the development is not so high as believed.