300E Cold starting problems... (-88)
 

Hello.

I have a -88 300E which is sometimes hard to start when the engine is cold and the outside temperature is approx zero Celcius (ca 32 F i guess). I have to crank it for ten-20 seconds and puch the accelerator until it starts and then the rmp is very low and the engine takes a couple of seconds to reach correct idle rpm... It runs smoothly as soon it has found the correct idle rmp...

When it is colder outside (approx -10 C or colder) it is easier to start it (strange). :eek:

I have checked the cold start valve and it works, the temperature sensor for the CIS-injection also seem to work (the resistance is approx the same as the "Haynes" guide tells), the idle speed actuator works, the EHA seems to work to... The spark distibutor and distributor cap and the spark plugs are new ones (changed this winter)... WHAT could possibly be wrong??
I will try to increase the CO-screw to enrich the mixture since the lambda sensor will force the engine to lean the mixture as soon as the engine is running anyway - right? Is this an good idea? Please help me...

I'm getting sick of not finding what is wrong!!! :mad:

Can it possibly be the OVP-relay. I've read on som www-page that this could cause cold start problems.... :confused: :confused: :confused:

/Ola, Sweden.

I had the same problem (and I am still working it). However I did manage to rig a temporary fix that works well for now. I ran a wire from the battery to a push type switch which I installed under the dash and then to the cold start valve. When I engage the starter I just push the switch for a moment and I get instant start. When the Temp is over 60 (f) I don't need this. But it sure saved my butt in January. If I ever figure out why it doesent work on its own I will let you know.

Monitor your Idle Control Valve voltage. I suspect the valve is gummed up and not responding well until engine heat loosens the internal lubricant. If, however, voltage to the valve is low or null during warmup, the problem is elsewhere. If it is the valve, clean it or replace it.

Steve

Mine had no voltage during start or warmup. I had bought a new valve but decided to check the current one before actually replacing it. Now I have a spare.

How did you test the cold start valve?

I was using a 10 kohm resistor (between the temperature sensor cable and ground) to emulate cold engine (minus 15 C approx (guess that is ca 0 F). Then I removed the cold start valve and refitted it in such a way that it sprayed out in the open air, then it was obvios that it was working... A fine spray of gasoline in the air (yes I know this i very dangerous....)

The voltage changes depenging on engine temperature and the idle rmp seems constant... if I, however, remove the power from the engine (at cold temperature) it lowers the rpm and at warm engine temperature the rpm increases.... So it SEEMS to work fine, but you never know, right?? ;-) ... But you guys think that cleaning the idle contorl valve will fix this? That might be a good idea... I will try it. What kind of lubricant do you need to put in there after cleaning???

*lol* Funny, I had that idea myself ;-)
Actually I was thinking about doing that, however, I never came to actually doing it, since my "test" of the cold start valve showed me that it was working.
On the other hand, there might be an idea doing such an arrangement, since the problem seems to come and go. Therefore one might think that the problen IS the cold start valve not being actuated due to bad connectrions somewhere....

I recently went thru the same scenario, on a 1993 300E. Would require 2 cranks to get her going when temp. below 40. OVP was replaced, & fuel pressure was good internally, everything else perfect. Donnie from MB Autowerks got a second look at it, overnight, went thru it, and traced it to the battery . I said no way, it is good. Was about 2 years old, tested fine...until under load. Then it would crap out. Dropped a new 850 CCA Interstate in, and all problems went away.

Simple.
Good luck
Q

No, probably not the cause based on your tests. Could be an air leak, though. Sounds like the ECU is trying to reduce airflow when warm, so it might be coming through elsewhere. Probably unmetered, too, which would lean out the mixture - a real problem when cold.

Also, Q-ball's idea about the battery is worth investigating. If voltage drops too low while cranking, it could upset initial conditions set by the ECU.

Steve

I have been searching for an air leak, but I haven't found any (using start gas), but this was never done properly - I will test it again.

The battery is a 88Ah battery (made for the diesel engine cars)and its only one year old so I think that it is almost over powered and ok (it sounds like the starter engine runs at high power), but I can of cource measure the voltage ot the battery while cranking... I've had that idea myself, but never tested it though - or actually I never really believed that that was the problem... but I will try.


Thanks guys! :-) :-) At least I have som ideas now :-)

What resistance do you get across the connectors of the start valve? The spec is 10-25 ohms.

Also, what engine model do you have - digits 4-6 in your engine serial number - 103.xxx? The CD shows at least two modesl for Sweden.

Do you have EGR?

A catalytic converter?

An O2 sensor?

Yes, I have lambda sensor and catlytic converter, but not any EGR-valve... after the weekend i will have the six digit number of the engine... Ok? (also, my car is originally sold in and importef a few years later from Germany)... I will check the resistance...

The engine is called 103.983 and the start valve resistance vas within specs.

I think you should run some basic checks on your warm-up enrichment by measuring the current at the EHA. There is special harness that you can use to do this, although you may be able to rig something up.

Here is what the CD says you should be seeing for a KAT engine as of 09-1987:

KOEO: 20 mA

20 seconds after start: 4 - 8 mA

60 seconds after start: 0 to -4 mA. At +20 degrees, the coolant sensor's resistance should be 2.3 - 2.8 kOhms

Temp at +80, fluctuating currrent reading, 230 - 290 Ohms resistance from sensor

My guess at this point is that your coolant temp sensor may be bad and is telling the brain not to enrich.

Good luck. Many posts on measuring EHA current.

Ok, thanks! :-)
Actually I think that the reistance of the coolant temp sensor is within that range, but I can always try again.
What do KOEO mean? Is it during startup?
I guess all of the current mesurments are is done when engine is started up from cold...

Thanks again!

Key On, Engine Off. It gives you the base current for the EHA.

Aye, then I know....
I will try tonight! :-)
Thanks!

I tested the current at the EHA and it was 19.9, engine off ignition on, about 7 mA directly after startup. 20 seconds later it was approx 4. When engine was warm it was fluctiating at about -3.5 mA (between -3.1 and -3.8 or something like that) (indicating that the lambda sensor was compensating for a rich mixture, right?) and I left it there since a little rich mixture would make cold starts easier according to another thread at this forum.

Also last night (a few degrees below freezing point) the engine was easy to start.

This morning however, it was in the garage (temperature about 5 Celcius (ca 40 F if I make a guess)) and this time it took a few second of cranking, then the idle rpm slowly rose to the warm up idle at approx 950 rpm....

I will exchange the fuel filter, I have bought a new one. But might the OVP relay or the fuel pump relay have anything to do with the strange behaviour at just over and at freezing point temperature.

By the way.... I did also connect a button (installed under the steering weel, hardly visable!) to a relay that I connected to the the cold start valve, will try to use it the next time the car doesn't start easily (to test if this is the problem).

Thanks! :-)

So now I am back to thinking the problem is in the power to the cold start valve.

From the wiring diagram - the coolant temp sensor sends resistance reading to the CIS/E, which relays it to the fuel pump relay. Fuel pump relay decides to power start valve.

So, possible problems:

- flakely reading from coolant temp sensor tells fuel pump relay not to send power to start valve

- bad wire from CIS/E control unit to fuel pump relay (tells fuel pump relay coolant temp sensor reading). Green/Blue from Pin 9 of CIS/E to Pin 2 of fuel pump relay.

- fuel pump relay bad (cold start valve circuit)

- bad wire from fuel pump relay to start valve

- bad start valve ground

I think you need to have the test equipment in place to measure voltage to the start valve when the problem occurs. Based on what you have already tested, the fuel pump relay is my prime suspect now, with the temp sensor in second place.

I agree with you, although the two last suggestions are probably not the right ones. The resistance between battery minus and the ground connection the valve is 0.3 ohms, the resistance between fuel pump relay and positive connection at valve is also 0.3 ohms. Not perfect connection, but not source of a very large "voltage loss" at load either.

I mostly agree with the suggestions that its about bad wirering or bad fuel pump relay or something like that, since the problem seems to "come and go". The cold start valve is working (i can hear it work so to say if i shortcut it (engine off, using the accumulated fuel pressure) or use my newly installed start valve button button I can hear a "click" from the valve and also the sound of gasoline being sprayed (if you understand what I mean). The wire from the computer (control unit is perhapps a better word for it) to the relay is good (?), I have tested it before (if its not a loos connection making it work sometimes and sometimes not).

The resistance between the temp sensor and ground is, if I remember this right, ca 4-5 kohms at 0 C, freezing point, (but Im not very sure about this value, haven't checked this for weeks) and about 180-190 ohms at working temp (87 C) (the latter is a little to low, I know that). Might this small deviation be the cause of all of this???
When I connected a 10 kohm resistance instead of the temp sensor the valve sprayed for several seconds (tied this out to open air, just to test if it was a fine spray or not).

Tried to disconnect the temp sensor, and then the engine was impossible (almost) to start if warm, probably due to flooding of fuel due to open cold start valve, and starts only with gas pedal to the floor... Then I got an idea!... Since the engine also sometimes starts easier with pedal to floor when i try to cold start it another question arises... Is it probable that the computer tells the relay to "over use" the cold start valve (using it to long time) thereby flooding the engine with gasoline (that is, the problem is to much gasoline instead of to little)??? And that THIS is my cold start problem... (I hope not)...

Many thanks to the answers so far, I have got several ideas about what to look for from you...

Last night I left the car outside just to test wether it would be hard to start or not. This morning it was very hard to start, took four tries including one where I used my button for 2½ seconds (0 degrees C) - guess its not because the cold start valve is not actuated then... this time I know that it was actuated since I did it myself....
Strange...
Eventually I floored the gas pedal and then after a few seconds of cranking it started...

/Ola

btw: i have a new ovp relay waiting for me at the post office now so i will fetch it after work and hook it up... Most probably i will need another fuel pump relay ta, I will try to get my hands onto one of those to... however they are quite expensive new so I will try to find a good used one...

Were you able to get a reading on the voltage to the start valve when it didn't want to start?

Maybe the valve itself is flakey?

BTW - my friend Jim Freeh, who bought my 300E last summer, is having this exact problem, and he has not been able to figure it out either. He has an "extra" 300E and he has swapped all the components in the loop withouth success in isolating the problem.

Mine sounded very simliar...



Now in the last few months- it starts and then dies cold or hot. I cannot be sure-but I think I smell fuel on occasions.

I was leaning toward testing/replacement of the FPR?

I can by pass the fuel pump relay and try a cold start. temp sensors are in spec. Seems to start fine when it's below 0 C.

It's better after an airflow sensor plate adjustment.


Michael

I will try to do such an adjustment... Thanks for the tip! :)
By the way, what is an airflow senor plate adjustment?? It not the CO-screw you mean, right??

There is specifications for the position(centering of the airflow plate) in the venturi. It must be ccncentric with the bore and then must have the zero adjustment.

it's procedure 07.3-1665 in the manual.

Michael

Never measured that for som time now, but it looked like it´s withins specs, but I will be forced to check that again....

Is there a check list of things to verify(?),

I feel like I'm arbitrarly testing components.

Can someone tell me what the test proceedure is to tell if it's a fuel pressure regulator vs EHA?



Thanks,


Michael

High again, new questions bringing this thread back to the top...

I did replace the fuel filter the other day, and now I have a problem with "walking" idle... sometimes it's normal and suddenly it can go up to 1200-1500 rmp or so... being there for som time... I did a new CO adjustment using the EHA current and it got better, however still wanting to be high from time to time... I have just tried a few times)... When I changed the filter the car was washed as well, might it be a problem realating the new filter (better fuel pressure) or might it be maiost disturbing the engine?
Any ideas anybody ;-)

BTW: I have only cold started the car once since then, and then it started just fine, although the idle was to high....