Loss of power - restart fixes it.
 

Since my car did this three times already this morning, I figured I'd start a new thread and see if I can get any input.

1991 124 will suddenly lose power just driving along. It even does it sometimes just sitting at a light. It will suddenly start to idle rough, the idle speed will drop under 500 and when you try to move, the power is gone, like you've got five ACs running. You can tell when it happens while in motion because the car starts to feel really heavy, the slow idle makes it slow down more quickly and when you stop, the tach is showing less than 500, rather than the usual 700 or so.

For a while now, the car would lose power sometimes but you could "fix" it by flooring the accelerator for just an instant and it would be cured. This new (?) problem feels different somehow and the floor-it trick does not work. But turn off the key, restart and poof! It's back to normal. For a few minutes, anyway.

What do ya'll reckon this could be?

Sounds like the beginning of a bad OVP. At least that's what happened to me.

Those symptoms happened to me. In my case it was a flakey oxygen sensor (nobody ever believes this.)
I diagnosed it by pulling the connector wire under the passenger carpet while stalling on the interstate... (not a method I would recomend)

One more note: A restart takes the computer out of loop for a few seconds when warm... so any 'leaning' fault in the mix might be temporarly overridden or be forced to find another intermitent window.

have you replaced the cranksensor pickup at the flywheel thru top of bellhousing? they are known to operate intermittently when going bad.

it sounds like a bad coil to me.

One More Thought
 

My '88 260 hesitated badly and was finally cured with a new EHA.

In my case the car ran fine when first started cold, would hesitate as it warmed up, and ran okay but had a rough idle when warm. Your case sounds a bit different - more like an intermittent hesitation.

One sure symptom of a bad EHA is leaking, there will be a slight gas smell under the hood after driving.

Today I replaced the O2 sensor and the OVP with spares that I had. You accumulate this stuff when you own two of these cars for years.

The coil and EHA have been replaced within the last year.

I also washed the car today...that always seems to make them run better.

We'll see what happens tomorrow.

Thanks for all the input!

Oh, well. No change at all.

Any other ideas?

Thanks

How often does it die? Might be time to hook fuel pressure gages up to see what's happening in the distributor when it stalls...

Oh one last thing before that... un plug the EHA (you probably know that that's the black box on the back of the distributor under tha air cleaner) and see if it runs without stalling... that would narrow it down to a mix problem if it continues to stall.

Well, it never has actually stalled. It loses power overall and the idle speed drops from the normal ~700 to ~400. If I happen to be at a light I can shut it down, start it right back up and it's normal again.

Mass Air Flow - maybe? Or the aged wiring to/from

I wonder if the Arthur Dalton Radio Shack code reader would reveal anything?

I'm really wishing for OBDII about right now.

My 560SEL was doing right about the same thing..no power and a re-start would fix it for a little while.. Turned out to be the KE control unit...it's open loop/regulated had lost it's mind, no longer able to provide worthwhile info to the EHA etc.. Re-starting kicked it into closed loop/unregulated and it could run OK that way.

Finding a 300E KE shouldn't be hard @ all..know anyone in the area w/ the same car? You can make the swap in a matter of minutes.

Jonathan

Go ahead and read the codes using Arthur's code reading tool. That's allways a good idea.

http://www.peachparts.com/shopforum/tech-help/107630-e320-16-pin-dtc-charts-inc-pin-8-pin-14-a.html?highlight=CODE+READER
Post #8 links to a PDF that list the codes for different model/years.

How about the coolant temp sensor ? This one sends signals to both the ICU and ECU and an intermittent disconnection of this sensor would put the car in a limp mode and restart would remove it from that mode until it happens again. So check its wiring to both the ECU and the ICU, make sure there are no loose connections or cuts in the harness. If that is fine, I would replace that sensor if it has not been replaced recently. Intermittent CPS operation is also a good possibility, but I would assume that the car would most likely stall without the CPS signal. Also hoping that there are no vaccum leaks, i.e. unmetered air entering the intake, that could cause the oxygen signal to reach limits and I am not sure what the ECU would do if it always gets a constant signal. But this you can test and see if indeed the oxygen signal is turning into a DC signal.

Good idea. Which of the sensors are you speaking of in particular?

I am not sure about your car but on my 87 260E, it is the last one on the engine, towards the windshield. It is a two terminal sensor. When the car is cold, the sensor's resistance should be between 2.5 and 3.5K and when the car warms up, at 80 degC, the resistance should drop to between 250 and 350 ohms. The exact values could be different but in general the resistance should drop by a factor of 8 to 10 when the temp changes from 20 to 80 degC.

brewtoo,
If you pull the EHA plug and drive it you won't have any computer input into the fuel mi_ this includes the computer itself, the o2 sensor the coolant temp sensor... you wont have cold start assistance or emmission tuning but the car should run perfectly well this time of year without it. If it does not stall with the EHA unplugged then you know the problem is mi_ related, if it does stall with the EHA unplugged then you must look elsewhere.

(Bad ecs key on this key board thus fule mi_)

Ohm's range from the temp chart in the /Engine Manual-Combustion Manual-07.3 Mechanical/electronic gasoline injection system (KE injection)-121 Testing electrical components of KE injection system. page 60

Temperature °C........Resistance k?
-20........................15.7
-10.........................9.2
0............................5.9
10...........................3.7
20...........................2.5
30...........................1.7
40...........................1.18
50...........................0.84
60...........................0.60
70...........................0.435
80...........................0.325
90...........................0.247
Same resistance values apply to KE intake air temp sensor.
I think this chart applies to these 2 sensors for all KE systems but don't know for sure.

Thanks for all the input guys.

I have to pause this project now and put in a new water pump. And belt tensioner. Doesn't it just figure? The joys of these old cars!

Looks like Mr. Patel may be the winner!

Preliminary tests with the Arthur Dalton Honorary Code Reader reveal:

CIS CODE 31 - Intermittent contact in engine coolant temperature circuit.

I wonder if it is the harness, a connector or the sensor?

Is it the 4-pin sensor all the way in the back of the head? Or the single-pin in front of that? Or the 2-pin sensor in front of that?

Thanks

OK, I have swapped that temp sensor between my two cars and deduced that the sensor is not the problem.

Could this be a ground connection problem?

Is there a document anywhere that shows all the geound connections thoughout the car? Seems like I've seen one somewhere.

The fact is though, that 4-pin connector does not use a ground connection and the temp sensor is isolated...really two identical temp sensors in one.

So, all things considered, it must be a wiring harness issue.

Agree?

Wiring harness is possible, one set of cables should go to the ECU and the other set to the ICU, check continuity. There was one person on the forum who tried everything and it turned out to be that the coolant sensor's wire were cut somewhere due to rubbing against a metal part.

My factory manuals only go thru 1987. Of course, by 1991 they had changed the temp sensor from a 2-pole to a 4-pole.

Where might I find the proper wiring diagram?

Thanks

Brewtoo, I have some pictures in my CD, but the files are large to post, send me an email address. Does your car have the mechanical fuel injection system, or the hot wire/hot film system ? What is the exact engine/model number ?

OK, one source says code 31 is the coolant temp circuit. Another source says code 31 is the outside air temp circuit.

Can anyone verify which is correct?

Heads up for anyone who may go down this road...

It would seem there is an error in this document: http://www.geocities.com/MotorCity/Downs/1039/mercedes/cs1000_mb.pdf which was referenced earlier. It is a big 169 page pdf file listing error codes...it's actually a manual that came with a code reader.

I have a CODE 31 which that document says is an intermittent coolant temp sensor circuit.

But now I notice that it says that both error code 28 AND 31 are a coolant sensor circuit problem. Obviously, that cannot be true.

The factory documentation says code 28 is a coolant sensor circuit problem, and 31 is an intake air temperature sensor circuit problem. However, there is plenty of confusion supplied in the factory documentation, too, because in the repair section they twice refer to air temp sensor as the coolant temp sensor, even though they call it by the correct part number.

So I've been chasing a coolant temp sensor circuit problem when, evidently, my problem is with the air intake temp sensor circuit.

Update:

I checked the lambda and the EHA current. It was very odd. Once it went into closed loop, the EHA current and the duty cycle would slowly climb straight up. The current would go over 20 and the duty cycle would seem to give up and fall to 0% (not 100%, nothing). I could the move the throttle briefly and it would repeat the same process every time. The 02 sensor was sending .7 to .8 volts all the time.

Since I've accumulated various parts over the years, I tried a different CIS computer, 02 sensor, EHA, coolant sensor and intake air sensor. None of these made any difference.

So, as I was about to give up for the day, I looked down at the meters and all was normal! So I adjusted the duty cycle to ~50%. I've driven a couple of hundred miles since and everything is perfect. Not just good, but perfect! No codes, duty cycle is on the money and it runs great.

I mean, it's great that it's fixed. But the problem is it just started working and I have no idea why.

:confused:

Correct me if I am mistaken but it sounds like the O2 sensor output was suspect. I think it should be sending above and below .5 v intermitently if the mix is correct. If this comes around again you might want to try a third O2 sensor to see if that solves it. Glad she's working for the moment. And remember if this should reoccur to pull the O2 sensor wire under the passenger carpet for an easy diagnosis. :D

Yes, I too would like to know the "proper" 02 sensor output.

Anyone?

It fluctuates between 0.1 and 0.9V, it should definitely be a changing voltage. If you get a fixed DC voltage, it is not working OR the computer just cant adjust the air/fuel ratio anymore, you then have to make mechanical adjustments so that you bring the electronic adjustments within the range of operation.