Engine stall after brief idle

[hnyc]

Hi everyone.

I see this is similar to another somewhat recent post on the board but I didn't want to take away from that car's attention.

After a brief hard start (one second or so more than normal) the car will have a lumpy idle which is verified by the pulsing exhaust puffs coming from the back. A few minutes later the car will unceremoniously shut off. No advance warning of it coming and then it shuts off. Adding throttle will bring the speed up smoothly and it doesn't have the same characteristic then as it does when left alone.
When you go and start it again, it will but this time it will shut off sooner and after a while it just will not start after a few cranks.

This car had a lot of work done to it if some of you remember from last year and I did give it a test drive before this started happening. During the test drive I noticed a small loss of power but something that seemed not too serious.

So that's the symptom(s).

Work done that would concern this (not all in conjunction with this problem but this was an involved refresh):
New ignition (from the dust covers/suppressors to the plugs )including them) + coils).
Wiring harnesses (upper and lower)
Rebuilt ETA
New injector seals and screens.
Vacuum lines (all)

What was done for the stall problem.
CKP checks ok with resistance
Fuel pressure
Ignition and fuel modules swapped

The codes (No check engine light):
1) TNA signal incorrect or open or short circuit
2) Intake air temperature sensor (This is me as I disconnected the intake tubes after this started happening)
3) Injectors open or short circuit
4) CAN data line to electronic accelerator/cc/isc module
5) Closed throttle position contact switch
6) Tn-signal is outside the tolerance range
From the Diagnostics module port:
7) ISC (idle speed control) incorrect. Fault from EA/CC/ISC
8) Intake air temperature sensor signal (from above)
9) Engine speed signal TN (rpm) or Engine control module defective
10) Fuel injector circuit open or short or self adaptation at limit.

Keep in mind that these codes were cleared (the ones that couldn't be erased stayed) not more than a month or two ago (only a few starts since then)

I've looked up and down these threads and whatever the solution seemed to be, I had already worked on it. I resisted writing here but right now I'm stumped.

 

[Kyiv]

Just to cross out the easiest stuff: are all 4 10 amp fuses in the base module intact?

 

[robm.UK]

Codes (4), (5) and (7) suggest there is no communication between EA/CC/ISC and control module or a fault in EA/CC/ISC, or the wiring between the two.

EA is the electronic throttle
CC is cruise control.
ISC is idle speed control.

Diagnosis is in this link

I believe you could check the supply voltage to the control module by pulling the ECU and testing voltage across socket terminals according to first checks in the above link. This may provide useful insight to your problem, or at very least would rule out supply to ECU.

Condition appears to get worse as engine gets up to temperature? Suggesting problem with EA.

As far as I know further checks according to the link require break out connectors to enable checks of the potentiometer reference voltages etc. Unless someone here knows of a tip/workaround.

The potentiometer and idle switch is built into the electronic throttle actuator.

 

[robm.UK]

Further diagnosis charts are HERE

 

[gsxr]

I know this is a long shot, but based on the TN errors (which are VERY unusual), I'd replace the crank sensor (CKP) unless it's recent/new. A failing sensor can cause some really weird problems despite the resistance testing OK.

Also, have you swapped the E-GAS module yet, if you have a spare for testing?

 

[hnyc]

Thank you for your replies. I'd like to say that it didn't work (because it was so easy) but kiev's initial direction so far seems to be the correct one.
I had a blown base module fuse.
This definitely threw me as there was always something pointing me to ignition.
One of the coils were throwing out low voltage.
One of the connectors in the distributor cap was broken (non manufacturing defect or issue). The wires having been taped up and showing some insulation breaks told me to change them as they were the last common denominator.
This is strange as this car didn't exhibit this before the work was done so I thought it had to be engine/ignition related. On top of that with everything else that was changed it didn't make sense.
It's frustrating but a relief that it was (so far) a simple fix. It's still idling a little lumpy but it's not shutting off so I'm assuming that when I attach the airbox and tubes on with the sensor that that should even it out.

Thank you kiev (and Rob and Dave) for your help. I didn't even think of those fuses but honestly I'm not well enough informed on these cars' systems that I would've been able to figure it out without the technical experience/knowledge present here.

 

[gsxr]

Good to hear it may have just been a fuse issue, and good work @kiev!

I'm confused about the distributor cap though. Your first post said the ignition system was new from stem to stern. How old (time/miles) are the ignition parts?

I'm surprised the car would run at all with a blown fuse in the GM/BM...

 

[hnyc]

Not to reveal too much (I still want to write up about it) but the cap was impact damaged while being out of the car.
One of the terminals was cracked and showing weak on a spark test which is how I caught it Visually externally it seemed fine.
The ignition parts are new as of this summer to yesterday (didn't replace everything at once) and the car hasn't travelled for more than a half mile since I test drove it after all the work was done (this summer)

 

[Kyiv]

This is strange as this car didn't exhibit this before the work w

You likely blew the fuse when reading/resetting codes. That's what happened to me, using a blink reader, I did not insert its wires in a proper sequence/or into a correct hole, causing a short. I can so feel your hope igniting when you saw that a fuse was indeed blown, followed by euphoric relief when the engine started and kept on working

 

[xfadmin]

For outstanding and meritorious services rendered to the 500E community, @kiev is hereby awarded Герой Советского Союза.


This gives him special forum privileges for the rest of his life.

 

[hnyc]

Well you're right. I remember crossing the pins (I used paper clips to connect) slightly when I was reading it and it set off the smallest of sparks but didn't think anything of it.
The fly in the ointment in all of this though is that I only started reading codes when this trouble started. Blowing that fuse would have come after the stalling issue as I was trying to get some answers.
So I might have flown high just to come right back to where I started.
Doesn't matter. First time since summer that I shut the car off actively rather than passively.

Dave I realize that I hadn't unambiguously answered your question. I replaced the plugs over the summer during the engine redo. The caps had age but not miles on them so I left them alone. The stalling started and through troubleshooting showed low spark. The wires tested fine so I figured new caps and everything behind them would be next up to remedy that. By that time though I had spent way more money on the car than I had expected to so I gave myself a moratorium on no more until the new year. As you can imagine, no one was happier than me when Jan 1 came around. That's when the new caps rotors and shields went in. I got my improved spark but still had stalling. Even though they tested fine I figured the wires must be the culprit as aesthetically they were nothing to beam about. Those came in yesterday and after installing them today the car still didn't want to do what I wanted it to which is what brought me here.

 

[robm.UK]

Kiev’s instincts are impressive.

Glad you found the problem.

Following my advice, by checking supply voltage to ECU (this was my hunch), you would have found 0V, which would have then led to checking the base module, and of course same result.

Incidentally, according to WIS, a faulty CKS would elicit limp home mode.

 

[Kyiv]

For outstanding and meritorious services rendered to the 500E community, @kiev is hereby awarded Герой Советского Союза.
View attachment 123197

This gives him special forum privileges for the rest of his life.

I'm incredibly honored to receive such a high award. Showing up without a prior appointment to a 500eboard HQ, opening a door without knocking with a leg kick, and proceeding inside with a lighted cigarette in my teeth, straight towards Gerry's megabucks mini bar, is just ONE of the ways I plan to take advantage of my new status.Though it'll be a while before I can march in the same column with the likes of his, GSXR, JC220 and others

 

[gsxr]

Incidentally, according to WIS, a faulty CKS would elicit limp home mode.

Rob, which sensor is "CKS"? Do you have the WIS reference handy? I'd like to take a peek at that.

 

[gerryvz]

J

Rob, which sensor is "CKS"? Do you have the WIS reference handy? I'd like to take a peek at that.

I think he means "crankshaft position" sensor (CKP)?

 

[gsxr]

I think he means "crankshaft position" sensor (CKP)?

That's what I thought, but I have never once heard of a crank sensor causing limp mode on LH-SFI systems, which is why I wanted to check out the WIS document.

 

[robm.UK]

Rob, which sensor is "CKS"? Do you have the WIS reference handy? I'd like to take a peek at that.

Dave, your force is strong!

Not a sensor. I meant to type CPS, not CKS (latter would lead to no-start). Closed throttle position switch S29/3.

HERE is link to STARTEKINFO diagnostic chart

‘EA in limp-home mode’. Possible cause ‘closed throttle position switch S29/3’.

 

[gsxr]

AH, got it! Yes, S29/3 will trigger limp mode. I don't see that often, but it does pop up occasionally. This should trigger a recurring code, and also requires the throttle linkage to be adjusted properly. IIRC even with a new S29/3 switch you could get a code (and limp mode) with the linkage mis-adjusted.

Note that there are different fault codes for closed-throttle position switch, and @hnyc didn't specify. One is for S29/3, the other is inside the ETA itself, assuming a car with ASR.

 

[hnyc]

A 24 hour update.
I seem to have had a couple of small issues that worked in concert to help throw me around.
The car stays on though. That is tremendous.
Idle though is still lumpy enough to send the vibrations through the cabin (new mounts so it's not that)
It improves as the car warms but it's still there.
My last repair should include new injectors. I cleaned them up when they were out of the car and either I didn't do a good job on one or two of them and dislodged some sediment (didn't test the spray before re-installing) or they're just faulty.
A rebuilt set is 'cheap' enough and given the car's miles and age doesn't make it seem like an unnecessary purchase/replacement.

Dave I'll assume that my throttle position switch has more to do with the linkage than the eta as that was rebuilt. It is an ASR btw. I'll look into that as there is still some tuning left to do.

 

[Kyiv]

Idle though is still lumpy enough to send the vibrations through the cabin

For what it's worth, I've had my 400E for 6,5 years and put on 39k miles on it (currently it has 137k). My cold idle has never been "perfect". My idle RPM generally don't hunt, but once in a blue moon it will, just a little. There has always been a slight, soft miss, when the car is cold, which goes away by the time I stop at a first intersection (45 seconds from departure). With all of the things that I have replaced - caps and rotors (several times), insulators, both coils, MAF and its rubber boot, LH module, vacuum lines, FPR and fuel pumps, spark plug wires, spark plugs (twice), cleaning throttle body, etc, etc, - (subjectively) nothing has helped more than driving my car more often.

Over the years, browsing the internet, I saw many people experiencing the same, without a definitive resolution. If, once you start driving your car, your gas mileage is what it should be, then that's an indication that your engine is working 99%+ correctly, which means you should not be stressing out too much about a somewhat lumpy cold idle.

This is what my engine sounds like when it has warmed up some. I think it sounds fine. Inside the car, if I pay attention, sometimes I can sense an ever-so-slight tremor. When I just start the car, I can see a slight, soft shake from the miss

 

[gerryvz]

That's an awfully yellow expansion tank......

 

[hnyc]

If my car idled like that I'd be plenty happy.
This isn't a princess and the pea type of situation.
On my car you can see the zoom (am I allowed to call them that?) tubes shake on idle.
Your idle is what I'm shooting for.

 

[robm.UK]

LH-SFI makes for a smoother idle. You should hear my CIS-e CFI 500 SL.

BTW your HT leads have a unconventional routing.

 

[robm.UK]

I would add that the idle on my KE CIS-e CFI is as good as it ever could be, but the KE has a characteristic rhythm due to the way the N3 KE ECU controlled system feathers the mechanically based fuelling.

I am fond of the KE Jetronic II, because it is primarily a mechanically based system, and if the N3 ECU or any sensors/actuators fail, the engine will keep running in a more or less normal fashion, but without closed loop functionality or cold start/full load enrichment. I believe fuel cut off on over-run would still operate, since it is controlled by the MAS. An ECU I cannot afford to have going south on me, so I carry a spare. It sits in the spare MAS holder in the ECU compartment right next to the operating MAS. I assume the spare holder was there ready for the V12 600 SL. Two of everything. The Germans were masters of planning, sometimes . I digress...

 

[Kyiv]

That's an awfully yellow expansion tank......

I have long noticed that a yellow expansion tank disrupts your aesthetical view of the world as much as URO parts It's actually on my next to do list.

Is it known how it usually fails? Does it explode, or just develops miniature crack/s leading to slow coolant and cooling system pressure loss?

On my car you can see the zoom (am I allowed to call them that?) tubes shake on idle

Have you gotten a chance already to drive your car for an hour or more? What is the hot idle like?

 

[gsxr]

The plastic expansion tank failure is usually a small leak somewhere, generally at the top metal neck where it's pressed into the plastic. If you don't replace it, at least replace the level sending unit. If old or original, those are usually shot. Relatively cheap insurance, as you'll get a warning light on the dash if the coolant level gets abnormally low.

 

[hnyc]

Dave is being kind.
What really happens when you drive with a yellowed tank is a result of Gerry's extensive dealer/repair/refilling network associations. Should you ever be asked to open your hood if you need a repair, a part at the dealer, or even for full service at the gas station and they see a yellow tank, you are initially refused service. A scarlet letter then gets stenciled onto your car on the driver's door and that's when the real fun begins. Food thrown at you in town squares, preferential treatment towards 15 year old Kias over your car at valet serviced restaurants, parking lots, etc. even when you got there first. It's really bad. Which if you're in the company of your beloved elicits a typical "Would it have killed you to change that tank? You knew this was going to happen. Next time we take the minivan."
I mean sure, the tank can crack and leak coolant, but that's really the least of your troubles.

I haven't run it for an hour. Maybe a half hour after the base module fuse replacement. The exhaust smells a bit and I don't know if I'm pushing gasoline through the converters. This car is close but there's something just off about it. I'm guessing injectors but I need to do more than guess.

 

[gerryvz]

It's not very common at all for the injectors on these cars to go bad.

 

[gsxr]

I haven't run it for an hour. Maybe a half hour after the base module fuse replacement. The exhaust smells a bit and I don't know if I'm pushing gasoline through the converters. This car is close but there's something just off about it. I'm guessing injectors but I need to do more than guess.

What are the current symptoms, now that it is running again? Have you driven it for an hour or so to get everything fully up to temp?

 

[hnyc]

I know but injectors are the only remaining thing that makes sense. Even if it doesn't make sense. And I'm thinking that maybe when I replaced the injector screen that a small amount of mesh or plastic separated from the part and is still in the feed well. Highly unlikely but is it impossible? I'm good on ignition. Good on fuel pressure. What is left?
The car is a salvage which means here it needs to be inspected to verify that the repairs were done with non stolen parts. The funny thing is that I haven't replaced a single body part on this car.

I think I might have to bring it there as is. It runs and they don't care about running rough I don't think. I'll have it passed then insured and all that so I can drive it on the road to then get to the bottom of it.
Maybe it just needs an Italian tune-up.

Current symptoms are the still running rough idle.

I'm going to go out and take a look at it a little later. If I can, I'll take a video and post to show you.

 

[gsxr]

Is it only rough at idle, but smooths out and has normal power at higher RPM's while driving?

Live data from SDS or other digital scanner (or, oscilloscope) would show if it's an ignition misfire or not.

 

[hnyc]

It's the strangest of things.
I've replaced virtually all known ignition consumables. Except for the wires, each of them were replaced due to time and miles and still had the same miss. I replace the wires and these are incremental improvements.
So while I know the car is better off than it has been due to it, there is something wrong that these parts being installed has not cured.
In addition to the miss on idle, it shows up as well when throttle is applied.
I might take it out tonight for a ride around the block or two.
Before that though I will clear all codes and see what comes back.
The unwanted shut-off fix is sure to have been the cause of a few of them. We'll see, and I'll report back.
But I'm thinking that a computer hook up will be in order.

 

[Kyiv]

Open up your LH module. See if any of the capacitors have leaked

 

[frall]

Open up your LH module. See if any of the capacitors have leaked

This is a good advice. The capacitors are old enough to replace.

Another possibility for unstable idle is the Starter lock-out/backup lamp switch (S16/3). You can make a measurement as shown at Test Step 12 in this document: STAR TekInfo.

Did you already check the fuel pressure?

 

[hnyc]

The LH module was swapped as I have an extra one. Same result.
The startek test I'm more curious about and are going to try. I am embarrassed to say this but as I pulled the engine and transmission out last year, one of the wiring connections was snagged and I didn't notice it. I wasn't feeling well enough to check under the car during the extraction and rushed it. When I put everything back together and the transmission was working, I assumed it was ok.
I'm going to check this next.

 

[frall]

And the fuel pumps/valves and pressure is ok?

 

[hnyc]

Oh, yes, sorry.
Among my initial checks was fuel pressure. Both measurements are within specification.
The valves if you mean combustion I'll assume are good. Not giving issue before this so I'll 'assume' they're good. This seems electrical more than mechanical so I'll pass on an actual reading of those for now, although they probably should be checked in the next round.

 

[frall]

Did you check for wrong air (Falschluft in german)? I would inspect the pipes in the attached picture.

 

[hnyc]

Those are all new and attached as depicted.

Is it possible to do the switch/wiring test without a socket box? I'm confused as to the connections to be made since I don't have one. I pulled out the egas module thinking I could just pin the connectors underneath but the test says/shows different.
I have a multimeter for readings. No Mercedes specific readers.

 

[frall]

Yes, you can messure the resistance when the module is out of the box directly at the pins.

 

[gsxr]

Note you can only measure certain items at the sockets with the E-GAS module removed. Some cannot be measured without the breakout box, i.e. voltages from the ETA for example.

This is where the SDS live data is useful... plus you can view data with the engine running and, hopefully, when it's acting up.

 

[hnyc]

I just did a measurement.
Park and Reverse are within the test result window but the rest of the gears are showing as an open (no resistance) circuit.
I'm going to check again tomorrow as it's strange that 4 settings are showing open.

Dave that engine does not stop acting up. This isn't an issue that I will have to 'look' for.

 

[frall]

Have a look to the switch. You can open it by drilling out the rivets.

 

[hnyc]

Unfortunately the wall of winter is upon me.
There's some snow on the ground and cold enough weather that this car being outdoors activates my 'too cold to lay on your back for a recreational vehicle' rule. The next 10 days don't seem to be looking any better either.
I will pause this for a bit and return with results when it's more pleasurable to do so.
Thank you for the assistance. I appreciate it and will report back in time.

 

[hnyc]

Well the weather warmed up a little and I was still very much wanting to get this car going properly.
The NSS wound up being functional and giving the right values for the right gears.
I purchased an Ebay SDS reader because if you hear 'live data' being mentioned enough times, you think that that's where you should logically go next. Yes Dave I'm talking about you

After a small mis-start with it (thanks to Gerry for clearing it up) I poked around and got a message that would indicate that something was off.
Under the LH1 and DM groups they both mentioned the fuel injection valves as being an issue. I checked the connections, they all seemed to be fine. I didn't want to unplug them all to use a meter to see if they were getting current so I used a screwdriver to listen for actuation. They all checked fine except for Number 6. I checked again and yes they all sounded the same except for that one. This is a new harness but the injector is old so of course I'm thinking that the injector needs to be replaced. Of course I'm going to test it but before that, what if I jiggle the connector around? Maybe that might jar something and I wouldn't have to do anything.

So I did, and although it seemed tight the first time I checked, this time was different.
The connector wasn't seated properly. So I properly connected it and wondered if it would be at all beneficial. Since the car was running during all this it didn't take long to find out.

This annoyance that was plaguing an otherwise enjoyable (mostly) rebuild was from one wire not being properly connected.
And how did this happen? I'm not careless enough to not click in an injector connector. But I am particular about hose and wire routings. And when I see one that has to be redone because that's not how it was originally, I remove the injector connector to facilitate its being redone.
I slip it back in without clicking it because what if I don't like the routing this way and want to redo it again? Of course I'll remember to properly connect it once I'm satisfied with the hose/wire routing. Of course I will.
Except I forgot.

So I had a rough running engine in three parts. I had:

1. a loss of current to ignition which gave a rough running no/low spark to one cylinder
2. a blown base module fuse which would shut if off after a brief idle
3. and lastly, a non firing injector which was contributing to the last of the poor idle and operation.

Sometimes you think the fix is in, but sometimes, it's just a coincidence of factors which throws things off.

Thanks again to all of you that offered help and guidance towards this elusive issue.

 

[sheward]

Happens to the best of us. Glad you worked it out.

drew

 

[gerryvz]

Nice detective work. Glad you got it all sorted out !!