Car shuts off

[Benz Dr.]

Cold start valve only comes into play at startup.  Not the culprit.

I agree, or at least doubt that it's at fault in this case. More likely to be the fuel tank or lines going from or back through the tank.

[Shvegel]

The easiest thing is to check for spark and rule out the ignition system entirely.  My general rule of thumb is if it seems like you turned the key off it is ignition related and if it feels like you ran out of gas it is fuel related.  You could very well be dealing with a cracked solder joint in the transistor module that opens up when it gets warm and closes when it cools down.  I see this all the time in my work life on boats.  I just fixed a module last month that caused the entire electrical grid to drop underway(no steering).  I have not opened one up but I am certain there is a big transistor in there that has grown and shrunk a whole bunch of times stressing the solder joints.

In short check for spark first.  If you have spark then put a fuel pressure gauge on it and drive it till it drops and see if you have fuel pressure with the key on.  No? see if you can hear the wildly expensive electric pump? If yes, something is blocked.  if not, you have an electrical issue that is hopefully not related to the $1700 electric fuel pump.

[BaronYoungman]

Hi , I do believe it is a blocked line. I know it is a hassle but you could test by filling gas tank all the way up so gas comes over the top of the flowerpot and see if it stops, if it does not then you know that your flower pot is not getting return gas to fill it up from the engine.  I am definitely NOT an expert in the fuel system, but I think I know a lot about the pumps and the pump will only overheat if it is not getting fresh gasoline to cool it.  The only way the pump would suddenly stop when it is already running is if it becomes grounded, I find this to be very rare.  Earthing (grounding) is a common fault which occurs when part of the winding becomes connected to the metal core of the armature.  More important and likely once the armature is overheated this can cause the solder connecting the armature coils to the commutator to melt. As the armature is spinning the solder is thrown from the armature, causing the connections between the armature coils and the commutator to be broken so then even if you fix the fuel flow issue , now the pump may not turn fast enough do to the bad armature to give you the proper flow.
Bob

[MikeSimon]

It is a little difficult to diagnose how the engine shuts off when you are driving. The car is rolling and the engine stops and you don't really know if it shuts off from lack of spark or lack of fuel. The car does not accept throttle or stalls and you put it in neutral and by that time the engine has stopped.
I am not that much of an expert with the W113 -actually none at all - but I have worked on cars and engines most of my adult life and the last 20 years extensively on motorcycles. What puzzles me, is, that the car starts back up after a waiting period. So the problem is most likely temperature related.
What electrical problem, other than the coil (already replaced) and the capacitor on the distributor (already replaced) would cause "no spark".
I don't remember anymore who told me some time ago, that the fuel pump under the tank would/could shut off when overheating. If it does not, I am not sure what would cause a fuel supply problem.
What other part/component of the engine operating/control system is temperature sensitive?

[Iconic]

I don't remember anymore who told me some time ago, that the fuel pump under the tank would/could shut off when overheating.

Mike,
It is not the electric fuel pump near the tank that overheats. It is the FIP, mechanical Fuel Injection Pump.
It "vapor locks" when the fuel gets hot. The flow of the electric pump should be 1 liter per 15 seconds. The car only "burns" a small fraction of that. The rest of the fuel is being pumped in a big circle cooling the FIP and running back to the tank.

1. Drive car till stalls, or even better, let it idle until stalls. It was about 10 minutes for me.
2. Pop open the hood/bonnet.
3. Slap on an inductive timing light to any spark plug wire.
4. Crank engine.
5. Look for flashing light on timing light. If not light, no spark. If light, then you have spark.

Now, disconnect return fuel line to tank at the fuel tank. Turn on electric fuel pump with car NOT running. Measure how much gas flows in 15 seconds. Check against spec (1 liter per 15 seconds).
If low, you found it and send me $200  . I don't know why I said that, I usually don’t kid around.
If not, more diagnosis is needed.
 

[BaronYoungman]

Thank you Mark, I should have specified, I am only talking about the fuel pump in the rear by the gas tank, I am first to say I KNOW NOTHING about the mechanical injection pump attached to the engine. Just trying to answer the questions I can. Bob

[Mike K]

Mike- It could be any of a number of issues, based on the input you've received from the Group so far.

I think it makes sense to start with the easy/obvious stuff first and follow a process of elimination, otherwise you're all over the place.
I had similar issues with my 280SL the day after I purchased it.

Based on advice received here I started off with the fuel tank. I'm not sure if you've done this already?
Unscrew the filter port on the underside of the fuel tank and drain the fuel into a large container and see what comes out.
If there's a whole lot of debris and rust, then you know which route to follow.....

The waiting time between restarting may have nothing to do with temperature, and possibly something to do with debris taking time to  settle in the fuel tank. That's what I had. This led me onto fitting a new tank, new fuel pump, blowing compressed air through the fuel lines, and new engine fuel filter = problem solved.

Failing that, if you haven't done so already, check ignition wires from the spark plugs to the distributor, you could have faulty cables, this has also happened to me and the problem got worse as the engine heated up.

If this is all OK then move onto the more complicated stuff....

Good luck and let us know how you get on.

Best,
Mike

[Pawel66]

I second what Mike is saying.

But I think the simplest possible test of fuel supply is: detach the return line from tank, put in into some pot, switch on the fuel pump and see how much fuel you get. You should get close o 1 liter per 15 seconds. If you do not - this is your issue. Then you check the components of the fuel system, starting from the other end - undo the hose from fuel tank to fuel pump - if you see fuel dripping rather than pouring out with a solid stream - you start from the tank, which is clogged.

The first test takes about half an hour, including lifting the car.

[MikeSimon]

I think this will be my course of action - checking fuel flow and supply from and back to the tank. I am planning to take the tank out anyway and have it redone as the car sat for 20 years and the tank is probably rusted pretty bad inside.
I will keep you all posted on what the outcome is, however, it will take a while until I get there as I do not have the time to work on the car every day.
Now I have to address the open motor first. Cleaning everything up, vapor blasting the intake manifold and valve cover, painting the exhaust manifolds, doing some jobs on the head and then putting this all together again.

[rgafitanu@gmail.com]

Hello, I have been living with this problem for 3 years. I followed the advice on this forum to find the cause. I would like to add from my experience:
- if by now the car is stalling consistently after 5-20 minutes of driving the fuel pump is full of rust particles clogging the intake screen. If you take the pump inlet connector out and if the screen has rust in it similar to the attached picture then you have your answer. Between the tank filter, the fuel pump screen and the engine bay fuel filter, the pump screen is the weakest link that gets clogged first causing stalling. Picture shows rust accumulated over 3 months.
- because I do not want to shell $800+ for a new tank I plumbed a fuel pressure gauge inside. When the pressure falls below 8 PSI it is time to change the screen.

[MikeSimon]

OK, so what "pump" are we talking about? The pump at the tank or the fuel injection pump? Where is that intake screen and how do you access it?

[mbzse]

.../...painting the exhaust manifolds.../...

Yes - an opportunity to paint them in the original off white finish 
See these postings:
https://www.sl113.org/forums/index.php?topic=16880.msg116511#msg116511
https://www.sl113.org/forums/index.php?topic=6139.msg38059#msg38059

[Pawel66]

OK, so what "pump" are we talking about? The pump at the tank or the fuel injection pump? Where is that intake screen and how do you access it?

We are talking fuel pump next to the fuel tank. Squeeze the fuel hose from tank to pump close to pump, so that fuel is not flowing when you disconnect it. Remove the clamp on fuel hose from tank to pump on the pump side. Get a relatively deep 27mm socket and undo the inlet fitting of the pump. You will remove the fitting with screen as you see on the picture two posts ago. If you do not have this screen in the fitting - place intermediate filter on the hose and look for a fitting with screen. It would be good if this intermittent filter has 10mm fittings as this is the inner diameter of the fuel hose. Most of the filters are 8mm, which is not enough, you'd have to add some hose pieces to make them fit.

Undoing this fitting with 27mm socket may be tough. I used an air impact gun.

You will need a 27mm aluminum sealing ring, do not use again the one that was there.

But do the flow tests first before you undo anything.

[BaronYoungman]

When you do the flow test on the fuel pump (the one by the tank), you have to let it run for a few minutes, the pump initially will spin very fast but after it warms to its operating temperature (usually about 5 to 10 minutes) and the counter magnetism acts on it  it will fall to its steady state that is where you measure the flow which should be 1 liter every 15 seconds. 
Bob
I usually test pumps for 6 cycles of 10 minutes, 20 minutes, 1 hour, and repeat.
I have seen ones that turn way over the specification for about 2 minutes then drop due to bad armatures or worn brushes

[mbzse]

When you do the flow test on the fuel pump (the one by the tank), you have to let it run for a few minutes, the pump initially will spin very fast.../...

Just a reminder; do not run (spin) these pumps dry... i.e. always run them attached to hoses/pipes and with fluid flowing through
And for sure Bob; I'm not reminding you     I know you know

[BaronYoungman]

Mike if it helps, you can send your pump to me and I can hook it up to my test bench, Gratis.  Just please God drain the gas out of it and cap both ends for there are so many crevices the gas can hide and I do not want to meet another postal inspector.
Bob

[MikeSimon]

Thanks, Bob. I will come back to your offer once I get there and need it. I have accumulated quite a few parts, among them 3 fuel pumps. One, I think is almost new.

[Shvegel]

Mike,
More than happy to lend a hand if you need it.  Have presure gauges etc.