Hi, I got a strange problem where the car after longer standing gets really hard to start; it is like there is no fuel, similar as if you just changed a fuel filter and system is not primed.
Once it starts it runs perfect and if stopped for shorter periods of time (1h for example) it starts right away.
If left overnight, it behaves again as if there is a empty fuel filter and it takes lots of cranking to get it to start.
Next, I disconnected lines going to tank and used two clear hoses to run it out of plastic bottle.
When this bottle was placed on top of the engine (higher than engine) the car would start perfect, even after standing overnight.
When this bottle is placed on the floor next to the car, lover than the engine, I get the same symptoms as running from my tank, but once the car starts, it just runs perfectly again.
I guess somehow somewhere there must be coming air into to the system during longer periods of not running.
Where should I start looking at?
Anyone had similar problems?

Additional things that I noticed but not sure if could be related are:
1) there is a small leak of fuel when car is running on top of the pump where lines for 1st and second injector are coming out of the pump.
2) as I used clear lines to connect to my bottle I notice that when car is running there are small air bubbles flowing in the return line; not sure if this is normal or it could show the problem of air being sucked somewhere?
In supply line there are no bubbles at all, so, I guess this air must come from somewhere to enter the system?!

OK, so just wanted to add something else I noticed that could be the clew.
I left the bottle with diesel next to the car on the floor and after about 2 hours try starting it. It was same thing, first just cranking, than slowly starting to catch-up and finally it started; once started runs fine.
What I noticed, that during this cranking process there was first mostly air coming back trough return line, then started to be diesel mixed with large bubbles and then, finally, once the car started to catch-up it changed to champagne-looking mixture of diesel with lots of very fine bubbles.
Finally, once the car was running smoothly I only notice a small line of fine bubbles flowing in return line, as described above.
So, what I can not figure out is where does all this air gets sucked into the system when car is standing with fuel tank level lower than the engine.
Any tips on best method to look for this leak?

Your eyes are your best help here. Since the car starts with the bottle in a high location, that means that it's replenishing a leak somewhere. Clean everything thoroughly, and carefully examine the flow path for leaks. Possibilities include the secondary filter, rubber lines, fuel heater, lift pump and primary filter. (Have your recently replaced the o-rings in the primary filter banjo bolt?) If there's no visible leakage, your lift pump may be leaking into the IP.

An '87 has no primer pump, it's an OM603.

In addition to what's suggested in Post #4, the fuel preheater thermostat is a *VERY* common point of air ingress. There's an O-ring in there that starts to leak. See if you have diesel weeping there, if you do, that's a sure sign you have air ingress at that point.

The weeping fuel at the delivery valves on the IP can cause missing at startup, but unless the leak is seriously bad, it shouldn't affect the rest of the fuel rack.

__________________
Current stable:
1995 E320 149K (Nancy)
1983 500SL 120K (SLoL)

Black Sheep:
1985 524TD 167K (TotalDumpster™)

Gone but not forgotten:
1986 300SDL (RIP)
1991 350SD
1991 560SEL
1990 560SEL
1986 500SEL Euro (Rusted to nothing at 47K!)

OK, thanks, will try to look really carefully.
This sort of happened overnight; was not connected to ay repairs or filter changes....
My model does not have a hand pump, so this is not the case.

Would it be OK to actually pressurize the system when car is not running (blow some compressed air in return line while closing the supply line) and then spray some soap water around to see where the bubbles are coming out, sort of like checking for a gas leak.
Could I cause some damage to IP by doing this?

Do you see any fuel leaks anywhere on the IP? There's a banjo bolt on the back for the return line that comes directly out of the fuel rack. If it's vibrated loose, you can easily leak the entire fuel rack down.

With a sudden change, you may be looking at cracked disc valves in the lift pump. They're made of plastic, and when cracked can allow fuel to travel "backwards" through the lift pump when it is no longer in operation. In normal operation, they serve as check valves.

__________________
Current stable:
1995 E320 149K (Nancy)
1983 500SL 120K (SLoL)

Black Sheep:
1985 524TD 167K (TotalDumpster™)

Gone but not forgotten:
1986 300SDL (RIP)
1991 350SD
1991 560SEL
1990 560SEL
1986 500SEL Euro (Rusted to nothing at 47K!)

OK, perfect, thanks!
it is dark here now, so will have a good look tomorrow.
I do have a visible leak on top of the IP where lines for 1st and 2nd injector exit the pump; could this be the place where air could enter in reverse when car is not running?
Is it normal to have some bubbles in return line when car is running or it should be totally bubble free?
Can I damage anything by tiring to pressurize the system and use soap water to find leak?

The O-ring seals on the delivery valves on top of the injection pump are a common leakage point on the 603. As said above, unless you're just pouring diesel out of there, they shouldn't cause the starting issues you're having.

Your return line should be 100% free of bubbles and foam. Usually air enters on the suction side of the lift pump, take a look at that fuel preheater thermostat and the rubber fuel lines.

__________________
Current stable:
1995 E320 149K (Nancy)
1983 500SL 120K (SLoL)

Black Sheep:
1985 524TD 167K (TotalDumpster™)

Gone but not forgotten:
1986 300SDL (RIP)
1991 350SD
1991 560SEL
1990 560SEL
1986 500SEL Euro (Rusted to nothing at 47K!)

Temporarily bypass the fuel heater - small filter directly to lift pump. If the problem goes away, air is likely getting in through the fuel heater circuit. If the problem persists, air probably isn’t getting in through the fuel heater circuit. You can isolate further with a temporary fuel source in the engine bay as you might have done.

Sixto
98 E320s sedan and wagon

Hi, thanks, this was a good suggestion!
Went from small filter directly to lift pump and still I get air flowing trough return line, so heater circuit is probably fine.
As I add throttle (rpm go up) there seems to be clear increase in amount of air flowing trough return line; supply line is totally clear of any air, so it must come in somewhere?!
I thought that part of the circuit after the pull-pump would all be in higher pressure compared to atmospheric pressure, so I am really puzzled why would air be sucked in; would rather expect the fuel to spray out if there is a leak????
Unfortunately my clear lines that go from lift pump to large filter and from large filter to IP are pretty yellow/brown so it is not easy to see if there is any air flowing trough there, but I did try lighting them from the back with flash lite of my phone while car is running and could not really spot air bubbles; although could be that they are too dark for me to really see them.


It now really seems it would be the most easy way to find a leak to pressurize the system between small filter inlet and return line lightly;y with compressed air and then try spraying soapy water around to see bubbles coming out; can I damage anything by doing this?!

Keep the air pressure low and you shouldn't have any problems. 15 or 20 psi should be sufficient to reveal the leak. The injection pump is designed to produce close to 2000 psi fuel pressure, but I don't think the lift pump would like getting a whole lot of air pressure applied.

__________________
Respectfully,
/s/
M. Dillon
'87 124.193 (300TD) "White Whale", ~392k miles, 3.5l IP fitted
'95 124.131 (E300) "Sapphire", 380k miles
'73 Balboa 20 "Sanctification"
Charleston SC

So there are months passed and the wagon is basically sitting unused.... I am going to give it one last try and solve this problem. Did start it today and once it works it runs absolutely fine; but after left parked for 2h and tried to restart, again the same, it acts like if I just changed the fuel filter and did not fill it with fuel; takes ages to start it.
Still running a clear return line from the fuel filter housing and can clearly see air bubbles inside, but really can not figure out where they might come from. Bypassed the whole fuel heater circuit, so am running small filter directly to the lift pump and am connected to temporary fuel canister to eliminate possibility of air getting in anywhere between tank and lift pump.
So, to confirm, the return line should be ABSOLUTELY free of air?
Mine has bubbles running while the clear line between the fuel canister and filter is completely bubble clear.
If someone had similar experience, please let me know where did you find the cause?
Thanks!

The return line should absolutely be free of air. Air in the return line means air is entering the fuel system somewhere. If you've bypassed the fuel heater setup and you're coming from a temporary bucket, you can rule out the hard fuel lines (although the fuel heater is still a suspect!).

The air can come from anywhere, the crush washers on the filter housing, the flare nut on the lift pump, the lift pump itself, the O-ring seals on the filter nut, etc.

Make sure you're using the correct size rubber hose (7mm) and fuel injection style clamps. Worm clamps and 5/16 hose can give you leaks that you just can't seem to track down.

__________________
Current stable:
1995 E320 149K (Nancy)
1983 500SL 120K (SLoL)

Black Sheep:
1985 524TD 167K (TotalDumpster™)

Gone but not forgotten:
1986 300SDL (RIP)
1991 350SD
1991 560SEL
1990 560SEL
1986 500SEL Euro (Rusted to nothing at 47K!)

Clean the top of the injection pump with brake cleaner. If it builds a coating of fuel, it means the delivery valve seals are leaking. It's not a hard job to repair, but you'll need a set of torx wrenches and a spline socket. There's a copper washer and an o-ring in each delivery valve that you will need to replace.

Thanks for suggestions, I do have a leak on top of injection pump, where line for first injector is exiting, but, I guess this could not be the reason for air entering the system?
My thinking is that here there is a very high pressure during motor running, so can not see how it could suck the air trough there?


I am now running straight from fuel canister into plastic primary filter (using clear pvc line) and this one is totally bubble free.
Preheater circuit is by-passed, so it goes directly into lift pump.
Return is connected from main filter housing back into the canister (also using clear line) and in here I can see clearly bubbles flowing.
The more RPM the more bubbles.....
Where I fail to find the entry point; wondering if there could be some good "trick" how to do that.
I suppose they can only enter in low-pressure part of the circuit?
Could return lines connecting all injectors be the cause; I do not think as in the past I had one broke and car was still running and starting normally; just sipping fuel there....
Any suggestion would be supper helpful...it is driving me crazy as I feel I isolated most of the circuit...