I have a friend with a 2004 SL600 M275 engine. He came back from a car wash 11 months ago, the next day the car would not start.

I looked at it and the car had no fuel pressure.

We put a volt meter on the pump connector and observed ~ 11.5V at the connector. Battery voltage for both batteries was ~12.5 and 12.8

Replaced fuel pump no change.

Could not figure out why the filter had 4 ports so we disconnected it and plumbed the FP right to the pressure gauge. Zero

Did replace the fuel controller with a used??? one no change.

My understanding is that this may use a PWM fuel regulator and in that case 12V measured at the connector may be meaningless.

I have limited access to WIS and STAR but don't really know how to use it well.

I did successfully confirm that the driver authorization system is in start mode.

__________________
To see my 129 parts for sale visit:
http://stores.ebay.com/The-Mercedes-SL-Store
John Roncallo

As the M275 has a on demand fuel system (not constant fuel
pressure vs intake pressure in the rail as previous cars has), and
that the car was washed, I would concentrate at the sensors in
the engine bay, giving input on what pressure is to be had.

I'm no expert on M275, nor do I have one, but please
read these documents;
http://fotifixes.com/M275-initial-training.pdf
https://web.archive.org/web/20060508104707/http://www.mercedestechstore.com/pdfs/512_Maybach/512%20HO%20M285%20Engine%20(Frech,%20W)%2008-06-03.pdf

As to find out whats wrong, I do not think you'll get anywhere
without error code / sensor value readings in STAR/Xentry.

br,
syljua

__________________
1998 CL500 W140
1984 230E W123

Thanks for the heads up. Real good info.

__________________
To see my 129 parts for sale visit:
http://stores.ebay.com/The-Mercedes-SL-Store
John Roncallo

Do you hear / feel the pump running when you key up or crank?

Using a digital meter can be deceiving. If a wire is nearly rubbed through / corroded and the circuit isn't loaded, a digital will show voltage that goes away once a load is applied. Using a incandescent test lamp is the way to go as it will load the circuit.

Is there a voltage rating stamped on the pump? I'd assume it is 12 rated, if so apply 12 v directly and see if you have fuel pressure / engine will start.

If this is a returnless system, the pump is slowed down to save power / wear on the pump / heating of fuel.

Any spark at the plugs / injector pulses? If not, I'd vote for a crank sensor issue.

This seems to explain alot. We will be applying 12V shortly. Second vote for CPS and he recently changed that so good place to look.

__________________
To see my 129 parts for sale visit:
http://stores.ebay.com/The-Mercedes-SL-Store
John Roncallo

Latest Info:

When the car was washed it did not include an engine wash.

Gets a cranking speed of > 120 RPM as read on an OBDII scanner. Engine will run for several second with a shot of ether. So does that rule out CPS?

Powered fuel pump directly and it runs fine.

__________________
To see my 129 parts for sale visit:
http://stores.ebay.com/The-Mercedes-SL-Store
John Roncallo

Runs fine as in:

1) Fuel pump only runs, engine won't ?

2 ) Fuel pump and engine runs ?

Fuel pump runs when a battery is directly connected to it. Does not run when installed it the car.

Occasionally I have heard the fuel pump run but never built pressure.

__________________
To see my 129 parts for sale visit:
http://stores.ebay.com/The-Mercedes-SL-Store
John Roncallo

Lets try this again. Have you applied 12 V to the pump while it is fully installed and tried to start then engine?

The goal is to determine if the issue is with pump control or something further up stream. ( Like ECM / injector control )

He has not tried that. That would make the pump run uncontrolled at full power. Any safety issues.

Maybe we could try that briefly to pressurize the system and see if it will just start. I believe it will.

__________________
To see my 129 parts for sale visit:
http://stores.ebay.com/The-Mercedes-SL-Store
John Roncallo

With most ( all ? ) cars using a single line system, the pressure regulator will be in the fuel tank ( or at least a blow off valve ) I don't see it to be an issue for testing purposes unless someone has 100 % knowledge otherwise. There is a chance the car will run rich during testing but I don't see that to be a long term issue.

Full power would produce a pressure of 9 bar. You need a lot of confidence in your lines and connectors to push it that far. The fuel filter acts as a pressure regulator, which is why it has the extra connection.

I think you need to begin by testing the sensor. It should output .5V at 0 psi.

Attached Files

sensor.pdf (296.4 KB, 157 views)

That's if the system goes that high in the first place. With your information that the fue pressure regulator is in the fuel filter, that is going to control max fuel pressure regardless of how fast the pump runs.

The goal of a variable speed pump is to reduce electrical loads / pump wear / noise / fuel heating. Trying to vary fuel pressure to control fuel mixture would be a losing battle due to slow response.

Did anyone say anything about using pressure to vary mixture?

9 bar is a lot of pressure, enough to be a potential hazard to the car and to the mechanic. Even if it's only between the pump and filter, there's a chance that a weak line can blow out. The system is supposed to work at a 40% duty cycle at WOT. It's a risk to run at full pressure, and the test is irrelevant.

He's already bench tested the pump, so we know it works. He can start the car on ether (not recommended, but what's done is done). So we can be fairly sure the engine works. What seems to be missing is fuel pressure, or more specifically, what's missing is power to the pump when it's in place. So the focus should be on the sensor and the pump control module. It's not hard to imagine how these elements could be damaged by running through a chassis bath at a car wash, as they are under the car. Something as simple as opening each electrical connector and spraying with silicone might displace any intruding water and get him going. Or it might that the filter has picked this moment to clog. Syljua's documents clearly explain how the pump system functions. Concentrate on pages 25-28 of the training document, and you're sure to find the problem.

I totally concur. If is starts by putting 12v directly to the pump I will still know jack. The possibility of a filter being so clogged suddenly that it wont allow even 1 PSI through is remote, but a pressure relief valve open in the filter could explain no pressure. But when the pump does run you can feel it run and generally it doesn't run. But the few times it did run it did not build pressure???? Although we measured voltage at the pump connector we didn't measure current which I believe we now have the capability to do.

The schematic shown in the training manual shows 3 ports on the filter but the car in question has 4 ports. This may have to do with the SL having a partition in the tank and needing to return to both sides.

And I just realized Syljua posted two links the first one is the key. The owner of the car is in FL now so this will have to wait a bit.

Many thanks for all the replies.

__________________
To see my 129 parts for sale visit:
http://stores.ebay.com/The-Mercedes-SL-Store
John Roncallo