Fuel Pumps - PeachParts Mercedes-Benz Forum

anziani · #1 05-29-2004, 11:32 AM

I have a '87 300E with 170K miles. I'm getting a little nervous about the fuel pumps. Does anyone have any ideas at when to replace these?

J.HIDALGO · #2 05-30-2004, 08:32 AM

I did!
Replace them when they start to make a loud buzzing noise (it is about to go) or test the resistance. Do a search on the right upper corner of this screen. If they are the originals, they are probably due.

stevebfl · #3 05-30-2004, 09:45 AM

I replace Bosch pumps when they fail. I had a shop wagon 280TE that the fuel pump screamed for years.

Many cars have pumps that are a real liability and real diagnostic techniques have been adapted to quantitatively judge these risky pumps.

I looked for an example of a MB pump but there were none. There were plenty of GM examples and I will enclose one. This technique has been refined on Domestic cars and lots of data can be defined from the scope patterns. The pattern is taken by placing an inductive ammeter on a power wire. It represents curent over time. The humps indicate peak current at each commutator segment. Characteristic shapes of particular "humps" can identify a complete revolution and thus determine rpms. Aftermarket sources have compiled tables of rpm values for these perishable pumps.

Here is an example: Before I explain any further I will show the pattern and see if there are questions. This is a GM pump:

sunil190e-1.8 · #4 05-30-2004, 11:46 AM

i had a noisy original bosch fuel pump onthe 190e with 120k on the clock.

it was buzzing for at least a year before i replaced it.

it never failed but the buzzing got louder and hence i replaced it.

if yours are not buzzing,then you still have some mileage in the pumps.

stevebfl · #5 05-30-2004, 11:54 AM

Just about every pump I have replaced had no history of noise. If you wish a preventive repair I would warn against waiting for the noise. But if it is noisy I might agree some preventative repair could be valid.

Duke2.6 · #6 05-30-2004, 12:09 PM

Very interesting technique and current traces.

Obviously the used pump has some bad commutator segments and several more that are marginal.

Duke

cc260E · #7 05-31-2004, 04:15 AM

Quote:

Originally posted by stevebfl
[Many cars have pumps that are a real liability and real diagnostic techniques have been adapted to quantitatively judge these risky pumps.

The pattern is taken by placing an inductive ammeter on a power wire. It represents curent over time. The humps indicate peak current at each commutator segment. Characteristic shapes of particular "humps" can identify a complete revolution and thus determine rpms. Aftermarket sources have compiled tables of rpm values for these perishable pumps.

Steve,
can this test be done with normal amper metre or rather an oscilloscope

what is the time scale of your graph?

anziani · #8 05-31-2004, 11:40 AM

Thanks to all that replied to my request for info on Benz pumps. I shall await the "buzzing"!

stevebfl · #9 05-31-2004, 12:19 PM

The technique I explained has to be done on a scope. I erased the picture that had the x and y axis so I can't say the time scale, but I can put it in perspective. Typically the armatures have 8-12 segments and the pumps spin about 6000rpms. So that makes 48000 to 72000 humps per minute. Divide 60 seconds by that number and you get the time length of one segment.

BTW, the y axis has some identification and it shows 40 and 80mv. The tool used usually is calibarted at 10mv amp or 100mv per amp. Since I know the range of fuel pump current draw to be around 6-8 amps it is plain that the current drops to four amps during the bad segments from around 8amps when performing properly.

cc260E · #10 05-31-2004, 03:58 PM

Thanks Steve.