Best fuel shop to overhaul an EPIC injection pump?

[Fordtractorman]

It is looking like i may well have to remove my EPIC fuel injection pump (DSA1190) from my 1998 transit.

After going through the troubleshooting pages in the 'LDV EPIC service workbook' from 2001 to tackle my error codes- P1171 and P1173 i always get to the same point in that booklet, which is basically saying 'Mechanical fault within the pump. Repair the pump and retest the vehicle'.

If anyone is familiar (really familiar) with that booklet, im on page 11, which is for P1173 and i get to step 6. My value when doing that test is 31. Anything greater than 4, it says to remove pump as rotor has 'issues'.

It is annoying because the vehicle behaves, up until the 'pump internal temp' gets above 35 degrees C. My rotor demand and rotor feedback values on laptop (Forscan) follow each other fairly closely up to around that 35/36 degrees temp mark. Then they get vastly out of synch with each other. Then the P1171 + P1173 is generated.

Ive tried loads of fuel additives, two stroke oil etc, but she is now definitely getting worse.

Who in UK is really good at overhauling these EPIC pumps?. Several fuel repair shops ive spoke to, will not touch them. Which seems strange as these pumps are found on FIAT, Peugeot, Citroen and many other brands all over the world.

Or have i got an ECU issue?

[andypdq]

I believe there is such a low volume in this speciality that it's not economic for diesel shops to take this on, sorry to say.

[marcrbarker]

That's the rotor position thingy, it's a servomechanism that moves the rotor axially back and forth that varies the amount of fuel injected. The throttle pedal doesn't control the axial position directly, it's totally done via the ECU drive-by-wire. If the ECU sees the mechanism isn't axially where it expects it should be it'll throw a fault code or shut down engine depending on how bad it is.
The ECU controls this mechanism very similar to an electric window, with a separate Up and Down signals (solenoids). the 'window' also tends to move up and down by itself anyway. The ECU is watching it like a hawk (through FEEDBACK) continually tapping the up & down solenoids (feed & drain) to put it where is supposed to be.

So what goes wrong is anytime the ECU can't get the rotor where it wants it to be , demand and feedback won't track, it'll just try the best it can using what residual control it has. And throw a DTcode. You can get an idea of what's going on by breaking into the loop by looking at the pulse widths (an LED / resistor on each solenoid is good enough) which relates to the feedbacck. The lights will dimly glow under normal conditions then suddenly one will go off and the other bright then somtime later a DTC on the scanner.

Might be something silly like a failed Oring under the actuator attached outside the pump, or could be internal wear.

[Chug]

If it is ok up until 35C maybe its the thermistor in the rotor position sensor,

From the LDV manual
QUOTE:
The temperature compensation resistor is built into the
rotor position sensor. Its thermal characteristic is in
opposition to that of the sensor, thus eliminating ’drift’ of
the rotor position signal due to temperature changes
ENDQUOTE:


this post by MikeyL gives some detail/pics of the thermistor.
viewtopic.php?f=4&t=197333&c=1#p1703115

[marcrbarker]

If it is ok up until 35C maybe its the thermistor in the rotor position sensor,

good call. With thermistors if it's got the correct resistance at 25 °C it'll usually be OK at all other temps.
As I understand the sensor is a first order parameter only, tells the ECU that warmer fuel is thinner and will pass though gaps quicker.
A friend on his Citroen replaced a sensor resistor with a potentiomer on the end of a wire then adjusted it to taste got van out of limp mode.

Edit: what's going on page 11 the [url]google.com/search?q=epicservicemanual.pdf+filetype%3Apdf[/url]. Steps 1-5 are checking the rotor springs back each time stopped, stays put and that there's no dicky connections with engine shaking the harness and connectors. Until the engine is moving the ECU can't move the rotor so the feedback will be way out. That's normal.
6. Is about the other end of rotor travel to make sure the fuel shuts off completely. And the next steps looking for dodgy pins or broken connections etc

[MinorMatt]

Try Rochester City Diesel?

[Fordtractorman]

After doing more testing today, my van is now miss-behaving even when the Epic pump temp is around 22 degrees. She is now not idling smooth, at any temp.

Then as she warms up, the usual errors appear. But always P1171 first. So decided to go through the LDV booklet for troubleshooting this on page 9.

Could someone confirm that this is correct and that there are no errors/typos in this or the Ford EPIC training booklet:-

1) ECU pins 9 and 6 should be 48 to 60 ohms, when rotor sensor is at 20 degrees C.

2) ECU pins 9 and 12 should be 160 to 350 ohms

Because im getting 150 for part 1, and 280 for part 2. And thats when tested at the ecu AND the pump round plug.

If no typos in the booklets, then it would suggest i need a new rotor sensor. After reading online, i wouldnt be the first person to ever need one!

I think it might be part number 7186-009?

Has anyone got a source for a new one? Or am i barking up the wrong tree with this??

[Chug]

I have both pdf's the LDV and the Ford one,

in the Ford one it says on page 45..... pins 9 and 6 should be 48 to 60 ohms, pins 9 and 12 should be 160 to 350 ohms
in the LDV one it says on page 9 to check pins 9 and 6 and pins 9 and 12, but on page 37 it only lists resistance of 48 to 60 ohms for pins 9 and 6 @ 20C

As I said checkout MikeyL's post on repairing the sensor with pics of the inside.

[marcrbarker]

I measured this same axial sensor a while back and I got 59 Ohms / 285 Ohms
here : viewtopic.php?f=4&t=210355#p1746909

Looks to me like a bad solder joint or something not connected well

P1171 gets set when the FEEDBACK fails to track DEMAND while engine running, they're supposed to follow eachother closely. That would happen when the sensor goes screwy