Fuel pump relay 85 problem

Hi all.
My name is Pablo and I am from Spain and this is my first thread.
I own a XJ12 form 1987 and I got a problem with it.
The main issue is that once it is started the engine shuts off. I noticed that when it shuts off the 85 on the fuel pump relay loses ground. It doesn’t lose the 30 or the 86, it just loses the 85.
I checked the wire from the relay to the ECU at the trunk, the wire is in good condition, I anyway did a jump wire isolating the original one but the same thing happens. I read the entire manual but i can’t find any explanation why the ECU stops grounding the relay. I used the oscilloscope checking the 85 of the relay and the injector signal and fist the 85 ground dissapears and after a couple of seconds the engine stops and loses obviously signal to the injectors, so it is clear that only the relay pump loses activation. The next thing I am going to do is to check what happens with the 50 signla to the ECU and the positive and negative of the ECU but to be honest I thing that I am not going to find anything.
Have you ever have guys this problem before? Could the ECU be the problem?
Thank you and sorry for my english.

Failures of the fuel pump control circuit in the ECU are common and well-known – in the 6CU. By 1987 I think you’d have a 16CU, though, and we’ve heard less of issues with the fuel pump circuit in those. You might want to check the part number of your ECU and perhaps post it here, maybe someone can tell you more about it.

Does this car have oxygen sensors? I’m not familiar with the emissions regulations in Spain in 1987.

One thing that concerns me is your statement that you bypassed the ground to 85 and there was no change. Presuming this is the orange wire from relay coil to ECU, connecting it from 85 directly to ground should have fixed the problem.

Hi, thank you for your respond Kirbert, sorry for the misunderstanding, by bypassing the 85 cable I meant that I isolated the original cable and I put a new one but still conected to the ECU, in order to discard the orange cable, obviously if I put the original cable to the gorund the pump runs fine. I have the car since a couple of months and I haven’t checked if it has oxygen sensors, back then in Spain you were not required to have them but I will check that and the ECU type as well, hopefully I will do it tomorrow.

By the way my car is a 5.3

The fuel pump circuit in the ECU is a safety feature. It shuts off the fuel pump after two seconds if the engine isn’t running. That’s presuming you have crashed the car, are unconscious, and the engine has stalled. If you’re willing to risk it, you can just ground that orange wire and call it a day.

From what I saw in the wiring diagram, in case of a collision or whatever the main relay is shut off so the pump relay wouldn’t receive 86 anyway. When I jumpwire the relay the cars doesn’t run as good as it should so I think that I am going to open the ECU and try to fix it in case there are more circuits damaged.
I wiil let you know how it goes.

A pump that keeps running is a real risk. In the event of a leak the entire engine bay can get soaked with gas, making an engine fire impossible to contain. Don’t ask me how I know.

(And it wasn’t a Jaguar!)


To possibly clarify, Pablo; the pump circuit is independent of the injector circuit…

As ign is turned ‘on’ the ECU grounds the ‘85’ and the pump should run for 2 seconds - does it? While cranking and running (engine turning); ‘85’ is permanently grounded in the ECU, but the ECU, on signals from the coil/ign amplifier, ECU grounds the injectors for the specific injector opening time.

That the engine stops after a couple of seconds after starting (and as the key is released from 'crank?) means that the fuel pump relay is not the main issue; engine should run on with fuel remaining in the fuel rail. Unless, of course, you have a rapid rail leakdown due to a faulty non-return valve.

Your analysis otherwise is to the point, but you should verify that injectors stops even while the engine is turning. Also, how did you jumpwire the relay; if you just grounded the ‘85’ (still connected to the relay) - the ECU may use this as a makeshift ground also for the injector circuit. Otherwise, with the pump running and good injector circuit grounding, the engine should run perfectly…

You may have a bad general ECU ground - pins #2, 16, 34 an 35 should all have ground, and may serve different circuits inside the ECU?

There is a couple of loose ends, but I can follow your logic - and the fault is indeed inside the ECU…

xj6 85 Sov Europe (UK/NZ)

Hi Frank, obviously I know that the activation of the pump and the injectors are independent, what I meant with that test is that the engine stops (a second or so after while the rail still has fuel) because the fuel pump relay loses the 85, just because of that.
You are right about the ECU grounds, thats the next thing I wanted to do as soon as I can before opening the ECU.
I will keep you posted.

By the way the ecu in the end is a 6cu, 83705B.

That means one plausible fix would be to upgrade to a 16CU, if you can find one.

I just recently repaired a DVR I’ve had for years. Turns out there’s a shop online that has fixed them before, knows what typically fails, and offers a collection of replacement parts for that particular model. Basically, open the box, replace the items in question, close up the box. The only thing you need to know about electronics is how to use a $5 soldering iron.

We need the same thing for that fuel pump circuit in the 6CU. Either a list of parts to replace, or, better yet, a little package of the parts to replace. Saves having to ship the entire ECU off to someone – possibly overseas – to fix this nagging little problem.

Does the pump run, as prescribed, for two seconds after the ign is turned ‘on’, Pablo, and the ECU is powered up - confirming that those parts are working? And does the pump and injectors work in ‘crank’, which they then should?

When the pump stops, the the fuel rail will immediately lose pressure if the one-way valve has failed - and the engine will stop. With the ‘85’ grounded, the pump should run continuously with ign ‘on’, the pressure in the rail will be maintained, and the engine should run if the injector circuit is working. This sort of indicates which circuit, pump or injector, is at fault - the ECU is an analogue set-up and the more info the better. In ‘crank’ the ECU may use specific circuitry to run the pump and injectors - taken over by other circuitry to provide ground for ‘85’ and injectors, triggered by coil signal when the engine is running - hence the need for specific testing in ‘crank’ and ‘run’…

All that said and verified, it may aid faultfinding - though a mere visual inspection when opening the ECU may reveal something obvious. Bad soldering is an occasional ECU problem - but with an oscilloscope you are sort of sitting prettier than most…:slight_smile:

xj6 85 Sov Europe (UK/NZ)

Yeah, the pump runs for two seconds and when the engine is started it is activated again, it keeps running for some seconds until it stops. Normally that is the way it fails but sometimes doesn’t even start and other times it does the 2 seconds run but it doesn’t work once you crank the engine.
As I said the only thing that I am gonna check before opening the ECU are the positive and the grounds.
As for the oscilloscope, I am mechanich and I mainly only do faultfinding and I couldn’t live without, but I have to say that even most of mechanichs don’t wanna use it.
I bought this Jag pretty cheap because it has a lot of job to do to it but to be honest I didn’t know anything about this cars so even though I guessed the ECU was the problem it is always better to ask people that have encountered this problem before.
Hopeffully I will be able to work on the car this weekend and I will let you know what happens. I will try to post some pictures of the ECU.

1 Like

That’s gold standard thinking, Pablo!

As a mechanic you know that similar symptoms may have entirely different causes - and I have tried to outline some alternatives. However, you are likely right in your conclusion about the ECU being the culprit.

As you say, the ECU controls grounding (provided that proper grounds are available) to operate both the fuel relay and the injectors. The main problem is that ‘how’ this is done inside the ECU is vague - reputedly, even the constructors were probably not that sure. Being analogue, or part analogue for the V12, it was a bit of trial and error. Analogue data from various sources to be correlated for a digital selection of ‘best’ injector pulse width - which of course varies with various parameters from CTS, throttle position, rpms and manifold pressure. The resulting pulse width is of course easily read with an oscilloscope at any of the injector plugs, or ‘somewhere’ inside the ECU - but test data is not generally easily available.

Of course, erratic or bad ground connections will certainly interfere with operations - as will of course other faults within, which will not be plain sailing. Basically, with a running pump the engine should run perfectly if the other elements of the ECU works to specs - triggering injections with the engine turning and ignition working. Like you, I cannot really see anything outside the ECU that can cause the symptoms you so well describe…

Best wishes!

xj6 85 Sov Europe (UK/NZ)

Wherefore comes this rumor that the 6CU is analogue? It’s called Digital P.

Sure, Kirbert - but inputs are converted from analogue sensors. And despite the name Digital P; it’s based on old technology - hence bulky, but at least some elements are ‘accessible’…:slight_smile:

xj6 85 Sov Europe (UK/NZ)

Someone makes digital sensors? Because I don’t think my 2000’s Japanese appliances have those.

They may also be digitalized by separate computers. Kirbert - how may computers do you have…?

xj6 85 Sov Europe (UK/NZ)

Hi, sorry for the long wait, I wasn’t able to work in the car until some days ago. Finally I resoldered the ECU but the problem was still there. I finally solved the problem installing a fuel pump relay from Opel, serial number is 1238563, here we call these relays taquimetric relays it gets signals from coil, starter, etc it basically does the same function the ecu does internally. It is running perfect now.
Thank you!