The D2 Forums

Rightthen wrote: Fri Oct 10, 2025 6:24 pm

MikkiJayne wrote: Fri Oct 10, 2025 6:08 pm
1, pull the pump cover in the boot, start it, and then run back to the boot and see if the pump makes any noise as the engine dies. If it starts whining / howling then its cavitating because it doesn't have fuel in the pump housing.

2, After it has died, loosen the banjo bolt on the top of the pump a little and see if you get a spray of fuel. The pump has a check valve in it to hold pressure. If the engine is dying through lack of fuel, there will be minimal pressure in the lines so you won't get a spray. If its dying because of other factors, the fuel will still blast out of that joint at high pressure.

I going to try this first thing tomorrow. There was pressure at the fuel filter today but that doesn't necessarily rule this out, does it?

Its a lot of pressure, up to 8 bar on the pump side. The filter will drop some due to gravity even if there's not much pressure, but it will spray from the pump banjo with considerable vigour.

MikkiJayne wrote: Fri Oct 10, 2025 8:19 pm

Rightthen wrote: Fri Oct 10, 2025 6:24 pm

MikkiJayne wrote: Fri Oct 10, 2025 6:08 pm
1, pull the pump cover in the boot, start it, and then run back to the boot and see if the pump makes any noise as the engine dies. If it starts whining / howling then its cavitating because it doesn't have fuel in the pump housing.

2, After it has died, loosen the banjo bolt on the top of the pump a little and see if you get a spray of fuel. The pump has a check valve in it to hold pressure. If the engine is dying through lack of fuel, there will be minimal pressure in the lines so you won't get a spray. If its dying because of other factors, the fuel will still blast out of that joint at high pressure.

I going to try this first thing tomorrow. There was pressure at the fuel filter today but that doesn't necessarily rule this out, does it?

Its a lot of pressure, up to 8 bar on the pump side. The filter will drop some due to gravity even if there's not much pressure, but it will spray from the pump banjo with considerable vigour.

Got it. There certainly wasn't 8 bar at the filter! Just a minor hiss and then a puddle. I'll be at it first thing and shall post my findings.

So if there's little to no pressure, the housing isn't being filled, then? Is that still applicable if there's nearly a full tank? And that being the case, is the culprit likely something along the lines of a badly seated o ring or similar? I was super careful when I replaced the housing so I'll kick myself if I missed owt.

Probably worth briefly reliving the full saga.. this all started during a cold snap at the beginning of the year. At this point I've replaced the battery, fuel pum and housing o rings, fuel pressure reg, fuel filter, been right along the lines, checked the connections/ water ingress in ecu box, dug out the spare key and rebuilt the current one.
This is all I can think of off the top of my head, but there's probably more.
Tbf it's only dragged on this long because I get no time to work on it. I feel like tomorrow is the day, though

When you reassembled the pump, did you make sure the return line went in to the pump housing alongside the pressure line?

It sounds like the pump is draining its housing dry then you lose pressure and the engine stalls. Enough leaks in through the pipework while its not running for it to be able to pump a bit on startup, but there's not enough in the tank / housing for it to keep going and get the fuel cycling round the tank. It often does this after a pump change and usually needs at least 20 litres adding to the tank to prime the housing, but with a 'nearly full tank' it should have filled the housing through the top.

Where is it on the gauge? If there's room, throw a jerry can of fuel in it and see if it helps. That said, if there's not room for another 20 litres it really ought to be working, so I wonder if there is something wrong in the pump assembly?

MikkiJayne wrote: Fri Oct 10, 2025 11:01 pm When you reassembled the pump, did you make sure the return line went in to the pump housing alongside the pressure line?

It sounds like the pump is draining its housing dry then you lose pressure and the engine stalls. Enough leaks in through the pipework while its not running for it to be able to pump a bit on startup, but there's not enough in the tank / housing for it to keep going and get the fuel cycling round the tank. It often does this after a pump change and usually needs at least 20 litres adding to the tank to prime the housing, but with a 'nearly full tank' it should have filled the housing through the top.

Where is it on the gauge? If there's room, throw a jerry can of fuel in it and see if it helps. That said, if there's not room for another 20 litres it really ought to be working, so I wonder if there is something wrong in the pump assembly?

There might just be space for 20l. I've put a jerry can or two in since the pump change. I'll try topping it up again if I'm getting no/low pressure from the banjo. I'm reasonably confident the pipework was correct. I've done that job quite a few times on various D2s now. I'm pretty familiar with it and was careful as i didn't want to repeat it. The thing with the pump was, when i put 12v on the old one it was still going strong, and the car basically behaved the same after the pump change. It seems likely to be an 'in the tank' problem because of the behaviour, but the history makes me feel like maybe I'm missing something elsewhere.

Is there a relay somewhere on the fuel system that could be dying? Or maybe another vacuum hose like on the reg? Blocked/faulty check valve? Or an air flow sensor that's shutting it down? (Although I'd see a code for that one, surely?)
I started this round chasing electrics, but the more i think about it, electrics would almost certainly come with other signs. This just dies until eventually the battery will drain from trying, then you get the ticking and the usual electrics nonsense.

I'll try the banjo soon as it's light. If the pressure is low I'll head for the petrol station. If still no joy, I'll maybe remove the pump housing and check everything over. Although if it runs with a full tank, that doesn't discount 'in the tank' problems as it could just be filling the cup from the top, right? Or does it sometimes need to have a totally full tank to prime everything? (This hasn't been my experience previously, just priming the cup with a little top up worked. But I think that was on PF if that has any relevance? )

I've you've put in a couple of cans of fuel since the pump change thats not going to be the problem. Likewise if the previous pump still runs well and there's no difference in behaviour that also implies something wrong in the pipework rather than the actual pump.

There is a relay for the pump. Failures are rare but possible. In this case it would have to be priming ok and then failing shortly after and not work again until it has a rest. An odd failure mode for a relay.

It will start with a massive vacuum leak from anywhere else so thats not likely to be the problem either.

Check if the pump is howling as the engine dies, check pressure at the banjo. Maybe one of those will tell us something.

MikkiJayne wrote: Sat Oct 11, 2025 10:26 am I've you've put in a couple of cans of fuel since the pump change thats not going to be the problem. Likewise if the previous pump still runs well and there's no difference in behaviour that also implies something wrong in the pipework rather than the actual pump.

There is a relay for the pump. Failures are rare but possible. In this case it would have to be priming ok and then failing shortly after and not work again until it has a rest. An odd failure mode for a relay.

It will start with a massive vacuum leak from anywhere else so thats not likely to be the problem either.

Check if the pump is howling as the engine dies, check pressure at the banjo. Maybe one of those will tell us something.

She won't fire today, just turning. Couldn't hear a howl but very little pressure at the banjo! Noticed the tank is just shy of 3/4 full. Would it still be worth a trip to the shell garage?

*there was pressure, but nowhere near 8 bar

3/4 full is plenty to fill the pump housing. Got to be something in the tank if there's no pressure in the feed line.

Is the line from the pump to the top of the tank still intact and tight? I'm wondering if either the union is loose, or the hose is split?

We have progress...


Put some fuel in to prime the housing. She fired up. I stuck my head in the boot and as she started to splutter, I heard the howl.

So is my next move to pull the housing and check everything? I'm asking the question because it's my least favourite job and I'm trying to avoid it

I'll post when I've got it out...

MikkiJayne wrote: Sat Oct 11, 2025 12:36 pm 3/4 full is plenty to fill the pump housing. Got to be something in the tank if there's no pressure in the feed line.

Is the line from the pump to the top of the tank still intact and tight? I'm wondering if either the union is loose, or the hose is split?

Ah maybe. I'll have a close inspection of everything. Strange that it's not priming unless I'm actually topping it up??