Do you know what pump you have fitted?
The main one?

If you keep pumping the primer, will it continue to run ?

You seem to be aware that HP fuel pumps have two stages.
A low pressure suction stage and a high pressure delivery stage that feeds the common rail.

I can't remember which make, I think Bosch but do not rely on what may be faulty memory, have a key drive to the low pressure stage drive and this is visible if a small cover is taken off the back of the pump.
And known to shear.

Personally, if I had to take the pump off, I would be fitting a recon unit rather then faffing around myself.
And do timing belt and water pump while there.

As for the dealers, I would be going back to them with my receipt and asking for a refund.
General rule for future on diesels, if there is a manual primer bulb, there is no fuel feed pump to get fuel from tank and vice versa.

 

Thanks for the quick response Dancingdad. It is a Bosch pump (CP3 according to Haynes).


It's difficult to operate the primer continuously to keep up the pressure as you can't get your hand round where it is mounted, but you can keep it going for 5 or 10 seconds pressing it with your finger ends. Using the pressurised sprayer filled with diesel at the tank end, it started straight away and sounded normal when increasing the revs.


I agree about getting the recon pump, but I don't particularly want to waste £300 if there's nothing wrong with it. I could get it tested at the local diesel fitter I suppose. The HP side seems to be working OK, so I thought it might be worth checking the transfer pump without taking the complete unit of the car. However, what I didn't want to happen was a load of springs, ball bearings or rollers flying out and disappearing under the engine when I take the front cover off.


As (bad) luck would have it, I had a big clear-out before Christmas and shredded the invoice for the work, otherwise I would have contacted the garage and asked them to explain. Clearly they didn't fit a new HP pump for £140 + labour.


I still wonder whether there is another pump hidden under the car that I can't see without getting underneath, despite the primer bulb. It was an imported car originally (as it saved me over £3,000 off the list price), so may well be a bit of an oddball.

 

I attempted an injector leak-off test today. There was no leakage while just cranking, but the small vertical tubes filled fairly quickly while hand-priming and starting. However, there wasn’t any significant difference between injectors.

My rule of thumb is 40cc/min at idle and no more than a third difference between the best and worst

 

Many thanks for your input. I've just bought the leak-off test kit so it was the first use. There was no maker's name on the box and no instructions (what can you expect for £18.50 including delivery) so I had to go searching on the Internet. According to something I read, the vertical manometer tubes were supposed to fit Bosch injectors, but they didn't seal properly so I guess they were actually Delphi. Even so, there was no leak-off while cranking, so I concluded that the reason for the rail pressure not rising was not injector leak-off. When I hand-primed to get the engine running, it took about 15 seconds for the tubes to fill fairly evenly to about 10cm, which according to my calculations from the tube's cross-section is about 16cc/min, which according to your rule-of-thumb should be OK. I'll do a proper test with the bottles when I eventually get the problem sorted.

When the transfer pump fails it usually leaves metal particles in the fuel system.

Hey madnoel, don't wish that on me - it would cost more than the car's worth.


As far as I can see, the fuel filter on the dci 1.9 is fed directly from the tank, and does not have a return flow from the injection system which goes back into the pump housing. I believe it is the 1.5 dci which has the feed back to the fuel pump. I have replaced the filter, and although it had a bit of debris sitting on top of it (the input side), it did not seem to have anything shiny there, so fingers crossed. Thanks for the warning though.

 

When the transfer pump fails it usually leaves metal particles in the fuel system - has the fuel filter contents been checked using a magnet?? If metal particles are present then the complet fuel system will be contaminated including serious damage to the injectors which ain't cheap.

 

You have checked the timing ?

 

No I haven't, because I'd convinced myself it was a fuel delivery problem to the HP pump. If it hadn't fired up when I pressure fed the fuel system, I would have looked into camshaft / crankshaft sensors etc. As I said in an earlier post, I have now confirmed it is a knackered injector pump, but thanks for your input.

 

Does your pump look like this one?
The pipe assembly at the back is lifted off at around 1.30
You can see the key

Another that shows that how it works.
The gears in the back end assembly, back end assembly... transfer pump.

It is a pump rebuild to get it working but can remove end and check in situ.

 

Does your pump look like this one?


......... It is a pump rebuild to get it working but can remove end and check in situ.

Thanks for the videos Dancingdad (nice music). My pump isn't exactly like that as there are no pipe connections on the transfer pump itself. However, I found this very helpful document yesterday, which is a tutorial about the workings of injection systems in general, and common failures. In particular, it shows a photo of the CP3 transfer pump removed.


After seeing that nothing was likely to fly out, I decided to bite the bullet, brave the sleet and took it off this morning. You were absolutely correct, as the end of the pump shaft had sheared off and was split into two pieces. I tried turning the slot in the transfer pump with a screw driver and it feels very graunchy, so something nasty has occurred within. The cover is riveted on, so I can't actually see what's actually happened.


So good news that I have eventually found the problem. The bad news is that it's going to cost a recon pump, and I'd be surprised if I got the £60 cash back on the old one in the state it's in.


Many thanks for your valued assistance, it is much appreciated.:smile2:
https://www.google.co.uk/search?biw...1.268.4.4.0....0...1c.1.64.psy-ab..0.3.206...0i8i30k1j0i8i7i30k1.0.IMdHfjYtlBw#

 

Bleah..... but at least you now know what is needed and can do some shopping around.
I would go for a rebuild on your pump or a recon unit, not a second hand unless you want to risk the sort of thing that Noel spoke of.

With the transfer side, I think the bearings break up so biggish pieces and not very small fines.
The latter is the one to worry on as they can scoot through the filter and infect the whole system.
They usually happens when the pressure side of the pump eats itself.

But do check your system, get a sample of diseasal into clean jar and let it settle.
Any gunge at bottom and if so, will a strong magnet outside of jar move it ?
You do not want a yes answer.

One final for thought.
I have heard of people who have removed broken bits, bolted it back on and stuck an electric pump in the fuel line to feed the HP pump.
If your motor is of an age/value where a full pump isn't "worth" it, may be an option to try.

You will be checking the timing BTW, when you strip off the pump and refit... best swot up on that :wink2:

 

Bleah..... but at least you now know what is needed and can do some shopping around.
I would go for a rebuild on your pump or a recon unit, not a second hand unless you want to risk the sort of thing that Noel spoke of.

I agree, but as the pump shaft is actually broken as well as the transfer pump, I'm not sure if it's economic to repair, or if they'd take it in PX. I found quite a few recon units on eBay, ranging from about £245 upwards which is a lot less than if I'd taken it to Renault for fixing. Even though the car is not worth much second-hand, its only done a low mileage for its age, so worth fixing properly. I'm hoping to get another 5 years (40,000 miles) out of it. I've had it from new, and know all its history. It's still a very smooth economical runner when it goes.


I think I might have got away with the debris because it was such a sudden failure. The broken bits of the shaft were totally isolated from both parts of the pump, and because the shaft broke, I assume the transfer pump stopped turning fairly instantly.


I was going to get the timing belt changed on its next service any way as it's still on the original. It's still got its original exhaust system and it was still on the original battery until this happened, so it's lasting well.


I've done a lot of messing with cars and bikes over the last 50 years, as I never like paying for anything which I can do (in many cases better) myself. I'm just getting a bit old to be grovelling underneath cars, and certainly don't enjoy getting cold and mucky any more.


One final thing. The link to the pdf file I included in the last post was just one I found by Googling 'Bosch CP3'. It would be vey useful for anyone who might be interested in the workings of common rail injection systems, but was deleted for some reason by one of the mods. Did I break one of the posting rules? :crying2:

 

UPDATE & POST MORTEM

I though it might be useful to summarise the symptoms relating to the sudden failure of my Bosch diesel pump, and the relevant part of the diagnostic sequence which led to its discovery.

The failure resulted in the engine suddenly stopping and refusing to restart with no prior warning, no rough running, no bangs and no smoke. The ‘STOP’ and ‘SERVICE’ lights displayed on the dash, but no relevant fault codes were stored.:crying2:

The common rail pressure sensor appeared to be functioning correctly as there was 0.5 volts on the centre pin with the ignition switched on. However, the voltage remained unchanged when the engine was cranked, indicating zero HP fuel pressure, and possibly no fuel delivery to the common rail.

While hand-priming, I could see fuel flowing into the HP pump, and found that the engine would fire up if the primer was operated while pressing the START button. This was a significant clue, and at this point I could have saved a lot of wasted diagnostic time by simply removing and examining the transfer pump. This is fixed to the back of the HP pump by four T30 torx bolts, and is very easy to access and remove. A cover plate riveted to the pump body prevents any bits falling out and disappearing under the engine.

I found that the end of the HP pump shaft which engages with the transfer pump had sheared off and split into two pieces (first attached picture). This normally slots into two specially-shaped cast ‘washers’ in the transfer pump which drive the first gear wheel. After punching out the rivets to gain access, I found that both the washers had also broken (second attached picture). Fortunately, the cover plate prevented the resulting shrapnel from entering the pump body.

The two gear wheels appeared undamaged, but would not rotate freely, as one was catching on tiny indentations around the inside of the pump body. However, as these exactly match the pitch of the cog teeth, it appears that they were caused by lateral pressure on the cog while stationary, and therefore a result of the failure rather than the cause. I am guessing that the flimsy washers (which seem like they are not up to the task, as they are barely one millimetre wide at the failure points) broke first. The broken pieces then jammed the pump spindle which sheared off. Luckily, the teeth weren’t rippedoff the aged timing belt instead, with more costly consequences.

I replaced the timing belt, the tensioner and the water pump (despite them all looking like they were good for another 80,000 miles) and fitted a Unipart reconditioned pump which I bought for £150 from a nice man on eBay. After bleeding the fuel lines, the engine fired up on the first push and ticked over at a nice steady 800 rpm.

I did an injector leak-off test, and after 6 minutes idling the readings from cylinders 1 to 4 respectively, were 65, 75, 65 & 85 ml, (averaging to 11,12.5, 11 & 14 ml per minute). So despite the significant differences, the leak-off quantities seem acceptable.

The ‘proof of the pudding’ was a twenty mile round trip test drive. While averaging just under 50 mph, the on-board computer registered 66 mpg fuel consumption, which is about as good as it’s ever been.:laugh: