Fuel pump circuit suddenly blows fuses

[Albie 0]

Don't know if this is in the right area, maybe a mod could move it if it is wrong.

 

Car is a K reg 2.0L 16V Corrado running completely standard. I have been having numerous electrical problems recently, (reversing lights, sunroof, rear spoiler, demister all stopped working and O2 sensor replaced), and got some great advice from you guys. These probs have mainly been dealt with now and are all working except demister. But I now have a more serious one come on overnight.

 

I had the car MOTd yesterday. It needed ball joints and a nearside CV boot replaced but got through ok. I used it last night for a good old burn up for 20 miles near home without a hitch. Parked it and switched off with no issues. This morning I tried to start and it point blank refused. On checking I found fuse 18, fuel pump and O2 sensor, had blown. I replaced it with another 20A but that also went immediately I switched on.

 

Another fuse later, I took out the fuel pump relay and unplugged the O2 sensor and everything stayed fine. Put the sensor back in, still fine. So no problem with the O2 sensor. I unplugged both the main and lift fuel pumps and replaced the relay, again the fuse blew! So the pumps themselves don't seem to be the fault. (Main is very quiet as it is pretty new and the lift pump is waiting to be replaced when I can figure out how to get the lid off the tank!!!) And the relay appears to pull in and drop out normally when the coil is just bridged across the battery with a wire.

 

Is there anywhere that is a common area for the wiring to the pumps to short out? I'm thinking of something like the front loom being cut where it crosses the inner wings which is a design fault! I have the rear seat up and the boot cleared and both seem to be fine. Any thoughts or hints?

 

Thanks.

[Supercharged 2]

It's usually the lambda wiring near the rear engine mount bracket that fails (engine loom side) - maybe worth a check but I would replace the FP relay first

 

Fuel pump cover - there isa special VW tool but you should be able to use a flatbed screwdriver and a soft faced mallet to break thge seal free - it just unscrews

[Albie 0]

Thanks for the advice, i will replace the relay as a matter of course asap.

Ive checked the system with the lambda probe disconnected at its lower connector - problem still occurs. I have chased wire up to the upper connector near the exhaust manifold and can find no problems.

Unfortunately lack of fuses means i have not tested to that point yet but i will do.

Thanks for the advice about removing the tank cover. I did wonder if brute force was the answer. Will get new lift pump in asap.

Once i get relay and additional fuses i will report back.

Thanks again

[Albie 0]

New info at last. I have just got back from an offshore course last week and got back under the Corrado. I have a meter and more fuses now so was able to investigate further. I can now confirm that the problem is right down in the O2 sensor itself, downstream of the connector down by the rear engine mount. Thanks for the tip regarding the connector up near the exhaust manifold, that was unexpected and it would have been difficult to find not knowing it was there. It wasn't involved in this case but for the future it is useful to know.

 

Everything is perfect up to that point, no shorts either to earth or across connections when measured at each of the fuel pump connectors. With the O2 wiring in place right up to that final connector the car starts up and runs! Connect the O2 sensor and the fuse blew as soon as I switch on, (that is even before I engage the starter motor).

 

Fortunately the sensor was just renewed recently from GSF, and it has less that 50 miles on it, and I kept the old one! It is a 3 pin type, one black and 2 white leads. Measured at its connector it shows open circuit from black to both whites and a dead short across the white leads! That is a dead short on the meter's 200ohm setting which should show even an ohm or two. The one I have just replaced shows the same except for about 6-8 ohms across the white leads. It's always difficult to measure such low resistances with cheap diy meters but 6-8 ohms doesn't seem a ridiculous value for this sort of sensor device. I only took the original out because it seemed stupid not to do so after over 50,000 miles on it when I was rebuilding the whole exhaust system. For the cost and use it has given it was better to replace when it was an easy job. Just as well I kept it for reference.

 

So I yanked out the new sensor to give it back to GSF with my blessings on them. Then I stopped to think. The connections to it are made with crimped tubing, soldering joints is NOT recommended for something of this low resistance in this sort of application. The individual joints are then shrink sleeved and a fabric based covering is slid over. I realised, shrink wrap sleeving near an exhaust!!! When I tried to peel back the outer cloth covering, there was resistance then a slight click as it came free and moved. On getting it free the shrink wrapped sleeving on the 3 joints had stuck together! They were carrying imprints of the fabric outer cloth in their surfaces. As I gently prised the 3 wires apart I found that 2 of the leads also had tiny perforations at their corners and were shorting. This almost needed a magnifying glass to see so it was a subtle problem.

 

As a hint to others like me not in the know, I would suggest that the method for replacing these sensors should include definitively staggering the 3 joints so they cannot possibly short out on each other even when the sleeving softens in the exhaust heat. I wish the instructions had shown that and not three leads of the same length!

 

As soon as I separated the joints the reading jumped back to 6-8ohms just like the original. It would seem that when I was moving, the air flow kept the temperature down, but when I parked up the heat in the tunnel around the leads, (which were well back from the exhaust as they should be), built up to the point where the sleeving softened and this could happen. Hence the problem occurred the morning after my original blast on the local dual carriageway. At least I will have a working new sensor in place and a spare for the future. I am renewing the joints as soon as I can get 3 more crimp type connectors and hopefully the thing will jump back into action then.

 

As a footnote another of my moans about poor design. What sort of engineer makes a unit requiring a 22mm ring spanner to go over it to remove and refit, then puts a connector on the end of the lead of about 24 mm across? Even a much bigger one would be not so frustrating at least. Wouldn't it have been nice to have had a connector which would allow the ring spanner over it to get it on the sensor? (Only moaning because I'm too tight to have a 22mm open ender and mole grips are so bloody clumsy!!!)

[Supercharged 2]

Quality parts!!

 

I think GSF sell an open socket to install with for not much £...