Fuel Pump failure

[barker_001]

Yesterday I had my first Failure to Proceed in 16 years of ownership... The car started to lose power on the A465 Heads of the Valleys Road at Pwnt Uchaf - the "Highest Point" - 1350 feet. 30 seconds later I was coasting down a (very opportunely located) slip road. It was rather cold and getting dark so not ideal for a breakdown... After pondering the problem for a minute or two I tried to restart the engine - it ran for a few seconds then died again. So there appeared to be a spark and no major mechanical issues, and we were able to limp into a car park in 3 or 4 attempts. I then noticed I couldn't hear the familar sound of the fuel pump.

Remembering what I had read about SU pumps, I removed the boot trim and protective plate, and gave it a few good hard taps with a socket wrench. Loaded everything back into the boot and turned the key. The pump ran, although the pulsing was very erratic to start with, then followed the normal pattern of slowing to a halt as the float chambers filled. Started the engine, and all appeared to work again. We drove another 20 miles home with no further issues. So it looks like the pump stopped for some reason, and the engine ran out of petrol.

The pump on my car is the electronic AZX1307EN, fitted by the PO in 2003. Never given a moment's trouble until now. Does tapping an electronic SU pump make any difference, or is this only a fix for the old points types, and therefore was it purely coincidental that it started running again after I hit it?

Another possibility I thought about was some debris in the fuel actually jamming the pump - can this happen? I thought there was a gauze filter on the outlet of the tank to prevent this.

Thanks for your thoughts gents, especially anyone with some knowledge of the SU pumps.

[rbgosling]

Hitting the pump I believe is a particular cure for the original points pump. I used to do it regularly with my MGB GT, where the pump is in a far more exposed position under the car near the O/S rear wheelarch! But that doesn't necessarily mean that there isn't a moving part that responded to "persuasion" with the electronic points replacement - I've not seen how they work, but there must still be something physically moving.

After my pump points melted, I replaced the whole pump with a cheap imitation Facet pump I found on eBay. Slight PITA tracking down some thread adaptors to make it fit, but it's been faultless for 5 1/2 years so far.

[Dwaynem884]

I have an AZX 1308 EN that gave up the ghost recently. Unsure as to when it was last replaced. It was poorly mounted by a p.o. . Took it apart and everything seemed to look ok mechanically, but with power to it, all it would do it make the circuit but not break it, therefore the diaphragm would not oscillate. Having no electrical experience, I bit the bullet & got a new one through LotusBits. They are the same price as going to SU Direct. I have the new one fitted and it's operating well so far. One point to note, the clamping bracket in the boot has 3 bolts, one to tighten the clamp around the pump, and two to mount the bracket to the body, one external (front) & one internal (rear). The rear internal is not a captive bolt. Mine, now unfortunately resides within some dark recess with the bowels of the car. If you can remove the pump without loosening the rear mounting bolt & nut, it would save similar heartache.
Dwayne

[barker_001]

Thank you Richard and Dwayne. Did either of you find any debris or fuel gum inside your old pumps?

[rbgosling]

Thank you Richard and Dwayne. Did either of you find any debris or fuel gum inside your old pumps?

My pump kinda caught fire. It was a horrible melty mess inside. I wasn't really looking for debris...

[Pete Boole]

My pump kinda caught fire

Pete

[rbgosling]

My pump kinda caught fire

Pete

Fortunately only the electrical points bit of it, no fuel ignited. But it wasn't a good moment. I still managed to bodge in a replacement pump by the roadside, and complete the journey!

[barker_001]

My pump kinda caught fire

Hmm yes, debris probably not at the top of your list on that day...

[barker_001]

Bump!

After the previous failure to proceed, the only thing I could find wrong with my old fuel pump was a slightly loose 12V connection. I tightened it up and all appeared well - until a couple of weeks ago. Same issue, fuel pump appeared to have stopped and the car was running out of petrol. Luckily I was able to get off the busy rush hour dual carriageway and gather my thoughts. After 5 mins I tried to restart, and the pump ran normally again, so I drove the car home (about 30 mins). This time, after looking inside the tank at the banjo filter (using a miniature camera) and changing the fuel filter (a few bits but nothing serious) I ordered a new AZX1308EN pump.

I'm about to fit it. What should be used to create a seal on the elbows where they fit into the pump body? In order to get them to the correct orientation (about 10 o'clock for the outlet, 6 o'clock for the inlet) they are not fully tight. I thought about blue Hylomar but I'm wondering if it will "set" sufficiently to hold the elbow in position. Some type of Loctite perhaps?

Thanks for your thoughts...

[barker_001]

The saga continues. Following on from the previous post, I fitted the brass elbows using a combined threadlocker and sealant, Loctite 577. This was recommended by Loctite themselves. I fitted the pump, and finally managed to get the fuel filter bowl to seal - this is much easier if you unbolt the bracket from under the wheelarch, and turn the filter housing so that you can check all the seals are seated properly. The new pump sounds far more lively than the old one, I need to get it seated properly in the bracket and foam so that it doesn't reverberate through the car, but that's a minor issue.

However - the pump doesn't appear to fully stop when the float chambers are full. It does the normal fast running as the float chambers fill, but it never quite stops. It pumps once about every 15 seconds, so fuel is going somewhere. I don't appear to have any leaks around the pump or filter, so I am wondering about the needle valves, and whether the new pump is overwhelming the seal they are supposed to make. The valves are the OEM viton tipped type, changed about 20,000 miles back. Should / could they have failed this quickly?

[Lotus-e-Clan]

DHLAs need between 2 and 4 psi ...3psi typical. If you haven't got an in line gauge then that's not any use to you. Doesn't take much to overwhelm the float valve. You could check the level in the carbs though? Just taking out the emulsion tube out might tell you if they're overfull or otherwise.

[John]

With the air cleaner off and throttle held open it may be possible to see if fuel is flowing continuously into the inlet manifold? You'd normally expect an initial flow from the accelerator pump on opening the throttle, but if it keeps flowing then the needle valves are not doing their job.

[dancroft1984]

I just fitted a new pump from powerspark. Was £67 and so far it seems good. Plastic, but it seems to work well.

[barker_001]

With the air cleaner off and throttle held open it may be possible to see if fuel is flowing continuously into the inlet manifold? You'd normally expect an initial flow from the accelerator pump on opening the throttle, but if it keeps flowing then the needle valves are not doing their job.

You've hit the nail on the head John. Fuel is continually dripping into the No. 4 barrel. Next step is to order some new needle valves then - thank you.

[barker_001]

I bought some new needle valves from Eurocarb and fitted them. Eurocarb advised that the standard 8809.170 viton tipped needle valves are no longer available, and recommended the 11135.170 valves instead. These have a slightly stronger spring and are apparently better suited to slightly higher fuel pressures, as on the Turbo Esprit. Fitting them stopped the flood into the No. 4 barrel. However...

Once the float chambers are full, the pump never quite stops pumping, which the old pump did. It slows to about one pulse every 20 seconds. I have verified that the needle valves have closed and no more fuel is entering the carbs, and have also tried disconnecting the output from the pump altogether and blocking the pump output connector. In both cases though, the pump continues at about one pulse per 20 seconds. Following the advice on the SU / Burlen website, I checked the tightness of the diaphragm screws on the pump and found they needed tightening by about ¼ turn, but this has made no difference. Also checked the tightness of the banjo connection onto the fuel tank and the input pipework, all appears OK - i.e. the same as when the old pump was fitted and came to a full stop.

Looking online at the Jaguar / MG forums, etc. as well as Lotus, it seems that some other (but not all) SU pumps do the same thing, and the owners have verified there are no fuel or air leaks. Has anyone experienced this - is this normal behaviour for an electronic pump, or could there be an internal issue with the pump that I should be concerned about? Thanks in advance...