the flyin' scotsman

Nice diag. Jim................got me wondering about the intank filter now?

Mrmerlin

the little bubble is part of the assembly, not sure why

AO

I'm sorry to say your new in-tank pump will look just like the old one, except in one piece. You may want to pre-emptively replace the hose with somethinge else (sounds like that's the plan anyway.)

worf928

Those pumps are pretty nifty little devices. Kinda wish there was a small strainer between the two pumps though.

Does it have a bit of red-ish color to it? Check the two electrical leads between the two pieces of the in-tank pump. I'm still betting you will find a bit of one of the leads missing and that the volume missing will be startlingly similar to the volume of the bit you show.

The first dead pump I ever dissected had valiantly chewed through most of both of the leads before seizing. On the track. On the first run of the first day, of a three day track event.

I've dissected two pumps since and both seized on bits of the in-tank pumps wires.

I don't know why. But, that's funny. Horrible. But funny.

jcorenman

Agreed. I've been looking for a small strainer or filter that could go between, but anything with 1/2" hose fittings is not small.

The piece of crap looks brown, not red. The leads are red but look perfect (except for the solder joints being ripped from the inlet fitting). There was shrink-tubing covering the solder-joints, also perfect except for detachment.

The fluid in the pump was red, which is the remains of the mystery oil that I had flushed it with prior to brutalizing it with a plastic hammer and 24 volts. The piece of crap that was jamming the pump looks like a small piece of brown plastic film, thin mylar or kapton or similar. It doesn't look like anything that was part of the pump assembly. The inside of the tank also looks really clean so I suspect it is something that floated in with a recent fill-up.

Interesting. I can see that happening if the in-tank pumps are loose for some time, especially on a racetrack. This trip was mostly highway miles (except for a spirited drive from Lee Vining through yosemite park...). But once the rubber hose separates then the main pump is unprotected, and it is just a matter of time before something gets in there.

Charley B

Jim that debri looks very similar to what i found clogging the pump on my '82 last year. Turned out it was the in tank weld material used to hold that seperator piece in the bottom of the tank. I ended up replacing the tank.

I would at least remove the sending unit from the top of the tank, and go in with mechanical fingers and a strong light and see if there is anything to fish out.

If the seperator is still solidly attached and you can find no evidence of further debri I would consider my guess wrong.

AO

I would think we could find some type of in-line fuel filter/screen that can fit that hose between the tank and the pump.

Seems to be plenty of length... Just have to find one skinny enough to fit.

sharknoob

3.8" long
1.7" dia

FredR

The in-tank pump will only put sufficient head [pressure] to get the gasoline into the main pump thus prevent it from cavitating. Thus if you fit something in between the in-tank pump and the main fuel pump to strain/filter the fuel you need to be careful and ensure there is sufficent head to overcome the newly installed kit that will invariably create hydraulic resistance.

Regards

Fred

Imo000

The older 32Vs didn't have an in tank pump. Why the need for the GT? I'm running a boosted engine with a single pump and it's running just fine.

FredR

S4's and GT's were fitted with the in tank pump from MY 1989 I seem to remember [for sure in 1990 models]. As to the reason why I suspect it is as detailed in my earlier post. On my 90 S4 I found that under hard acceleration and left hand bends I suffered fuel starvation as I experienced in an autocross event when trying to "run light". Porsche do not usually modify something without good reason nd perhaps this was an attempt to mionimise this problem although I suspect the additional pump makes no real world diference.

Regards

Fred

dprantl

My '91 starves for fuel MUCH less than my '86 did when the tank is below 1/4 full on left-hand turns. The '86 actually was very annoying in this respect.

Dan
'91 928GT S/C 475hp/460lb.ft

FredR

Apologies- I meant right hand bends in my earlier post- the fuel shifts to the left and exposes the suction outlet or so it seemed.

Fred

jcorenman

Wise, this one is. Sure enough, the new pump had the same molded rubber hose (at least it looked the same, I assume it is the same crappy self-destructing rubber).



I cut off the clamps (Oetecker 20-23mm, about 1.5mm thick), replaced the hose with 1/2" weatherhead fuel hose from Napa and new clamps (Oetecker 18-22mm clamps from McMaster, item# 54105K53). I got the 304 stainless clamps, these were slightly thinner (1.2 mm). In retrospect the zinc-plated steel would have been a better choice as they are thicker (1.5 mm, same as the original), McMaster # 6541K53 (18-22mm) or 6541K42 (19-23).

I also picked up the "official crimp tool", McMaster # 6541K67 (about $20) and also made by Oetecker. Basically end-nippers with a slight radius instead of sharp edges. The clamps tend to compress into a "football" shape (especially the thicker ones) so I gave them a squeeze across the "ears" with pliers after a first tightening to make them round again, and then tightened them further. They seem nice and secure.



I un-soldered the wire connections to make it easier to get the new hose in place, although it might be possible to squeeze the hose in carefully. Be careful about too much heat, I noticed the plastic around the pump terminals starting to soften.

I still think it would be a good idea to protect the main pump from any debris being shed by the in-tank pump, although with the proper hose that shouldn't be an issue. I don't think a small filter is the right answer... the in-tank pump pushes 4-5 psi (IIRC) which is plenty to push fuel through a small filter, but if the in-tank pump comes loose again then it stops pushing. Sucking fuel through a small filter is a good recipe for air-lock and getting stuck again.

I did find a Barry Grant mini-filter with a 2" dia screen rather than a paper element, that would stop the big bits that would jam the main pump without adding any significant resistance. And it is easily disassembled for inspection. But no versions with 12mm hose-barbs, that will require some adapters or machining. And I don't know if it will fit when all that is done, there is not much room in that area with the cover in place.

Here's a link: http://www.jegs.com/p/Barry-Grant/Ba...61189/10002/-1

I've got one on the way, and will update this when I find out if it will work.

So with all that back in place fuel pressure came right up and was stable, engine fired right up and ... ran crappy! It backfired, stumbled, didn't want to idle, generally very cranky. Fuel pressure good, WBO2 said the mixture was in the right ballpark,

It was running perfectly right up to the moment that the fuel pressure disappeared, but this didn't seem like a fuel-related problem. And backfiring is either ignition (crossed wires?) or cam timing, right? So we pulled the distributor caps and upper valve covers and checked the timing (perfect!), caps and rotors were worn but clean, but what's this? When I pulled the coil wire out of the left-side distributor cap, the terminal stayed with the cap... the wire-end was completely toasted and disintegrated and the terminal was welded into the cap.

Swapped that wire and cap with a spare and viola! Ran perfectly!! But how did the fuel pump failure cause the coil wire to get fried?? All I can think is that this has been going for some time, and with a fully-charged battery there was enough spark volts to arc over the disintegrating terminal. But with a lower battery state (sitting for a few days, doors open and lights on etc) it didn't have the juice to jump the bad terminal and then fire the plugs reliably.

Anyway, that's my story and I'm sticking to it. Thanks for all the help and ideas, and I'll update this if I come up with anything else.

hans14914

Funny how there are always contributory factors to a secondary failure. Last week I was gluing some new carpet in, and it caused a gas-leak (hose from the intank to the primary went bad), but it wasnt leaking the day night before, or before I started the job, just having the key in the ignition and cycling it to move the seats around was enough to cause the line to fail. I happened to have 2 replacement hoses, so all was good, but I never would have thought gluing carpet would cause a completely unrelated failure. Glad you found the troublemaker quickly.