IP Leak at Banjo Bolt?

[avocado]

I've been trying to track down the diesel leak that my IP pump developed a couple of weeks ago, and I have some new data since I parked my bus in the garage. I have a couple of quick questions, and the background info is below:

(1) How many copper washers are supposed to be used on the banjo fittings? One or two?

(2) Are leaks at the outlet banjo fitting common?

Background: Engine is a 1.6TD from an '84 jetta that's been transplanted into a 1971 VW bus. I've developed a significant diesel leak after I've put about 350-450 miles on it (after the engine more or less sat for 9 years), and I haven't tracked down where it's from yet. Leak started in the middle of a 250 mile trip, and it's coming from the area near the outlet banjo. When I held a towel paper over the head side of the IP (where the metal lines go into it), diesel was spraying up into it.

First and foremost, it's leaking while it's sitting, but that's probably because the tank is nearly full and the tank in the bus is elevated, so it's got pressure on the outlet side (the solenoid keeps it from having pressure on the inlet side).

Second, when I packed paper towels around different places near the IP and let it sit for a few days, the fuel appears to be pooling below the outlet banjo bolt. Removing and replacing the paper towels and turning the engine over, I saw the same thing.

I replaced the copper gaskets on the outlet banjo (are there supposed to be one, or two?) and did a quick test on the cloth return line from the injector (blowing into it with the end capped to see if I could get it to leak), but the cloth line didn't seem to leak and the new gaskets didn't stop the diesel leak. I'll be replacing the lines anyway, but that was a first pass test.

Any thoughts on what to check next, other than replacing the cloth return lines?

So

[Quantum-man]

There are 2 copper washers on both outlet and inlet banjos.

Your leak likely as not is from the central bolt that you insert the timing gauge into. That bolt sees the maximum pressures from the piston, and needs to be fairly snug.

That washer is so small that it work hardens quickly with reuse, and so either needs to be re-annealed, or replaced every few experiments.
Officially every time, but I never did even after buying 200 on Ebay for less than the price of one from the stealership.

I had one go on a trip and it's a little frightening to see the squirting diesel when on the side of a busy interstate...

[TylerDurden]

The IP has the aneroid and the outlet fitting is on the side?

I would suspect the small return line. When they are cracked, they spit while the engine is running. With the engine off, they usually let in air as the tank return line tends to siphon the fuel toward the tank (in sedans).

Not sure there would be fuel leaking from the timing port when the engine is not running.

[avocado]

There are 2 copper washers on both outlet and inlet banjos.

Your leak likely as not is from the central bolt that you insert the timing gauge into. That bolt sees the maximum pressures from the piston, and needs to be fairly snug.

That washer is so small that it work hardens quickly with reuse, and so either needs to be re-annealed, or replaced every few experiments.
Officially every time, but I never did even after buying 200 on Ebay for less than the price of one from the stealership.

I had one go on a trip and it's a little frightening to see the squirting diesel when on the side of a busy interstate...

I will replace that with a new washer as soon as I get home tonight...thanks for the tip!

And +1 to the leak developing on a trip...very scary and very happy it wasn't a gasser, otherwise I suspect I would now have a burnt out hulk.

[avocado]

The IP has the aneroid and the outlet fitting is on the side?

Hmm...? Not sure I understand... The IP has the solenoid somewhere between the inlet banjo and the pump header (I presume that's where it blocks the flow) so I'm thinking it's likely that would block off the gravity feed of fuel before it hit whatever part is leaking - although I could be wrong!

I would suspect the small return line. When they are cracked, they spit while the engine is running. With the engine off, they usually let in air as the tank return line tends to siphon the fuel toward the tank (in sedans).

Not sure there would be fuel leaking from the timing port when the engine is not running.

I ordered the small return line from Jack last night, so if it's that, I'll find out soon enough! Thanks for the tip..

With regards to it siphoning back, there's a slight incline between the pump and the tank return, so I think the fuel only makes it back with the pump running, otherwise that line stays full of fuel, which can seep out any leak.

Thanks!

[Dakotakid]

Another alternative to the standard copper washers, are the Zexel (Japanese-made Bosch (I believe)) brand rubberized coated washers. I think the local diesel shops carry these. I used these back when I love/hated Mk. I cars. They worked very well for me.

[Fatmobile]

Yeah, the pressure inside if the "OUT" bolt can be pretty high but it's not too high once it's outside the bolt.

Hard to believe it could spray.

[TylerDurden]

Yeah, there's lots of little pulses of pressure in the small lines. They can spritz... not really a stream.

[avocado]

It's essentially a spritz or a pulse.

I'm pretty sure it's not the timing bolt (after inspection) because it's seeping from above that...

[Quantum-man]

Ok so you are connfident it's the out banjo area... Check that the 2 copper washer's are not scored. Check that the banjo itself is not scored....Wet 'n' dry with fine paper

[bscutt]

Avocado T. - the pump I loaned you has a bunch of good copper washers on it. Try swapping them in on the out bolt - I know they are good and seal up well.

[avocado]

I replaced the copper washers and it's still leaking. I'll check the banjo bolt next...

thanks volks!

[Quantum-man]

Get youself some +3.5 dioptre reading glasses, small mirror etc and try and get right up close to find out exactly which interface is leaking. Twist the banjo to see if the leak moves with it or stays still. If it seems that the groove is on the pump body, then there is a seal that you can make.

Look for my posting in the 'Pump Sticky' about the Plastic Bottle Hone Errr; I mean the Polythene Bottle Banjo Out-bolt Internal Pressure Measuring trick...

[Quantum-man]

Good to see someone come to your aid with a little sense, on the other forum. Obviously pressure outside the bolt, although probably a tenth of what you have inside the pump [which is 50 to 150psi incidently]

As a demonstration; pulling the supply from the injectors will cause a spout of diesel 3 or 4 ft skywards out of the banjo despite the orifice being 1/20th of that to the tank.
This is because of the pressure drop needed to force some half litre of diesel back to the tank per minute, and that's just at idle.

Additionally, because the out bolt hole is tangential, but not likely to be in line with either of the banjo barbs , it acts as a kind of cyclone which also resists flow, and gives pressure Not sure what the Reynolds Nos would be.

[Quantum-man]

Advocado;
Don't let CAD-man persuade you otherwise over on the other side, you are seeing pressure outside the bolt/inside the banjo He seems to have lost the plot...

The test for a good pump is passing nearly 0.5 litre @idle.
With a 3mm I/D pipe back to the tank that equates to a flow rate of 3.8 feet per second.
Gravity may give 6" per second at a guess when vertical, or lets be generous and say 1ft.
Definitely not down 10ft of tubing on a 10% decline.
Now go for a drive, and the internal pressure of the pump can treble; so you may be looking at 10ft per second of forced flow...