Bill Ball

I thought if the impeller and pulley were in the correct location on the shaft, there's little room for the whole shaft to move much. If there is a lot of clearance between the pulley and impller and the casting, then that's another failure mode I hadn't considered. I don't see any significant clearance at the impeller side in the photos Roger has above. Hard to tell about the pulley side. I should go out to the garage and look at a pump I have, but it looks like if the shaft moved it would only be rearward, unless the impeller slipped as well.

Bill Ball

Yeah, Andrew, and it appears to be random as to what you get when you order a rebuilt pump, metal versus plastic impeller.

The plastic impeller would seem safer in terms of machining the block, but they appear to lose their grip and come loose more often than metal (although that's anectdotal). Also, I know of one instance in a track vehicle where the engine was destroyed by a loose plastic impeller. The damage was done by overheating. I guess that can happen pretty fast or it is possible the temperature rise was not noticed right away.

mark kibort

Bill, scots moved inward and the gap on the pump was only about 2mm at the most, and the front pulley wasnt rubbing. I wish we were having this discussion then, to see what really moved and where the failure was. I think the entire shaft moved inwards, but thats just a guess, but i do seem to remember the shaft being properly positioned on the impeller.

mk

ROG100

The drive is through the pully so it is quite possible that the pully could slip on the shaft under certain conditions - hard acceleration.
It would be good to see pictures of Greg's failure mode.

I am still contemplating the spot weld. It is about 30 years since I did any serious welding so would like some input from the experianced out there about the disimilar metals - anyone.

dr bob

The pulley is a plated steel piece, the shaft is cold-rolled steel, and the metal impeller is a cast steel piece. The dissimilar metals issue may not exist.

The method for "welding" is pretty simple, since it is just a tack-weld. You'll likely want to use your TIG torch with no filler metal. Flow a small section on each joint face, a very small section so the shaft doesn't heat up enough to damage the seals. Let it cool, then tack again 1/3 of the way around from your original tack. Cool again, then the third tack on each end. The tacks will be very small, just a puddle-width or two. Doesn't take much -- just enough to keep it from moving.

ROG100

Hi Bob,
Some of the pulleys are steel and the others are plated - either should be no problem?

I will practice on one and post the results.
Roger

blown 87

Some low quality pics of my pump
This first one is at the impeller side, it looks to have a step or counterbore in it and there is no evidence of it moving here.



pulley side



gap at impeller to housing.

mark kibort

looks like an internal problem with the pump. this is probably the more common problem, without a cure!

mk

worf928

I'm not so sure. If you buy a pump through Porsche you will get a plastic impeller. If you get one through 928 Specialists you'll get a plastic impeller.

I think 928 International ships metal impeller pumps? Jim???

My understanding of the common failure modes of the two (non-Lasso) types is as follows:

Metal - for whatever reason TBD, the impeller walks into the block just after it starts leaking. So, you have to be very, very observant of leaks down low.

Plastic - for whatever reason TBD, the impeller starts under-rotating with a consequent loss of flow and increase in operating temperature. So, you have to be very, very observant of operating temperature.

Thoughts?

mark kibort

yep, my plastic impeller pump failed very slowly . it would spin on the shaft and the pump would not work and the engine would overheat. i thought it was my sticking thermostat issue, but upon restart it would grab and start to work again. I was able to race most of the season with it like this. THEN, at laguna , it just failed in qualifying. so, when i got it off, the plastic impeller was just spinning free of the shaft and could be slid right off.

mK

ROG100

Dave,
Having looked at the 3 types I have available - Laso new - Porsche rebulit - Porsche original - all with metal Impellers, the issue appears to be the method of attaching the pulley and the impeller to the shaft.
They are all identical in this manner.
On the latest failure the metal impellar has migrated aft about .1" causing interferance with the block.
With the metal impeller it would seem pretty easy to either spot weld it to the shaft or drill a hole to take a pin. Same with the pulley.
The plastic impeller concerns me from the issue of how it is made to stay on the shaft. Does the plastc impeller have a metal sleeve where it attaches to the shaft. Or is it plastic to shaft?

I beleive most of the suppliers of WP can supply both metal and plastic. I can but personally prefer metal.

blown 87

You can bet I am going to tig my next one if I can not come up with a better fix.
I have been making some drawings of exactly how to do this.
Next step is to find another used pump to tear down and see how the shaft is made.

SeanR

One concern I would have with a quick spot weld would be balance. These things are spinning many times a minute, does anyone else thing that it might unbalance the shaft, causing premature wear and failure?

ZEUS+

That's what concerns me also. Two small spots 180 apart may work. Too much could upset the bearing.

ROG100

Blown 87,
I am pulling one apart this weekend and will share pictures.
I have a few cores hanging around.
Roger