I've been investigating the leak from my HP pump. from what I can see, the leak appears on the shaft, immediately behind the pulley. Obviously the pump itself is full of hydraulic fluid and - from diagrams and the parts books - there are two rubber seals at points along the pump shaft.
I'm currently exploring three possibilities:
- I incorrectly assembled some parts/ seals
- The parts are too worn to seal
- The seals are the wrong size or damaged and ineffective
I can't investigate the first two yet, but I have been looking into replacement 'O' ring seals. I've been trying to understand how fluid would end up on the outer shaft behind the pulley - the routes it could/ would take. Here's that diagram again with what i think are possible leak routes added....
That seal fits again the angled face of a 'deflector' plate. Perhaps the clue is in the name. I think the angle serves to compress the seal both down AND inwards inwards around the shaft as the pulley nut tightens the 'deflector' to the bearing. The old seal I removed was certainly shaped by the angled face.
Part 35 fits into a channel around the outer circumference of a bronze (or brass?) bush.
The parts book call that bronze part a 'bush' and repair books describe it as a 'seal carrier distance piece' - implying it's secondary to the rubber seal - not the other way round. However that spring definitely pushes the part up against the bearing - so the carrier must be doing SOMETHING. And it's the carrier that contacts the shaft, not the seal.
When assembling, the instructions suggest that the bush and seal can be pushed in place and will hold the spring compressed. But as you can read HERE, that was not my experience: the spring forced the bush out of place. I now wonder if that was because the seal was the wrong size and not tight enough? If the seal was not tight enough, fluid might get past it and into the balls of the ballrace - bypassing the seal that is part 37.
I've been seeking other opinions from forums. Now. It's the carrier that contacts the shaft, not the seal (35). The seal seems to hold the carrier steady in the body (i.e. stopping it rotating) while the shaft spins within it. That being so, the consensus is that it's the face of the bronze bush that does any sealing - against the bearing. To achieve this, the surfaces must both be mirror smooth and perfectly parallel. If not - you get a leak....
Talking of 'other way round' I check my photos from my original rebuild back in 2018 and I'd definitely put that seal carrier bush in the right way round.....
However the shaft spins in the carrier so surely that means there is scope for fluid to get through the middle? But I don't think it could get any further: I recall that the bearing outer was a tight fit in pump body and that, when I pressed the pump shaft through the bearing, the bearing got pushed out a little and needed to be re-seated - meaning that the pump shaft was tight on the bearing. With little scope for leaks?
If fluid did get past, perhaps that is where part 37 - the 'backstop' seal came into play? Maybe the bearing didn't re-seat correctly? Maybe the bronze carrier (by now hidden behind the pressed bearing) was in the wrong position? Maybe, maybe, maybe......
The opinion I sought yielded some useful information:
- Leaks from the shaft were pretty common after a rebuild - a bit of a lottery
- Even the big repair companies had some 'fails' with these
- It was the face of the carrier, not the seal that did the (majority of the?) work
- At least one person said they were aware of over-sized plastic and metal seal carriers made up to replace the bronze carrier.
- Some people made up their own 'O' ring repair kits by buying near-equivalent sized seals
There was little scope to muddle the seals up, so I didn't think that could be the cause of my problem, however I did think about just replacing those two particular suspect seals - in case they were damaged or faulty. I wanted to avoid the expense of a whole set as I don't need all the seals and am not even sure that these two seals are the source of my problem.
From the parts manuals,
- Part 35 is meant to have an internal diameter of 21.3mm and a seal cross section of 3.6mm, giving an outer seal diameter of 28.5mm
- Part 37 is meant to have an internal diameter of 12.1mm and a seal cross section of 2.7mm, giving an outer seal diameter of 17.5mm
I was surprised to find that the EXACT sizes specified were quite hard to find. I had to look them up with a couple of specialist retailers. As you might expect for something so small, per unit they are very cheap. BUT - to justify their vast ranges and make any sale worthwhile, the specialist manufacturers require you to buy in bulk. To find seals in exactly the right size, it looks like I will need to buy in quantity: 7 in one case......
![]() |
A harder, '75 shore' Viton seal has been recommended for part 35 |
.......and 132 in the other!
In both cases it works out the same - I'll have to spend about £10 per seal type! But even with postage, that's still cheaper than a complete seal kit.
Because of cost, I did wonder whether they go for nearest convenient size, rather than exact size. But I am reasoning that if a near equivalent is under-sized, it stretches, reducing the seal cross section and affecting the outer seal size. If a seal is over-sized it's compressed and may 'wrinkle' or buckle on its inner diameter.... whether that is correct or not I don't know.
Maybe the problem is that - for convenience and cost - seal kits only contain near-equivalents of these two crucial seals? Conversely: if the correct seals are cheap but simply need to be bought in bulk, then if you are putting together many seal kits, how hard or costly would it be to include one each of the exact right size? I think my glass is half-empty on this and I suspect near-equivalents are being used.
I'm going to have a think about what route to take - exact size or near equivalents.
No comments:
Post a Comment