Workaround for bubbles in LHM return flow

[white exec]

If it's that aerated, and the feed pipe to the pump isn't leaky, I think you might have to remove all the return pipes at the reservoir, temporarily extend them with some of your clear pipe, and find which of them is feeding heavily aerated fluid into the reservoir.

Another cause of massive bubbling in the reservoir could be that one of the return pipes is pumping abnormally large quantities of fluid at speed, maybe due to a leaky hydraulic component somewhere.

[CitroJim]

The temperature should not make a huge amount of difference Xantos... One property of LHM is that its viscosity (as defined by its Viscosity Index value) remains more or less constant over a very wide temperature range...

[Mandrake]

Really good that you've seen that improvement by simply doing those things and nothing else. Would seem to take a while for the air to work its way out of the system, which makes sense.

What I found interesting is that I expected that after applying the sealant (and giving it a few hours to fully cure) that if it was going to work at all, that after a few minutes of up and down Citrorobics any pre-existing bubbles would be purged and the ride would then immediately be good - it was not.

I gave it 3 fully up/down exercises over a 15 minute period then went for a drive and there was no real improvement on how the ride had been for the previous several weeks. Still quite crashy and a little bouncy. Naturally I was disappointed and was thinking "well that didn't do anything useful..."

It wasn't until a couple of days later (with no more Citrorobics in between) that it dawned on me that the ride was now significantly better than the day I applied the sealant. Over the following few days of normal driving (it only gets about 20 minutes a day) it improved further then levelled out where it is now.

So my conclusion is that Citrorobics may only flush out large obvious bubbles - such as you might have in the system after changing spheres, disconnecting the pump feed hose etc, or perhaps a really bad air leak such as Xantos's car. And it probably flushes them out by returning them to the tank where they become trapped in the return filter and disperse to the surface of the oil and then to the atmosphere.

But perhaps on mine the air leak was small but persistent and the system was therefore full of dispersed, microscopic bubbles instead of large ones, and if they are small enough perhaps they can pass through the filters and back into the system, so lifting the suspension up and down repeatedly doesn't do anything to "de-aerate" the fine bubbles in the the oil - it just pumps them back to the tank, through the filter meshes and then back into the suspension relatively unchanged.

But once the air leak is fixed, over time and normal use these microscopic bubbles will eventually disperse to the surface of the oil in the tank and into the air (whether inside or outside of the return filter) and thus purge themselves. Just a theory but it would fit the observations.

If a small air ingress in those places is all it takes to compromise the ride, then it might explain why, with age, many cars just seem to lose that magic carpet ride, even though all the other things (spheres etc) have been attended to.

Great result!

There is no doubt in my mind that air bubbles do worsen the ride, causing both crashyness and a loss of damping and body control (for reasons I've explained at length in other threads) and I think the sealant experiment helps prove this, as I didn't touch anything else or even attempt to remove the hose - just wiped a fillet of sealant around it and let it set, so minimal disturbance and changed variables.

I suspect the sources of air ingress into the hydraulics are numerous on old cars - as the rubber return lines and feed pipe start to age and go hard its unlikely that they remain truly airtight - especially if the pipes have been disconnected and reconnected for any reason. By this age the rubber will be hard and moulded to the shape of the spigots, and if you don't get the pipe and clamp positioned precisely where it originally was, there will be an air leak due to the shape of the pipe not fitting the spigot and no flexibility left in the pipe to adapt itself to the new position.

On my previous Xantia V6 I replaced the two 50mm "joiner stubs" at each end of the feed hose to the pump (the one at the tank end had split underneath) as well as doing the 10 minute mod and the ride went from really crashy to exceptionally good and stayed that way for about 3 months and then deteriorated somewhat again.

I believe the reason for the deterioration was that the hose that I had bought, which turned out to be rather stiff, thick walled, high pressure hydraulic hose (since the softer grade I originally wanted wasn't available for a few weeks and I had to get the split hose fixed ASAP) just wasn't getting a good seal on the spigot after a while - especially with that odd spike on the end of the spigot mentioned earlier in the thread, which I should have filed down somewhat. Basically the hose was too stiff and not a perfect diameter match to keep a perfect seal for long.

As long as its not split, applying a little bit of sealant seems like a much easier and more successful solution than replacing that stub of hose at the tank - as it's very difficult to get a grade and diameter of hose that is soft enough and a good fit on the spigot... Of course on the pump end the heat and inaccessibility would make sealant impractical, a leak there would require a new stub of hose of exactly the right diameter and a wall that is thin and soft enough to mould well onto the spigot. It's easier to get a good seal at the pump end however as there is no odd bump on the end of the spigot like there is at the tank end.

[Mandrake]

Found some pictures of the replacement hose I fitted at the tank end on my previous V6:



You can tell just by looking at it that it's too thick and stiff compared to the original: (which had split)







But it's all I could get hold of at the time. It worked OK for a few months but it was never an ideal fit. Never got a chance to try replacing that piece of pipe again before I sold the car.

You can really see how much the original pipe moulds to the shape of the spigot and hose clamp, before going hard in that shape!



By the way, garden hose makes a good emergency repair if this piece splits - it's a perfect fit on the spigot and relatively soft. If it had been oil resistant I probably would have left it in place:

[CitroJim]

That's really good stuff Simon and another excellent piece of original research

I know well the splits the V6 stub pipes can suffer from...

Just one thing about using sealant... A lot of it is not proved to be LHM-proof and may well melt and thus ruin... It's a good idea to test it's compatibility before using it...

[Mandrake]

Good point about the sealant compatibility Jim, I hadn't really thought about that. I assume by testing it you mean squirting out a blob, letting it set then putting it in a glass of LHM for a few weeks ?

It was not intended to be a long term fix originally, more of an experiment to see what happened, as I was getting fed up with the crashy ride. Mind you, in theory it should not actually touch any oil where it is located - the pipe is under slight suction and the sealant is only on the outside where the air would be drawn in, so even if incompatible it may not be exposed to the oil.

Worst case scenario is it "melts" and starts leaking air again I suppose. If that happens I'll have another go at finding a supply of soft, pliable LHM proof pipe to replace the stub. And file down that stupid one sided bump on the end of the spigot!

Edit: a quick Google found this black silicone sealant that claims to be fully oil resistant:

http://www.halfords.com/workshop-tools/ ... ge-sealant

It's not the one I used (I used black B&Q stuff I had to hand that I had used on a bench top) but this would appear to be a suitable choice!

[white exec]

Lots of good points in your post, Simon, particularly how the original hose becomes moulded and hard, and hence problems of refit.
To improve the seal on the tank spigot (apart from using a hose clip that stays circular!), don't see any reason not to use two clips side-by-side, and some good joint adhesive on the spigot as well - something like blue Hylomar, which seems pretty immune to most oils, but which can be dismantled when needed.

[RichardW]

there is still a little bit of suspension clatter over really broken surfaces so I suspect I need to do the droplinks as well, (and possibly the lower balljoint on the right hand side...)

I emptied my scrap bin at the weekend, but I kept the old joints from the 307 for potentially welding to the stripped head on the joint in your Xantia I think I've got a spare hub nut or two lying about that may also be suitable.....

[Mandrake]

there is still a little bit of suspension clatter over really broken surfaces so I suspect I need to do the droplinks as well, (and possibly the lower balljoint on the right hand side...)

I emptied my scrap bin at the weekend, but I kept the old joints from the 307 for potentially welding to the stripped head on the joint in your Xantia I think I've got a spare hub nut or two lying about that may also be suitable.....

You really are a glutton for punishment.

If you insist. I'll be in touch some time after a certain not yet announced event has occurred...

This time I think we take it out and put it in a vice and use a scaffold pole instead of the impact driver though!

At least with a 4HP20, no gearbox oil needs draining and refilling.... (right Jim ?)

[CitroJim]

[At least with a 4HP20, no gearbox oil needs draining and refilling.... (right Jim ?)

That's right Simon

And congratulations on the yet to be announced event

I can make a good guess....

[RichardW]

It's the OS hub, so D/S not too much of an issue, as the centre bearing holds it in - but nice to know the darned gear oil won't run out when you get it wrong - changing the ball joint on the 307 is the only time I have forgotten to put a bowl under the diff.....and the only time I have dislodged the shaft

You really are a glutton for punishment.

Yes, but I do feel at least partially responsible for epic phail the last time! And I have actually done the job now

[CitroJim]

Oh yes Richard, I've dislodged driveshafts without putting a bowl under the diff.. The mess is horrible...

Worst of all is how gearbox oil stinks so...

[xantos]

Back to diameter 10 pipe and all is well, except for bubbles...

[chinkostu]

I might put a new length of hose on mine for the pump feed see if it helps. Theres a definate small weep at the pump end!

[Mandrake]

I don't want to jinx it, but the ride seems to have improved even further.

It's actually to the point now where on roads that aren't broken it feels like its wafting over the road and hardly touching it, but at the same time it doesn't wallow - the damping is spot on, and the car feels really stable. I also noticed on the motorway the other day when looking out the rear view mirror that the rear of the car is super stable - even on undulating motorway the rear end is as steady as a rock with almost no movement relative to the car behind it. This was not the the case before applying the fix.

I do still notice the struts creak and groan if I lift the suspension up fully, despite the recent strut lube (Grr) but they only do it once and then seem to be OK for subsequent raises, at least for a few days. On broken roads I do hear a little bit of clatter that is probably droplinks but you hear it more than feel it.

In short, super pleased. I can't wait to see how it rides once the drop links and lower balljoint are done, and it has some new tyres on it!