Xantia -project nearly finished, still have suspension issue.

This is the Forum for all your Citroen Technical Questions, Problems or Advice.

Moderator: RichardW

Dom_81
Posts: 23
Joined: 13 Jan 2017, 09:08

Re: Xantia -project nearly finished, still have suspension issue.

Post by Dom_81 » 07 Jul 2018, 18:24

HI Chris,

Yes whatever I did seems to have fixed it, but at the same time I'll only be fully confident after driving for a few days with no relapses. I've got a long drive tomorrow which will be a good opportunity to test it.

I'll double check all the hoses, especially the one you refer to and will make sure the seal looks good, I'm half tempted to go out with a silicone gun but i'm guessing that's not a good idea, potential contamination of LHM, and i guess the engine gets so hot that it would perish quickly. Yeah that's interesting about those filters, I'm generally impressed by how well these cars were made, so many clever little details - I wish they would start making them again but with a few upgrades.

User avatar
white exec
Moderating Team
Posts: 2888
Joined: 21 Dec 2015, 13:46

Re: Xantia -project nearly finished, still have suspension issue.

Post by white exec » 07 Jul 2018, 19:20

Advice would be to remake the connections on that hose, if they seem less than snug.
If there is any spare length, trim back the hose end and remake afresh.
At the reservoir end, the (10mm-ish) plastic snout (on the XM) had an annoying small moulding ridge running along it, which didn't improve the seal, especially if the joint there was repositioned. The ridge was easily removed with a very fine file or emery paper.
Genuinely circular hose clamps needed to avoid distortion.
Sealant not normally needed, but if necessary I'd recommend Hylomar Blue, not silicone.

Good luck; hope it runs well.

Dom_81
Posts: 23
Joined: 13 Jan 2017, 09:08

Re: Xantia -project nearly finished, still have suspension issue.

Post by Dom_81 » 21 Jul 2018, 18:18

Thanks white exec.. predictably the ride has gone bad again after the citrobics and my wishful thinking wore off.
Is it the hose nearest the front of the LHM reservoir which is under suction? And if so does that hose supply the LHM pump?
Are there any other potential weak points which might allow air in? I've got a bit of time tomorrow so will see if i can get it all sealed properly.

User avatar
white exec
Moderating Team
Posts: 2888
Joined: 21 Dec 2015, 13:46

Re: Xantia -project nearly finished, still have suspension issue.

Post by white exec » 22 Jul 2018, 16:04

Sorry for the delay in answering...
Click on Diagram to view clearly
Xantia hydraulic reservoir, fluid out.jpg
So far as I know, you were right, the (suction) supply pipe to the pump is the front one - arrowed red.

I also remember that someone a while back reported a leak on #5 (Reservoir Cap) and its O-ring #12, which allowed air to be sucked in. Cap no longer available, but the O-ring is. I think it was easily re-sealed.

>>> If I'm off the mark here, one of the team will correct the info.

Air can get into the fluid either as a result of this suction hose being leaky at its joints (or, and rare, a cracked or leaky reservoir top where the hose connects) - or because excessively aerated fluid is being returned to the reservoir from the rest of the system, and the de-aerating filters just can't cope. You can assess whether the fluid in the reservoir is aerated simply by dipping a sample of the fluid in a small clear container, and taking a shufty at it. Best done after the car has been on a bit of a run (say 10mins' drive). The fluid you'll be looking at will be what has been through the large (returns) filter, and is 'crossing the tank' towards the conical output filter to the pump.

There are lots of return pipes on the car, which tend to join up as they head towards the reservoir. If any of these are leaking, you'd be aware of it - drips on the floor. All the return pipes are non-pressure, and none of them are suction, so a leak will show up as exactly that.

Check also that the fluid level is correct in the reservoir. Your Xantia owner's handbook will show the correct procedure. If level is too low, air can be sucked in instead of pure fluid. Worth checking that the float/indicator is free and doing its stuff. If your indicator is the same as the XM one here, the indicator can be removed by rotating it anti-clockwise through about 90deg, and then gently lifting it out. It's a mechanical device, with a float-bulb at the bottom end. Check it's free to move ok by dunking it back in a few times.

Dom_81
Posts: 23
Joined: 13 Jan 2017, 09:08

Re: Xantia -project nearly finished, still have suspension issue.

Post by Dom_81 » 28 Jul 2018, 19:44

Thanks Chris,
This is all really helpful. The circular clips seemed to improve the ride for a while, now its gone back to be being bouncy and plonky.
I tried to remove the round central piece which all the octopus hoses attach to, but it seems that's not possible without either removing all the hoses, or removing the reservoir? I can't get the angle needed to pull it out.
I noticed another thread whereby aerodynamica said that he fitted makeshift downpipes to the return hoses in order to prevent excess frothing when the fluid comes back to the reservoir so thought I'd give that a try. I can't see any visible damage to the nozzle housing thing, I'll check the Reservior cap, although it feels pretty snug. Does it have a no return valve somewhere as the hose attached to the cap is open at one end anyway so presumably that would allow air to enter?
LHM level is correct and float is working. I used worm hoses at the pump and reservioir end, with 2 used at the reservoir end for good measure. I might use some of that hylomar which you mentioned. Will go check that cap, otherwise do you reckon the downpipe idea is worth pursuing?
Thanks again, much appreciated :)

User avatar
white exec
Moderating Team
Posts: 2888
Joined: 21 Dec 2015, 13:46

Re: Xantia -project nearly finished, still have suspension issue.

Post by white exec » 28 Jul 2018, 20:14

Dom_81 wrote:
28 Jul 2018, 19:44
Does it have a no return valve somewhere as the hose attached to the cap is open at one end anyway so presumably that would allow air to enter?
No return valve. The hose should not be open to air at any point. It's attached to the pump at one end, and to the reservoir output snout at the other. The snout turns vertically downward (blanked off by the 'cap' and its O-ring) and descends deeply into the fluid, and inside the output (conical) filter. As such, it should only be sucking up non-aerated fluid, for use by the pump.
>> Did you see any air in the reservoir fluid, after a dip-test?

It's possible that there's some annoyingly obstinate air lurking in the system. Hell Razor5543 recently posted a very detailed set of instructions for Citrobics...
viewtopic.php?f=3&t=60566#p581694
Might be worth running this, if any air/bubbles show up.

I did follow Graeme's (Aerodynamica's) thread about extending pipework in the reservoir, but wasn't convinced this was a general way to go. I must own up to not having had to wrestle with the problem on Xantia, and it may be Graeme was needing to sort out a problem peculiar to his car.

Getting the reservoir top ('octopus') out isn't easy, as there isn't a lot of slack in the fairly rigid pipework. On XM, I've found it necessary to remove the pipes from their snouts (label them!) to lift it out. The sealing of the return pipes to their snouts isn't hyper-critical (all they can do is leak gently), but the sealing of the suction pipe to the pump is, which is why decently round clips, an undamaged hose ends on that pipe, and maybe some Hylomar, is.

User avatar
xantia_v6
Forum Admin Team
Posts: 5518
Joined: 09 Nov 2005, 23:03

Re: Xantia -project nearly finished, still have suspension issue.

Post by xantia_v6 » 28 Jul 2018, 21:01

white exec wrote:
28 Jul 2018, 20:14
Dom_81 wrote:
28 Jul 2018, 19:44
Does it have a no return valve somewhere as the hose attached to the cap is open at one end anyway so presumably that would allow air to enter?
No return valve. The hose should not be open to air at any point. It's attached to the pump at one end, and to the reservoir output snout at the other.

I think that you are talking about different hoses. The hose attached to the filler cap is a breather hose, to let air into and out of the reservoir as the LHM level rises and falls. It is open at the other end.

I also don't subscribe to the theory that air bubbles are significantly created by theLHM dropping into the tank from the return lines. Any micro-bubbles have already been formed long before that.

User avatar
white exec
Moderating Team
Posts: 2888
Joined: 21 Dec 2015, 13:46

Re: Xantia -project nearly finished, still have suspension issue.

Post by white exec » 28 Jul 2018, 21:49

Agreed - the filler cap is vented (small open hose), quite different from the pump suction hose.
Tank is vented to allow air in and out as the tank level changes. Tank operates neither under pressure nor vacuum. By design, there is a large air space above the fluid. Reservoir also functions as a de-aeration chamber - bubbles bursting as they pass through the filters.

Also agree about there not being an issue with returning fluid causing aeration. In general, fluid returns in modest quantities, and not under any great force. Majority of returning fluid is from the power steering.

User avatar
Mandrake
Posts: 7603
Joined: 10 Apr 2005, 17:23

Re: Xantia -project nearly finished, still have suspension issue.

Post by Mandrake » 29 Jul 2018, 10:53

xantia_v6 wrote:
28 Jul 2018, 21:01
I also don't subscribe to the theory that air bubbles are significantly created by theLHM dropping into the tank from the return lines. Any micro-bubbles have already been formed long before that.

I'm not convinced by this theory either - my testing more than a decade ago on my first Xantia suggested that one of the design flaws was that the main return pipe - the one that comes from the octopus, actually goes too deep into the oil inside the return filter.

The result of this is that the turbulent return flow of oil with potential air bubbles in it exits the pipe at the bottom of the filter and is only a couple of inches from the bottom of the suction of the pump at the bottom of the supply filter with only the two micro filters to block the bubbles, so large returning bubbles tend to get sucked directly back into the pump instead of being able to disperse in the oil and rise to the surface of the tank. The filters can block very small foaming bubbles but not large bubbles.

In short the design error in the Xantia tank is that the pump inlet and octopus return are right beside each other. Ideally they should have been at opposite ends of the tank with some baffles between them so that returning oil could not be directly sucked up again, but would have to slowly wind its way horizontally along a series of baffles, during which time it would be able to deaerate or at least allow the air bubbles to float nearer the surface rather than being low down where the pump inlet sucks from.

The return pipe from the octopus does have a slit down some of its length to allow for changes in oil level but most of the oil still seems to come out the hole at the bottom. So in my first Xantia I fitted a U-tube that was strapped on the bottom of the plastic pipe that diverted the flow back up to the surface of the oil - the idea being that returning oil would be added to the top of the oil where the air bubbles could easily escape, and it did seem to work well.

On my first V6 I did a different version of the mod suggested by cc101 (?) which was simply to cut the return tube to about half its length with a 45 degree at the end, so that the oil comes out of the pipe higher up inside the filter. This did also seem to work, however one side effect was that when the oil level in the tank is low - such as when the suspension is fully raised, the oil actually comes shooting out of the pipe in mid air hitting the top of the oil, actually generating bubbles by churning up the surface of the oil - and because the oil surface is so low it was close to the suction input of the pump, thus bubbles were generated and sucked in.

In short, that version of the mod was OK at minimum or normal suspension height (where the shorter pipe is still under the surface of the oil) but problematic when you lifted the suspension to high - such as when doing citrerobics, where it might end up introducing more bubbles than it removed, and that did seem to be the case. For this reason I did not do that modification on my current V6.

Giving it some thought just now a couple of alternative approaches would be worth exploring - possibly in combination.

The flow rate through the return pipe is quite high through the small diameter so it may be beneficial for it to have a flared nozzle to reduce the flow velocity and spread it out over a wider area before it enters the tank. So cut the pipe short then attach a cone shaped end such as a tiny plastic funnel that is most of the diameter of the filter. That should cut the flow velocity down dramatically at the point it emerges into the filter and help avoid stirring up the oil or forcing a jet of oil directly to the bottom of the filter as happens now.

The second modification would be to implement a baffle between the return and suction filters that goes right to the bottom of the tank and most of the way forwards and aft, so that any oil passing through the return filter cannot be directly sucked into the supply filter, but must travel slowly around the outside of the baffle via the extremities of the tank, and thus partially solve the issue of lack of baffling by giving that oil time to deaerate.

Actually doing this would be very challenging though as the baffle would have to be inserted into and attached to the bottom of the tank (how ?) and line up perfectly with the filter assembly when it is inserted. I can't think of any easy way to do that. Perhaps a baffle that is small enough to pass through the hole at the top and thus be attached to the filter assembly instead of the tank would still give adequate baffling performance to avoid direct suction of the aerated return oil ?

It's been a few years since I've had a close look at the filter assembly.

User avatar
white exec
Moderating Team
Posts: 2888
Joined: 21 Dec 2015, 13:46

Re: Xantia -project nearly finished, still have suspension issue.

Post by white exec » 29 Jul 2018, 11:53

Have just been down to the workshop with a return filter, some LHM, glass jars and other bits - to repeat the tests I did on aeration and the action of the fine nylon filters.

Can confirm that large bubbles (anything bigger than the mesh size) do not pass through the filter, but remain inside it until they break down into something smaller. The only bubbles that do pass through are micro ones (presumably smaller than the mesh). A leaky, badly fitted or damaged filter would obviously change the situation.

At atmospheric pressure, it is actually quite difficult to get LHM aerated by, for example, bubbling into it through a 4mm plastic tube. The bubbles immediately rise to the surface and disappear. I can get LHM aerated heavily - large and small bubbles - by taking a container of it and shaking it up vigorously. It was this shaken fluid that I poured into a return filter, placed in a glass jar.

I think you are right though, Simon, about the short and over-direct route that fluid can take when passing from the first (return) filter to the second (suction). Bearing in mind the large size of the tank, it seems (in theory at least) that a larger separation/route would have been a good idea.

But, having said that, our beloved large LHM tanks have been around for a while now (!), and with the same relatively compact octopus arrangement. One notable difference between Xantia and XM tanks seems to be that Xantia has far fewer return pipes from the system (3?), where XM has six (of very varying diameters). However, all six discharge into exactly the same return filter as Xantia . . . but XM is not particularly known for fluid aeration problems which haven't been tracked down to either a leaky pump/suction pipe, or a damaged octopus pick-up tube or its cap.

One question remains for me, though. Heavily aerated (shaken) fluid - on the bench - does allow thousands of micro-bubbles out of and to pass through the filters. These seem to remain in standing fluid for quite some time, and are really slow to rise and disappear. Does this happen in normal driving? Only a dip-test of the fluid will tell, but what I do know is that when working normally and faultlessly (ie maximum comfort), our XM fluid contains no bubbles, even micro ones. On rare occasions where crashiness has occurred, micro bubbles are sometimes present.

Large amounts of air in the system are obviously not good, interfering with the proper action of brakes, steering and suspension. But I still don't know how well micro bubbles are tolerated before problems show up. By the looks of the reservoir design (and the fineness of the mesh), some transfer of micro bubbles round the system looks almost inevitable.

PS Inverted funnel-shaped spout I don't think would work: it would behave just like a tube end. A submerged 'sprinkler' (issues with back-pressure when flow is large?) might deal with large bubbles, but not small ones; in any case the existing filters do that already.

User avatar
Mandrake
Posts: 7603
Joined: 10 Apr 2005, 17:23

Re: Xantia -project nearly finished, still have suspension issue.

Post by Mandrake » 30 Jul 2018, 08:59

white exec wrote:
29 Jul 2018, 11:53
Have just been down to the workshop with a return filter, some LHM, glass jars and other bits - to repeat the tests I did on aeration and the action of the fine nylon filters.

Can confirm that large bubbles (anything bigger than the mesh size) do not pass through the filter, but remain inside it until they break down into something smaller. The only bubbles that do pass through are micro ones (presumably smaller than the mesh). A leaky, badly fitted or damaged filter would obviously change the situation.
Sorry I got the point about the bubble sizes back to front - you're right, the very small bubbles do pass through the filter mesh and the large ones do not. As for leaky/damaged - how good is the seal between the filter and the rest of the housing when clipped into place ? Is a gap there a possible route for bubbles ? It's just moulded plastic pushed up against moulded plastic after all.
I think you are right though, Simon, about the short and over-direct route that fluid can take when passing from the first (return) filter to the second (suction). Bearing in mind the large size of the tank, it seems (in theory at least) that a larger separation/route would have been a good idea.
If you look at the design of industrial hydraulic systems such as diggers, hydraulic cranes etc, I believe that not only do they have elaborate baffling in the reservoir tank, they also have explicit air removal systems in the return lines. So it obviously can be a problem.
One question remains for me, though. Heavily aerated (shaken) fluid - on the bench - does allow thousands of micro-bubbles out of and to pass through the filters. These seem to remain in standing fluid for quite some time, and are really slow to rise and disappear. Does this happen in normal driving? Only a dip-test of the fluid will tell, but what I do know is that when working normally and faultlessly (ie maximum comfort), our XM fluid contains no bubbles, even micro ones. On rare occasions where crashiness has occurred, micro bubbles are sometimes present.
I would say that a tiny air leak on a main return line - such as the PAS return would easily generate micro bubbles, as there is a constant heavy flow. In this instance, a tiny air leak may actually be worse than a larger air leak because a larger leak may generate larger bubbles that are easier for the filter to deal with than the smaller bubbles.
PS Inverted funnel-shaped spout I don't think would work: it would behave just like a tube end. A submerged 'sprinkler' (issues with back-pressure when flow is large?) might deal with large bubbles, but not small ones; in any case the existing filters do that already.

Why would a funnel behave just like a tube end ? How about a size adaptor which has a tapered section that increases it to a larger diameter and then parallel walls again for a certain distance ?

Complicated by the fact that the return line is near the edge of the filter not in the middle from memory. :?