Workaround for bubbles in LHM return flow

[MatPL]

Hello. I haven't written in a while but the fight with the air continues. I bought a new Citroen LHM pump and fitted it to the V6 mount. It didn't change anything. The power steering was a bit worse than on the original pump from V6 (bigger pistons in V6). I also bought new connections to the tank (xantia) to eliminate cracks in the plastic, etc. Unfortunately, I can still see air in the pump supply line. At idle, there are a lot of them and they are microscopic, but when I press the gas, it draws large air bubbles (like in the video), which the pump breaks into even more microbubbles, which causes worse and worse driving the longer I drive.



I have looked at this pipe and the air is coming out from inside the tank, not from under the band. I watched it after about 12 hours of suspension rest, where there should be no more microbubbles in the tank. In the morning, after starting the engine, I immediately looked at the cable, which already had microbubbles in it. How is it that air gets into the hose supplying the pump within about 3-5 seconds from starting the engine despite new connections at the tank? I wonder if Mandrake found a problem because his symptoms were similar to mine.

[white exec]

Hi Mat,

One XM owner found a very small crack in the right-angled hose connection on top of the reservoir - the one which connects to the pump supply pipe.

To remove this possibility, try running a clear plastic pipe from the pump inlet to the reservoir, with the end of the pipe pushed well down into the fluid, rather than using the connection on the reservoir top. Make sure that the hose clip on the pump inlet is absolutely firm and decently circular.

Try running like that for a while. If there are no bubbles in the reservoir fluid, no bubbles should be being sucked on by the pump.

One other finding was that the reservoir supply connection could have a small ridge along it (I think on its underside) where the black plastic snout had been moulded. This ridge can be carefully filed flat, to make the surface properly smooth.

The member with these issues was Graeme (posting as Aerodynamica).

[MatPL]

I used to drive with a hose inside the tank once and there were no positive effects. Now, after replacing the steering gear, there is a big improvement in the process, so I will try again, but only after replacing it with a fresh LHM to save the LHM pump. I leveled the threshold at the connection before installing new connections.

[MatPL]

Unfortunately, driving without a suction filter and with a suction pipe directly in the tank does not give good results. The ride itself was worse, rougher and more nervous without the filter. Probably some of the air stays on the filter screen. After changing the LHM, I noticed that the air in the hose in front of the pump appears only 3-5 seconds after starting the engine, which would suggest a problem somewhere on the power steering. Return from the conjector comes only when it lifts the suspension, which in the morning takes a total of about 30 seconds before it fills the suspension.
I noticed that on the new LHM, air releases from the oil much faster. I also measured the time to turn on the conjector after driving in the city, which shows that the air disturbs the work of the hydraulics:
8 sec; 13sec; 16sec; 17sec; 21sec; 25sec; 28sec; 35sec; 51sec; 1: 16 min.
I also did a little front axle test because I have a problem with the front axle dropping down after a night. I have already used 5 sets of actuators, a few height correctors, a few SCMAC valves and unfortunately there is still a problem.
Yesterday before the night I went and started the engine for about 10 seconds, and I directed the return cable from the power steering / conjector to the 1L bottle. I repeated this 4 times. This morning the front is up and there is pressure on the actuators when the doors are opened.

[white exec]

Air is getting into the fluid from somewhere, by the sound of things.
LHM fluid thickens very slightly with age, so fresh fluid may allow air to remove itself more quickly.
The two tank filters are also there to break up returning air bubbles, and prevent them from being recirculated into the system. It's a two-stage process - large filter first (for the return pipes), then the small one surrounding the pump pick-up tube.

[MatPL]

The two tank filters are also there to break up returning air bubbles, and prevent them from being recirculated into the system. It's a two-stage process - large filter first (for the return pipes), then the small one surrounding the pump pick-up tube.

I wish you happiness and a lot of comfort in the New Year

I also have new filters. They can't handle air microbubbles. I did the 10min mod unfortunately without cutting the tube. I used copper elbows and directed the liquid to a piece of sponge (it retains microbables very well and connects them into larger ones that flow faster to the LHM surface). My wife said today that XM is enjoying a ride. For me, this is not the suspension work he can do. I ordered an air separator used to heat apartments. I wonder if the oil is also able to precipitate air during the flow.

[white exec]

Interesting, Mat.

All the fluid returns are fed into the interior of the large filter, and the fluid has to pass out through the nylon mesh, which is there to break up/destroy large bubbles, and to persuade small ones to break down. The microbubbles which do manage to get through the mesh are supposed to cross the reservoir slowly, and hopefully rise to the surface and dispel.

The smaller filter attacks bubbles for a second time, to prevent them being sucked into the pump and recirculated.

I'm interested in your additional sponge filter. Which of the return pipes was it fitted to?
It is obviously helping, as the ride is reported to be improved, and suggests that this return pipe is the source of the air.

One of these systems working properly (no air leaks) will not have any air (even microbubbles, which are usually tolerated) visible in the reservoir fluid.

[MatPL]

From my observations, the microbubbles that float in the tank with the engine off, very slowly rise to the surface. Probably because they have low displacement. With the engine running, the pump draws in oil, making it difficult for the micro-bubbles to get to the surface and draws them into the pump.
I put the sponge filter loosely (as in the photo) in front of the outlet from the return of the steering gear / conjector. I left some space for the oil to seep through the original filter and so that the sponge filter does not clog, but catches the microbubbles and fuses them into larger ones.
I think the steering gear circuit is a problem, because after starting the engine, the most fluid returns from there and under the highest pressure. I have already replaced the steering gear 4 times together with the heads on new seals.

[white exec]

Agree that the steering is the biggest flow in the system.

Could you put a length of clear tube into the steering return pipe (where it joins the reservoir) to check whether steering returning fluid is aerated? Ideally try this using a supply of non-aerated LHM - but watch you don't let the reservoir overflow (I have done just this; very messy!).

I wonder whether the steering mechanism is capable of letting air in? It is fed from the pump under pressure, so any leak ought to show up as just that - a fluid leak. The return pipe from the steering rack could conceivably be leaky, but even this is under very low pressure, so a leak should show up, and not suck air in.

The only part of the hydraulic system under suction is the pipe run between the reservoir and the pump input, and we know that air can get sucked in along this, where joints are not inwardly leak-proof.

Another possibility for air getting into the fluid is the existence somewhere in the suspension system of an pocket of trapped air, which slowly gets pushed into the fluid. Thorough bleeding of the system (by the correct procedure at all four brake calipers) should sort this. A very rare cause of aerated fluid is a gradual leak of nitrogen from a sphere, but this doesn't usually go on for long, and results in the sphere going flat, which hopefully would be noticed.

[lcwin]

Matpl you must remember almost no air can actually go into the high pressure hydraulic lines once the hydraulic pumps build pressure . If there is a microscopic leaks you will see oily pipes without fail. It is usually due to the intake circuit from the LHM reservor to the pump inlet or pump itself.
PLease google my comments on this as I am too lazy to repeat it again. I solved this problem many years ago but now not using a xantia. There is a lot of folklore about the bubbles . On a well running system there is almost no bubble visible in the tank. Good luck

[MatPL]

Czy móg?by? w?o?y? kawa?ek przezroczystej rurki do przewodu powrotnego uk?adu kierowniczego (gdzie ??czy si? ze zbiornikiem), aby sprawdzi?, czy p?yn powrotny uk?adu kierowniczego jest napowietrzony? Najlepiej wypróbowa? to z zapasem nie napowietrzonego LHM - ale uwa?aj, aby nie przepu?ci? zbiornika (w?a?nie to zrobi?em; bardzo ba?agan!).

I cannot separate the power steering and conjector return systems. The recovery system is more extensive on the ES9J4 engine. An external steering bleed valve at 160bar is attached and plugged into the return system of the conjector and additionally the steering gear. I've used a transparent tube on the return from the steering gear / conjector and you can see air there. When turning the steering wheel, the air flow increases.

Matpl, musisz pami?ta?, ?e prawie ?adne powietrze nie mo?e dosta? si? do przewodów hydraulicznych wysokiego ci?nienia, gdy pompy hydrauliczne wytworz? ci?nienie. Je?li wyst?pi? mikroskopijne wycieki, bez w?tpienia zobaczysz zaolejone rury. Zwykle jest to spowodowane obwodem dolotowym ze zbiornika LHM do wlotu pompy lub samej pompy.

On the supply line to the tank-> pump does not draw air into the system. I can see in the pipes that the oil with the microbubbles is running out of the center of the tube and into the pump. The power outlet in the tank has been smoothed out. I suspect that air may be sucked in the return line from the steering gear / conjector with a fast oil flow but low pressure. I'll build that line back again and see if it helps.



I bought the air separator from this video. Yesterday, after unpacking the shipment, it turned out to be too large for use on the return line from the steering gear. For now, I'm looking for a smaller one to see if it will be able to detach the air microbubbles from the oil.

[white exec]

You have pointed to the return pipe from steering gear to tank, and suspect that this is drawing in air.
As you say, this pipe has high flow, but low pressure.
It is theoretically possible that a fast-flow pipe can create lower pressure sonewhere along its length (a Venturi effect?) if there is a restriction somewhere.
What you might be able to do is gently restrict the flow from this return pipe at the reservoir, and see whether the pressure in the pipe (which will be higher is you move the steering wheel) is enough for a leak to be seen/found. That leak will probably be the source of air ingress.

[lcwin]

Matpl try this, You leave the car depressurise for a day to clear the bubbles . Get a thick Clear PU hose with internal diameter of the intake connector of the hydraulic pump . It must be nice tight fit without need to use any fasterner. The other end of the hose you directly insert into the LHM container without need to go into the filter but make sure its way below the level of the LHM oil.
. Start the car n see if there is any bubbles now. This way you straight away rule out LHM intake to pump problem. Dont worry about compatibility of LHM oil with the clear PU tube . Reactivity of LHM oil is very slow so using PU tube for a few minutes is no issues at all

[lcwin]

if there is a very small leak in any of the return pipes you will see wettness around the area . Its was a commom problems in citroen BX and even then there is not much bubbles with those leakage.My brother used to drive a BX and that car return pipes is probably biodegradable!

[white exec]

That is exactly what was suggested on 11 December, after Mat posted his video of large bubbles in a clear pump intake tube.