Bleeding hydraulics

Not exactly model engineering, I know, but there's such a wealth of knowledge and wisdom here...

I have had to replace the seals in the rams of my tractor's front-end loader. The single-acting rams are arranged with the cylinders pivoting on the frame fixed to the tractor, and the ram rods extend outwards and upwards, to lift the loader arms. The hydraulic seal is at the 'open' end of the cylinder, the seal bearing on the rod. There is no provision for bleeding.

The cylinders always point upwards, to a variable degree, thus any air in the system will rise to the seal area, potentially making the seals run 'dry', apart from any oil carried to them on the surface of the moving rod.

I can't see how it's possible to be sure there's no air in the system. The inside of the cylinders (luckily not working surfaces) have rusted a little, suggesting water and air (mayonnaise?) have contaminated the hydraulic oil.

Local tractor engineers say they never attempt to bleed such external hydraulic systems, after they have been opened - 'the air just works its way out'. I think this is Kiwi 'She'll be right' engineering - dangerous! The only way out for air in the cylinders would be past the seal. Is this likely?

Whilst I'm fairly sure I can arrange things to get the system air-free (it will be a bit of a performance), I'd be interested to know what 'proper' engineers would do about such an hydraulic system that has been made with no provision for bleeding.

The workshop manual for my mini-digger says you should fully extend and retract all the rams several times to expel air after working on the hydraulic circuits. In a lifetime of working with farm machinery that always seems to work for me. I guess the seals are oil-tight but not air-tight.

Mike.

Hydraulic systems in wind energy equipment are also 'self bleeding' You "just run them" and the air comes out I was told.

Having stripped and reassembled hydraulic systems mainly on diggers with double acting rams I have never had to bleed them working them after assembly normally clears air from the system. Just make sure that they are fully extended and retracted to expel the air. Check the levels in the tank while doing so.

None of the hydraulics on the recovery trucks I worked on needed bleeding. We just operated each ram to and fro as necessary until it worked properly.

I'm guessing that as it's only single acting it's an "older" style front loader and possibly an old tractor? Most of these tended to have the hydraulic pump immersed in the gearbox oil and which nearly always immulsified . Our old Fordson and Massey tractors always had milky hydraulic oil.

Years ago when I fitted new seals to our old mucking out Dexta, the rams were primed on the bench and then fitted. With a ram removed from the front loader fully collapse it and pour oil through the connection hole and when you're sure that the chamber is totally full and with the hole uppermost connect its already primed hose. Do the same with the other ram and then fit them. There should not be any significant air to affect the seal. The difficult part might be collapsing the ram, maybe with a big sash cramp or just push it against a wall or other solid surface.

Double acting cylinders are no real problem as they can only retain a small amount of air as the cylinders can be force emptied, carrying any air at least to the supply pipes.

These systems are not like brake pedals (with limited volume supply and very limited travel) - there is/should be an infinite supply of oil at working presure supplied by a pump that will only need to compress any air in the system each time the valve is operated. Even the amount left in a single acting ram is fairly insignificant after a few full operations. Any air which becomes entrained in the oil will be carried back to the reservoir and have time for some/most to escape the liquid before re-pumping. Air compressed to a couple hundred bar, or whatever, will operate the ram just the same a the liquid fluid - except there may be a slight amount of ‘bounce’ at the bucket over rough terrain.

Seals are seals. They do not leak air.

If these installations required bleed points, they would have been installed as original equipment?

The thing which needs protection is that of avoiding cavitation in the oil pump.

Second the extend/retract a few times advice. I'll add that the working pressures in a tractor hydraulic system are huge (e.g. 2500psi), and any attempt to bleed them would be potentially hazardous. Resist the temptation to loosen any fittings when the engine is running or if there is any residual load on the system. When I park a machine up I try to work all the levers after switch off to depressurise it.

+1 for NDIY's advice. Even if there is a trapped air pocket in a dead end, the amount of fluid and the velocities with which it travels will carry any air away readily. It's good advice to sweep the stroke of a newly installed cylinder a few times to exchange the oil and pass the oil with entrained (or even dissolved) air back to the reservoir tank. But normal operation will clear any remnant air in the fullness of time anyway.

Also bear in mind that even if the actuator cylinder is full of air, introducing fluid at pressure will pressurise it anyway, and effectively there is no limit to the amount of liquid available. Thus the hydraulic system will reach enough pressure to overcome the load (within the normal working constraints of allowable applied pressure) even if there is trapped air in the system. It just means that more oil will be needed to achieve working pressure that would be the case if there was no air in the system. The hydraulics operate, but may behave oddly.

Nothing could be further from the truth with conventional mechanical hydraulic brakes, where the available volume of fluid is that of of the (part) stroke of the master cylinder. In consequence of the limited fluid displacement there is a very limited (if any) exchange of fluid each operation, so any air included is not swept back to tank where it can be rejected by the fluid.

I did some experiments once upon a time with the behaviour of air in brake fluid. Actually it dissolves into the fluid quite readily at pressures of typically 2 - 3000 psi, but then comes straight back out of solution when the pressure is released.

I made a hydraulic log splitter ,double acting ram,running off my 1973 International 474 tractor ,which had single acting hydraulics so fed the hydraulic fluid back into the gearbox via the gearbox filler,and operated the ram via a modified control valve, the gearbox ran on Internationals own Hytran hydraulic fluid ,ran the tractor for 35 years and the logsplitter unit for about 15 years and never had any air in the system and never had any problem with oil going milky which seem to occur with Fordson majors and others.

As others have said a new hydraulic setup should be cycled a few times which will cause the air to be encapsuled in the oil and once it has circulated through the system the air will naturally expell when in the reservoir. We used to assemble about 10 loader combinations a day and we never had a problem with trapped air.

As an aside a friend of mine worked for a very prestigious sports car manufacturer and they had a tool that they fitted between the seat frame and the brake pedal which pushed the brake pedal down and it was left overnight. When it was removed in the morning the feel of the brake pedal had improved out of all recognition and they didn't have to spent a long time firming up the pedal!

Thanks everyone for your advice and warnings.

I'm sceptical that air will work its way out in a reasonable time, either as an air-oil foam or as air bubbles: the dead space in the cylinders and pipework is quite large, compared to each cylinder's stroke volume. Some fluid exchange will of course occur, but there'll be a lot of shunting backwards and forwards, and the air will surely find its way to the high points - in this case, behind the seals, I appreciate the system will (sort of) work as an hydraulic/pneumatic system, but the pipe lengths are very different to each side's cylinder, so there would likely be different degrees of 'compressability' on each side of the loader's arms.

I plan to (try to) pre-fill the cylinders and pipework, before re-installing the cylinders. I have plumbed in Tee unions at the input to each cylinder, so can use these as filling/bleeding ports. I can move the piston rods manually, to use the cylinders as pumps, with a bit of added grunting. I take the warning about the danger of bleeding under pressure, but I can crack open the diverter valve, so there's a substantial pressure drop across the valve, and be careful not to allow the circuit to be working against load: i.e. bleeder open and cylinders at minimum stroke.

If any of this sounds stupid, please let me know!

It sounds like I may be over-thinking this. It's a bad habit. Perhaps that's why I get so little done...

Hi, Kiwi Bloke.

I must agree with you that you are thoroughly over-thinking the (non-existant!) problem!

I did some work for the worlds largest digger company and they did not see a need for bleeding the system in production. By the time the linkages were fitted up and everything was checked out to the various tilting and opening and closing buckets the job was considered finished.

Believe me if bleeding was required they would have done it and to the ninth degree!

OK, I'll stop worrying. It's obviously not a significant problem in practice.

However, the thought of a log-splitter with a lot of air in the system is frightening - all that stored energy to be released as the log 'pops'! Some of my eucalyptus logs go with a final bang, as the stress in the splitter's frame is released.