Inverted hydraulic vangs, problems, Revelations, ruminations, and esoterica....

DDW

Super Anarchist
6,707
1,231
I’ve expressed an opinion in the past that I hate hydraulic systems, because they always leak - either now or in the near future. But I’ve been unable to come up with an alternative which didn’t require extensive modification/re-engineering of the rig. So I have spent about one year attempting to fix, or at least understand, the issues. This is esoterica, but to the extent it may help or entertain someone:

The inverted vang designed for my rig was described in the earlier thread, the gas pulls up and the hydraulic pressure pushes down - upside down from a conventional vang. It turns out that this is significantly more challenging for hydraulic design. Seals do not want to be dry on either side, Navtec recognized this and instead of putting gas pressure under the seal (virtually guaranteeing a dry wall) designed in a remote gas accumulator in a separate chamber above the cylinder proper: gas over oil in it pushes oil through a transfer tube to the underside of the piston. Seemed like it would work, provided any gas could be bled from the underside, for which they specified a procedure.

The First Revelation is that bleeding gas from an open hydraulic system (that is, one with a free surface exposed to gas) is a fools errand. It is understood within hydraulic engineering (perhaps deep within) that gas will defuse into the fluid over a period of just a few hours. And not in small amounts: approximately 10% of the volume of the fluid. It does not change the volume, it occupies the space between the molecules of oil. The most familiar example is a carbonated soft drink. When you open the top and release the pressure, the dissolved gas expands and fizzes out until the drink is flat - but the fluid level in the drink does not change. Hydraulic fluid does the same thing and responds the same way. The gas is dissolved in it at the local ambient pressure, and is released by the same conditions: a drop in pressure, a change in temperature, or agitation. It’s pretty easy to observe this with a little equipment. Put some fluid in a pressure chamber, add 600 psi of argon, wait 24 hours, then depressurize and pour the fluid into a clear container. It will fizz for many hours afterwards. If the gas volume is 10% of the fluid volume and diffused at 600 psi, when you reduce that to 14 psi you will get 4 times the fluid volume in gas out.

The result is that gas will be everywhere in a hydraulic system regardless of your bleeding efforts, and agitation or changes in pressure (from pumping or releasing the vang) will release free gas. If the gas has a way to be trapped, it may build up and never go back into solution. In a conventional vang, this is of little consequence: the gas chamber is on top, there is a puddle of oil between it and the seal, gas can dissolve and come out again, you don’t care. The hydraulic port on the bottom is drilled up through the rod to the top of that chamber. Any gas bubble there will exit first as the cylinder is extended, but for the small amount above the waterline defined by the port, hence it is self bleeding to a great extent.

In the inverted vang with a hydraulic accumulator, gas released under the piston has no way to exit as the port is necessarily at the bottom. It continues to collect, pumped in a little at a time dissolved in the fluid and released by agitation and pressure change, until it pushes all of the fluid out. The pressure for operation is maintained, but the bottom of the seal is now dry, does not seal well, and wears quickly.

The Second Revelation is that the pressure in a conventional vang is (almost) always higher under the piston, and lower in the gas chamber, so that it is impossible to leak gas. Consider a vang cylinder partially extended: the pressure on top due to gas acts on the whole piston area, the balancing hydraulic pressure below acts on the piston area minus the rod area, static equilibrium requires the pressure there to be higher by the ratio of active areas. Add a clew load and there is even more hydraulic pressure. Only if the boom is abnormally heavy, or the cylinder is allowed to extend to the stop, does the gas pressure exceed hydraulic. If the piston seal leaks, hydraulic oil will leak into the lower pressure gas side, the gas will not leak into the hydraulic side.

On my inverted vang, the gas pressure is underneath on the rod side of the piston, and will be greater than hydraulic pressure while in the berth. This is reversed when loaded by the clew, but the great majority of any boat’s time is in the berth. The gas wants to leak out, and can if the seal is imperfect in any way. Gas leaks across the seal much more easily than oil, and in this design there is likely to be gas at the seal due to the problems mentioned above. Once in the hydraulic side, when the valve is opened to ease the vang, it expands many fold, blowing the hydraulic fluid back to the reservoir.

We had a Third Revelation which should not have been. On what is the 6th disassembly of the cylinders (3 times by Navtec and 3 by the rebuilder), it was discovered that the bore on the pistons which accepts the coupling spool (sealing the rod drilling through the piston) was machined oversize, resulting in diametric clearance for the O-ring of 0.017 on port and 0.009 on starboard, where 0.003 max was called for. It would literally fall out of the hole by its own (very light) weight. it was a difficult thing to find due to the assembly sequence, I’m quite happy the rebuilder noticed it. In machining new pistons to replace them, I’ve a pretty good guess how it happened.

In this application, if a sealed gas accumulator was used - where the fluid and gas are separated by a floating piston or bladder - the fluid could be degassed, bled, and stay that way. No easy way to convert this design. Had the cylinders been larger, the rod could have had two passages in it, allowing the fluid and gas ports to be at the top of their respective chambers making them self bleeding. Again no way to convert this design.

I’ve developed solutions to patch this up, cylinders are back on the boat, and we are hoping they will be less trouble now. The conventional mizzen vang has had none of these issues, though the same age and living the same life.

I still hate hydraulics.
 

mgs

canoeman
1,170
272
maine
I’m amazed you’re going through this again. Bout time to just keep the vangs and get a new boat isnt?
 

DDW

Super Anarchist
6,707
1,231
Wouldn't that be the wrong way 'round?

Until now, I've always been able to use the boat. The last year has had other distractions, so a good time to sort it. Many of our assumptions (mine, the rebuilder, and the current owners of Navtec) had been most probably wrong, so I am now armed with more knowledge. I built a few test fixtures allowing me to simulate conditions on the boat in my shop, developed methods to monitor pressures and fluid levels, more tools to test the bleeding procedure, machined improved the pistons and couplers, changing the mounting orientation which turns out to be signicant. For about the first 6 months the questions were piling up far quicker than the answers, things happening that seemed clearly impossible. A big change to that was understanding the nature of gas diffusion in hydraulic oil, then observations began to make more sense.

A conventional vang is a happy accident of design, where there are many things going your way. Turning it upside down turned a lot of things upside down. If starting fresh, I think I have a couple pf non-hydraulic design which might work. It takes some surprisingly heavy pieces to do it though.
 

mgs

canoeman
1,170
272
maine
No I’ll stand by my statement, although I am fully committed to my thoughts in your previous thread.

You’ve clearly established yourself on the path of making these vangs work, spending considerable time and resources to them. Why would you scrap them now? I haven’t heard much else about the boat so you can’t be too attached to it.

In all seriousness though, yeah it might be time to put these inverted vangs aside, or convert them.
 

DDW

Super Anarchist
6,707
1,231
I'm pretty attached to the boat, there's been plenty written about it. The vangs are just a detail.

On this rig, possibly the simplest alternative is to do away with the whole concept of a boom vang and fix the boom angle. Then adjust leech tension with a downhaul. Pretty simple, except it has to be duplicated for each of 3 reefed clews.
 

Caca Cabeza

Super Anarchist
Not to sound too harsh, but maybe that’s why the Navtec (or other) engineers specified that in the directions.

They had already tried to optimize.

Nothing is perfect. There is only “as good as we can get”.
 

DDW

Super Anarchist
6,707
1,231
Specified what?

The Navtec engineers thought this would work. They are being used as they designed them. Knowing what I now know, there are many design improvements that could have been made - but that is the price for going unconventional, you are on your own.
 

Laurent

Super Anarchist
2,322
1,986
Houston
I don't understand why you have a free surface between the oil and the gas. A hydraulic accumulator is doing exactly what you want to do as far as I know: a pressurized gas chamber maintains pressure on hydraulic fluid. In a standard hydraulic accumulator, the gas is inside a rubber bladder that expands and contract with the gas, depending on the hydraulic pressure, inside the accumulator. You do not have any direct contact between the gas and the hydraulic oil.
Would something like this work? See image in the left.

1655762707536.png
 

El Borracho

Verified User
6,832
2,789
Pacific Rim
What purpose does the pressurized gas serve? Why not oil for both actions. Seem to work for about a billion pieces of industrial equipment – and my little tractor.
 

DDW

Super Anarchist
6,707
1,231
I don't understand why you have a free surface between the oil and the gas. A hydraulic accumulator is doing exactly what you want to do as far as I know: a pressurized gas chamber maintains pressure on hydraulic fluid. In a standard hydraulic accumulator, the gas is inside a rubber bladder that expands and contract with the gas, depending on the hydraulic pressure, inside the accumulator. You do not have any direct contact between the gas and the hydraulic oil.
Would something like this work? See image in the left.

View attachment 524428
That is how it would preferably be done. The oil can then be vacuum degassed and stay that way. However that is an example of a closed hydraulic accumulator, with either a bladder, diaphram, or floating piston separating the gas and oil. This one was designed with a free surface. Converting these to a remote bladder accumulator, or adding a floating piston in the upper chamber was considered, but has many complications as a retrofit. With sufficient capacity it is too large to be mounted on the rig, so would need to be plumbed below decks (complicated by the rotating mast). Not to mention the difficulties in the vacuum degassing and bleeding operations, on deck during install.
 
Last edited:

DDW

Super Anarchist
6,707
1,231
An update here. I machined new pistons with the correct O-ring clearances and had the cylinders rebuilt yet again with the new ones. Bled them, charged them, put them on the boat, went back to Alaska (on the other boat) for a couple of months. Returned to find them holding gas pressure, no gas in the hydraulic side, no droop, no pumping 50 times just to get pressure. But: first the starboard side had a very slow drip from the rod seal, now the port side has started. This has no excuse, the rod is just a rod like any vang. Well, there is a small difference: On newer Navtec vangs the nut is apparently Delrin, on these aluminum and has to be because the rod is relatively large for the cylinder size, wall thickness on the nut are too thin for Delrin to survive.

I guess they will go back to the rebuilder again. I've only had one rod seal leak, back in about 2011 in LIS. Took another whack at reimagining them as mechanical without the hydraulics, and once again come up empty.
 

The Q

Super Anarchist
Generally an inverted vang is known as a GNAV, very common in the dinghy world.
hydraulics, I stay away from, too much complication , the excesses of temperature changes combined with salt water is not a mix.
 

mgs

canoeman
1,170
272
maine
Okay, so where is the leak actually? The rod seal on a standard vang would be the bronze gland just below the cap. Is the leak coming from the jaw/rod connection or from the body itself?

Thanks for making the thread easy to find.
 

Ease the sheet.

ignoring stupid people is easy
20,356
2,353
Is it possible to change the weight of the fluid?

A heavier/thicker fluid may have different "leaking" properties...
 

DDW

Super Anarchist
6,707
1,231
The leak is from the gland on the rod, not the oil jaw. Either the sliding seal or the o-rings sealing it, usually the o-rings are reliable so best guess is the sliding seal. There is a dust seal and nut in the way so I cannot differentiate between the two possibilities.

I did change to 30W from the 10W a couple of rebuilds ago hoping to improve the gas sealing performance.

I've not historically had the rod seals leak (as I mentioned, one side 10 years ago). The plan is to disassemble and polish the rod surface, though microphotographs of them look OK. Apparently you want a certain surface Ra for best seal performance, not mirror polished, not rough. I don't think they were polished last time apart, the rebuilder does not have the fixtures for this rod - because of the upside down construction the rod is in compression not tension, so was make bigger (1.25" in a -17 cylinder). The thread diameters at the ends are different than any other Navtec 1.25" rod. I kinda think Navtec cheaped out on the rods, I'd have expected they'd be something a little more exotic (and harder) than 316. But apparently 316 is what they typically used for rods. The rod is guided by the gland, not too much side load - just the weight of the cylinder and strut - but the gland is also hard anodized aluminum, considerably harder than the rod.

Yeah, I know it's called a gnav. I don't like hydraulics either, but in the dinghy world you don't need 13,000 lbs of vang pressure either.
 




Top