JCB 803 Saga

[awemawson]

Matthew, as far as I can tell the crack has propagated from the root of the female thread that that fitting I highlighted with a red arrow on a photo up thread.

My current hair brained thought is that IF the crack is within the length of the male threaded fitting I can just assemble it with loctite thread seal.

Currently the hydraulic seal is an o ring in a recess under the hex of the fitting, so the 3000 psi pressure is acting between the male and female threads, if that space is full of thread seal I might affect a cure.

Second hair brained thought if that fails is to make up a clamp that bridges the crack, and bed it down with that magic Titanium Putty I used building up the dipper bush

[mattinker]

Andrew, another thought, if it's the female that has split, bore oversize re-tap and loctite in a stainless ( or even mild steel) sleeve, threaded inside and out. The "O" ring would then be between the sleeve and the coupling and the pressure taken by the sleeve. A lot would depend on how much space there is around the coupling!

All the best! Matthew

[awemawson]

Yes Matthew all those thoughts have been going through my head over night 

There's actually not much space - the female thread is very close to the top of the block.

However, this morning I have managed to get a picture definitely locating the crack - I started the engine and very gently lifted the boom slightly until oil seeped rather than torrented out, and took this picture.

When the rain stops I'll crawl back under, pull out the fitting and measure how deeply it penetrates. I imagine that the female is somewhat deeper than the male on the fitting, so I can possibly make a longer adapter if the threads don't reach in deeper than the crack.

[mattinker]

Andrew, pin pointed!
I'm on tenter hooks! Hows it going?

All the best, Matthew

[awemawson]

Well I unscrewed the original fitting, and by measurement there seems to be only 2 threads 'beyond' the crack. So I came up with a cunning plan. Indeed the female looks to be threaded deeper in than the crack, so by making an adapter with a longer male part, just maybe I can seal it all up with Loctite  Hydraulic Thread Seal.

The hole in the block is threaded 7/8 UNF - 14 tpi and it is made from 28 mm hex bar. I decided to make a male / female adapter rather than try and copy the part that the hose goes on.

So first job - find some 28 mm hex - well I ended up making some from 32 mm stainless bar

[awemawson]

Then cut the male and female ends to form the adapter.

I decided to use those 'bonded washers' to avoid threading too close to the shoulder

[awemawson]

Then with loads of Loctite Hydraulic seal it all went back together.

I'll leave it overnight to set fully as I reckon it's only a 25% chance it'll work anyway and no point in putting 3000 psi on it unless the gunge is fully cured

[mattinker]

With baited breath!

[awemawson]

Yes - me too Matthew it needs 18-28 hours for full strength (Loctite 542)

Mean while I've bee trying get to grips with exactly what is inside this ridiculously expensive block of aluminium. In the attached picture the area marked '1' inside a dotted line represents the block:

1F are dual non-return valves

1B is the Digging and Tracking Main Relief Valve

1D is the Pressure Reducing Valve


So presumably high pressure oil from Pump 3 and Pump 2 pass straight though the block onward to various spool valves, but they both can vent back to the tank via the '1F' non return valves, and the '1B' Digging and Tracking Main Relief Valve, onward through the oil cooler '24' and  the Return Filer 17 (which has a bypass).

Now to the bit I don't understand: there is a dotted line implying a control dependency, between 1B and 1D, and the input to 1D comes from Pump 1 via '23A' which is a 'Pressure Maintenance Valve' and this is associated with a 'Solenoid Lock Valve 21' that I know is operated by a micro-switch in the left hand hand rest preventing the servo circuits working unless the arm is folded down.

So everything in the block '1' could be replicated with off the shelf bits if I understood the linkage between 1B and 1D. Unscrewing the 1B Digging and Tracking Main Relief Valve there is visible a peg protruding from the device on the other side of the block that I suspect bears on the plunger of 1B opening it up. So I think that that is the physical linkage - there is a bit of verbage in the manual thus:

"The main relief valve for P1 is located on the inlet side of the four section control valve block and is set to open at 172 bar (2500 psi). The main relief valve for P2 and P3 is located in the Main Relief Valve Block (my problem child!) and has dual setting of 207 bar (3000 psi) and 175 bar (2550 psi) depending on the hydraulic load on the engine. The valves are designed to protect the pumps against over pressurisation when a selected service is stalled or the ram reaches the end of it's travel"

[mattinker]

I agree, in the "verbage" I  read that  these are mechanical interlocks to prevent loss of pressure or damage through back pressure when two valves dump at the same time. The three circuits are running at different pressures but they dump through a common filter circuit. I imagine that there is some other security device so that if one is dumping another can dump elsewhere. How many of these circuits can function at the same time? What about dumping through separate circuits or adding non return valves to avoid the need for an interlock?

Thoughts! Regards, Matthew

[awemawson]

So what was the result?

Well, rather like the Parson's Egg - good in parts. The leak is very much reduced, the photo below was taken with the main boom ram fully extended so the the pressure was maximum and the relief valve was working. So the theory worked but the practise didn't 

Frankly if it was like this you'd not notice the leak, but that crack is quite long and can only propagate further and the leak will undoubtedly go torrential just at a critical moment. So I think I'm going to have to stump up for a new block 

I've not yet checked the JCB actually have one in stock, just got a price. Maybe if they are out of stock I'll give it another go but I don't want an impending disaster looming!

[lordedmond]

Andrew

Now this may be a dumb idea but could you not clamp a patch/ plate on the outside with some suitable gunk on it at the least it would stem the flow in case of a failure

With your cnc machines you may be able to cut a o ring grove into the plate for added seal

Just thinking out loud

Stuart

[David Jupp]

Since crack is only a couple of threads down and female thread deeper than male - is there scope to counterbore beyond the crack?  Move the whole fitting inboard a small distance?

Trouble with Aluminium is that there is no 'knee' in the fatigue curve - on cyclic duty it will fail eventually, no matter how low the stresses.  If re-making the whole block, consider steel. 

If you have an industrial hydraulics design company nearby, they probably make custom manifold block routinely - could be another option if you have to replace.

[awemawson]

Stuart,

I had considered (see post #50 above) making a clamp plate that straddled the block with bolts down the side of the block into an under plate, and cutting a slight rebate where the leak is. Then bedding it all down using the magic Devcon Titanium Putty that I used to set the main dipper bush, which is pretty amazing stuff.

All these things are bodges though and I'm not too keen on bodges.

BTW it almost looks as though the block is made of layers bonded together, but I can't imagine it really it 

[awemawson]

David,

The extended adapter I made goes in as far as is possible without fouling a counter bore coming in from the side, so it's not really feasible to sink in further.

I suppose it's a design error, as this appears not to be an uncommon failure mode - the block should have had more meat in it above this particular threaded hole.

I suspect that a custom manifold would be a close match in price to the proper one by the time they'd done the design and machining. But thanks for the suggestion.

[mexican jon]

All these things are bodges though and I'm not too keen on bodges.

I don't see them as bodges  A bodge is something that you do with no intention of putting it right when time or parts allow  What you you would be doing is a temporary repair 

That's the way I was taught during my early years of keeping machinery running 

[David Jupp]

Pity about the side bore! - maybe a groove and cross drill in an extended fitting?

Manifold block - though not cheap, getting a block made (in steel?) by somebody that routinely makes manifolds might not be too horrendous.  Where possible they will adapt standard general purpose (off the shelf) manifold blocks to keep the cost down, you might be lucky with any critical dimensions of the block....

I'm sure you will find a way.

[awemawson]

If I manage to escape from family commitments I'm going to make that straddling 'bandage' today. Top piece is a simple plate with holes for the bolts. Lower one needs rebates to accommodate the three sticking out blanking plugs, the four tapped holes for the mounting bolts.

[mattinker]

I hope I'm wrong, but I can't see your straddling block working. Although you've stemmed the flow, the leak will eventually get up to working pressure. If your strap is very strong and you machine a groove as Stuart suggests in both the strap and the block, you might get sealing from the outward pressure pushing the "o" ring into the joint.

Now that you've shown exactly where the leak is, you've got below it with the extended fitting, how about sleeving as well? I doubt that you have room to put an "O" ring at the bottom of the hole were it would do most good!

More photos of the block once it's out please!

Frustrated for you, regards Matthew

[awemawson]

Matthew,

I've decided to bite the bullet and buy a new block, as I agree the fault can only get worse. However, I've made up the straddling clamp as a temporary measure, bedded it with Devcon Titanium Putty, and lightly tightened the screws. When it's hard in the morning I will tighten them down properly.

Have some pictures - you'll notice I had to move one of the clamping bolts as it fouled a fitting. The recesses are to clear various plugs in the base of the block

[mattinker]

So the worst has come to the worst! having made the decision to buy a new one, why not pull it pre-heat the block and TIG it? Tig the split and weld a reinforcing plate over the top, you might just resolve the design fault!

As things are, you at least have time to find another one!

Regards, Matthew..

[awemawson]

It's likely to take a while for the new block to arrive - I'll order it Monday first thing - (well second thing actually as I need to order some Electronic sheep tags!)

I've patched it as I have a pile of road planings that I need to shift asap.

[Pete.]

Buying a new block is a good way of ensuring the temporary repair out-lives the owner

[mattinker]

It's easy to make suggestions, it's another to deal with the problem and the associated imponderables like moving pig poop^or whatever!

all the best, Matthew.

[awemawson]

Well the temporary repair has at least allowed me to shift 5 ton of road planings and squash them down. It did leak a bit but nothing like it had been, and it didn't gush anything like the bucket hose I caught in the tree  Easy hose change and I might even have the bits to crimp one.

I ordered the replacement block which should be here next week - only £1126.27 with the various seals and carriage and VAT