L48 split for HST repair

[MHarryE]

Any idea as to how the bearing came loose? How did you diagnose it prior to removal? Not surprising the parts made it into the filter doing no damage as the cradle is in the case and not in the high pressure loop so the parts would stay in the case if too large or would get flushed out and would need to go through the filter before once again entering the system. Cradle bearings need to be very low friction so the swash plate can pivot easily. All the units I work with have needle bearings but I’ve never been inside a low cost CUT transmission.

 

[radios1]

Here's a WSM page showing where the cradle bracket goes in the unit. After bolting it in I solvent cleaned the surfaces and put the port block back on with Permatex Ultra Grey (no gasket). HST is ready to reinstall.

after blowing up the picture 400% it still looks like brown paper over the bearings, you can even see it crinkled..

 

[MHarryE]

Any idea as to how the bearing came loose? How did you diagnose it prior to removal? Not surprising the parts made it into the filter doing no damage as the cradle is in the case and not in the high pressure loop so the parts would stay in the case if too large or would get flushed out and would need to go through the filter before once again entering the system. Cradle bearings need to be very low friction so the swash plate can pivot easily. All the units I work with have needle bearings but I’ve never been inside a low cost CUT transmission.

Ремонт гидронасоса хода Rexroth A4VG56 катка Komatsu - YouTube

I’m afraid that after posting I searched YouTube for one of my favorite hydros and came up with this. The video shows fretting on the cradle from the needle bearings. Interesting look at the hydro guts.

 

[radios1]

Ремонт гидронасоса хода Rexroth A4VG56 катка Komatsu - YouTube

I’m afraid that after posting I searched YouTube for one of my favorite hydros and came up with this. The video shows fretting on the cradle from the needle bearings. Interesting look at the hydro guts.

look here. NSK - Fretting

 

[rbargeron]

after blowing up the picture 400% it still looks like brown paper over the bearings, you can even see it crinkled..

The pic was an online image. The new part looks the same - its coated with light brown grease (for that crinkled look) and sealed in plastic wrap. I cleaned it, coated it with UDT and bolted it into the case. The bearing faces are bronze - high copper, low zinc, maybe a couple percent silicon for hardness.

 

[rbargeron]

Any idea as to how the bearing came loose? How did you diagnose it prior to removal? .............Cradle bearings need to be very low friction so the swash plate can pivot easily. All the units I work with have needle bearings...........

Rolling bearings trade low friction for high contact stresses - fretting (same as potholes - shear failure below the surface). Shell-type bearings spread out the load better than rollers. But they're not perfect either.

With everything immersed in oil, the unit will run a long time even with worn cradle bearings - as long as they stay in place. In this case the head of the shell's retaining screw was gone (worn away).

My guess is root cause was debris in the fluid sometime in the past. I don't know the machine's history - prior owner said the dealer serviced it. When I opened it up, one bearing was out of position in the corner blocking swash plate travel.

With a new bracket and bearings it all glides freely now - ready for a couple thousand more hours.

 

[arto98607]

......................................... ready for a couple thousand more hours.

Interesting rebuild story, thank you!

How many hours in the unit now?

 

[rScotty]

I can never resist talking about sealers and gaskets. So anyone who finds that boring this is where to tune out. When I assemble without a gasket I get really picky about the RTV. I also like the Permatex Ultra Grey and use it a lot. I also like Permatex fuel resistant and hi heat silicon RTVs. Great value for the cost. For me, the trick to using them is to apply a thin layer to both sides and then not assemble until it is very tacky and nearly cured. You don't want to force strings of RTV out of the joint - especially not in a HST.

For the ultimate silicon goo, there is always Dow Corning 3145 Grey. It's an industrial product not often sold retail - but it is the "go-to" grey silicon for high$$ hi-tech and for aerospace applications. I get it from Pilot Gear Online.com....$33 for a 3 oz. tube and worth every penny. Handled carefully & resealed with care it will last five years or more. Very sticky, capable of high temp and pressure. it cures to a tacky surface quickly & then takes overnight for final strength. It is also non-toxic, and doesn't affect sensitive surfaces since it cures by a different catalytic process so that it doesn't outgas any acid fumes (that distinctive organic acid smell). 3145 has no odor and can be safely applied and shaped with a wet finger.
Great stuff, but crazy expensive.

Looking at the pumpshaft PDF you posted, that's just the kind of place I'd use 3145 if available, but I haven't had or seen any problems with Permatex Ultra Grey either. Very thin layer, though....and wait for tacky before assembly.
I'm currently doing a set of vintage Norton motorcycle engine/transmission cases that way.
rScotty

 

[rScotty]

The pic was an online image. The new part looks the same - its coated with light brown grease (the krinkled look) and sealed in plastic wrap. I just bolted it into the case. The bearing faces are bronze - high copper, low zinc, maybe a couple percent silicon for hardness.

Thanks for posting that explanation, I was beginning to join radios1 in worrying if it really could be paper.
BTW, I think that is an area where I would prefer shell bearings to roller bearings... as long as there was a dedicated clean oil feed to the sliding surface. Given good oiling, I really can't understand why rollers would have any advantage in a high load oscillating movement like that.
rScotty

 

[mopardude318]

I can never resist talking about sealers and gaskets. So anyone who finds that boring this is where to tune out. When I assemble without a gasket I get really picky about the RTV. I also like the Permatex Ultra Grey and use it a lot. I also like Permatex fuel resistant and hi heat silicon RTVs. Great value for the cost. For me, the trick to using them is to apply a thin layer to both sides and then not assemble until it is very tacky and nearly cured. You don't want to force strings of RTV out of the joint - especially not in a HST.

For the ultimate silicon goo, there is always Dow Corning 3145 Grey. It's an industrial product not often sold retail - but it is the "go-to" grey silicon for high$$ hi-tech and for aerospace applications. I get it from Pilot Gear Online.com....$33 for a 3 oz. tube and worth every penny. Handled carefully & resealed with care it will last five years or more. Very sticky, capable of high temp and pressure. it cures to a tacky surface quickly & then takes overnight for final strength. It is also non-toxic, and doesn't affect sensitive surfaces since it cures by a different catalytic process so that it doesn't outgas any acid fumes (that distinctive organic acid smell). 3145 has no odor and can be safely applied and shaped with a wet finger.
Great stuff, but crazy expensive.

Looking at the pumpshaft PDF you posted, that's just the kind of place I'd use 3145 if available, but I haven't had or seen any problems with Permatex Ultra Grey either. Very thin layer, though....and wait for tacky before assembly.
I'm currently doing a set of vintage Norton motorcycle engine/transmission cases that way.
rScotty

I am assembling a hydraulic pump and the gaskets determine the gear to case clearance. The first set of gaskets I made were .004” thick and they compressed.002” when I tightened the pump body studs, the clearances were too tight and the internals bound up. Would not turn. I may double up on my gasket material to get the correct clearance.

Would you happen to know the compressed thickness of that 3145?

 

[rScotty]

I am assembling a hydraulic pump and the gaskets determine the gear to case clearance. The first set of gaskets I made were .004” thick and they compressed.002” when I tightened the pump body studs, the clearances were too tight and the internals bound up. Would not turn. I may double up on my gasket material to get the correct clearance.

Would you happen to know the compressed thickness of that 3145?

No, I don't. But since it cures to a rubbery solid I would imagine that compressed thickness is going to depend a lot on compression force. And I wouldn't trust any real measureable thickness of gasket goo with high hydraulic pressure. Frankly, as much as I like the stuff, I would be careful in that application. Much better to remove some shims if you have any on the gear shafts of the pump... but if not, here's my thinking...

In a high pressure hydraullic pump I'd go with a single thin high quality cross-woven fiber gasket or maybe two of them since yours are so thin. If I used any goo, it would be applied just as thin as I could get it. Less - way less - than the thickness of a single coat of paint would be what I'd aim for. Just as thin as you can do it. You can do that, it's not impossible. And I see what you are wanting...but if it were mine I'd get the clearance by stacking gaskets and use only enough gasket goo to fill in the highs and lows.
luck,
rScotty

 

[rbargeron]

This story scares me.
I am looking for a low time L48.

Of the responses to this thread, this one is important to anyone considering a Kubota with HST. The unit's reputation for very low failure rate is undeniable. In fact, many dealers are unfamiliar with their internals, perhaps from rarely if ever needing to get inside one.

In this case the tractor has 2200 hours on it. Its cradle bearings didn't fail abruptly - looking at them it's clear they've been wearing for a period of time. According to the prior owner the machine showed no symptoms until it suddenly wouldn't reverse. We now know now the wear was bad enough for the bearing to come loose, blocking swash plate travel.

So what's a take-away for the potential tractor buyer? In pre-purchase inspection, its nearly impossible to know the condition of the cradle bearings without opening the HST housing. Even with 50% thinning, performance could still be completely normal. And nobody's going to take it all apart just to check.

Oil service history is a big deal on a used HST machines, especially those with cabs or under-frames with high tear-down cost. The next step up would be a trace wear metals test of the trans fluid. But if the filter and fluid are changed the evidence may disappear. Dick B

 

[MHarryE]

The Bosch Rexroth YouTube video I showed with a tear down of an AA4VG56 hydro - the unit is older now but was considered the elite hydro transmission in its time used by virtually all the high end hydraulic excavators. Fretting of cradle bearings on these units is unusual but any roller bearing subjected to dynamic loading when the bearing is not rotating is subject to fretting. My memory shelf at home has pieces of rotating groups I’ve inspected - from a 30,000 hour unit with no visible wear and a 113 hour unit from a machine with a plugged oil cooler with melted piston slippers so the ball ends of the pistons were ripping into the swash plate. The YouTube video was a Komatsu excavator. Caterpillar uses the same system. They are the 2 largest Heavy Construction Equipment manufacturers in the world, both rely on quality to keep their name, and both use use Bosch Rexroth hydros on their premier product, the reason I selected their video.

 

[RickB]

Hot, dirty oil will kill any hydro no matter whether it is a Rexroth or a low end Hydrogear.

 

[fried1765]

Of the responses to this thread, this one is important to anyone considering a Kubota with HST. The unit's reputation for very low failure rate is undeniable. In fact, many dealers are unfamiliar with their internals, perhaps from rarely if ever needing to get inside one.

In this case the tractor has 2200 hours on it. Its cradle bearings didn't fail abruptly - looking at them it's clear they've been wearing for a period of time. According to the prior owner the machine showed no symptoms until it suddenly wouldn't reverse. We now know now that wear was finally severe enough the bearing came loose, blocking swash plate travel.

In pre-puchase inspection, its nearly impossible to know the condition of the cradle bearings without opening the HST housing. Even with 50% thinning, performance could still be completely normal.

Oil service history is a big deal on a used HST machines, especially those with cabs or under-frames with high tear-down cost. Dick B

If available, an oil analysis, prior to purchase, might be a good idea?

 

[sixdogs]

If available, an oil analysis, prior to purchase, might be a good idea?

That is a great idea on a used tractor. I wouldn't do it on my own equipment but sure would on a used purchase. Seems it would save a lot of potential pain.

 

[rScotty]

Of the responses to this thread, this one is important to anyone considering a Kubota with HST. The unit's reputation for very low failure rate is undeniable. In fact, many dealers are unfamiliar with their internals, perhaps from rarely if ever needing to get inside one.

In this case the tractor has 2200 hours on it. Its cradle bearings didn't fail abruptly - looking at them it's clear they've been wearing for a period of time. According to the prior owner the machine showed no symptoms until it suddenly wouldn't reverse. We now know now that wear was finally severe enough the bearing came loose, blocking swash plate travel.

In pre-puchase inspection, its nearly impossible to know the condition of the cradle bearings without opening the HST housing. Even with 50% thinning, performance could still be completely normal.

Oil service history is a big deal on a used HST machines, especially those with cabs or under-frames with high tear-down cost. Dick B

One of the things that interested me is that the swash plate itself didn't take much damage even after all the work it did on the cradle and the cradle bearing retaining screw. Lucky one there.....
& was there significant off-center loading on the swashplate as one cradle wore more than the other?

Dick, looking back on your other thread with this L48 back in 2013, do you think that the cradle bearings were involved back then or was that issue strictly about the regulator being whacked? Curious on your thoughts about causes.
rScotty

 

[rbargeron]

One of the things that interested me is that the swash plate itself didn't take much damage even after all the work it did on the cradle and the cradle bearing retaining screw. Lucky one there.....
& was there significant off-center loading on the swashplate as one cradle wore more than the other?

Dick, looking back on your other thread with this L48 back in 2013, do you think that the cradle bearings were involved back then or was that issue strictly about the regulator being whacked? Curious on your thoughts about causes.
rScotty

rScotty: I should have been more clear in post #31 above. This weak reverse condition has existed since 2013. Back then, to avoid the project of dismantling the machine I tweaked the regulator and servo piston to favor reverse - so reverse worked again - but it's never been 100%. I've put the current job off for years but the cradle bearing damage was already done. I just didn't realize it, having only diagnosed it from the outside.

When I got it apart I found the swash plate had severely rubbed into both bearings, and both cradle retaining screws. One was a bit worse than the other - both more than 70% gone - very little remaining thickness.

No idea why one side was worse than the other - the servo piston pushes from one side so there are uneven operating forces on the swash plate.

The failure came when the screw on one side let go, the bearing crumpled and jammed at the servo cylinder, blocking it from going past half travel. The tractor abruptly lost full reverse and the condition hasn't changed since.

Reverse doesn't need to be fast so the tractor has been operating with damaged cradles for several years. It's an extra tractor around here and doesn't get used much. It usually has a brush grapple mounted so with the hydro fixed it may get more use.

On the question of no apparent damage on the swash plate, the bearings are a much softer material. The flat head retaining screws are steel, but also softer than the swash plate. Hence no scratches.

I'm not at peace with the design to retain the bearings. Each has one small screw. If the oil is clean they see very little shear load, but any foreign material that increases friction can threaten the unit's longevity.

 

[dfkrug]

Here's a picture of a new bracket with bearings.

Interesting saddle bearings, Dick. The way it is secured is a little surprising...Philips head machine screws.

Here a is comparison saddle bearing on a McCormick GX50H (Landini), a similar sized 50hp tractor with servo-controlled swashplate. Most of the exotic material (bronze?) seems to have worn off mine. Pins retained the bearings, and were unlikely to back out. The maker is Bondioli Pavesi, if anyone cares.

Thx for posting. Any pix of the worn out parts?

 

[rbargeron]

Interesting saddle bearings, Dick. The way it is secured is a little surprising...Philips head machine screws.

Here a is comparison saddle bearing on a McCormick GX50H (Landini), a similar sized 50hp tractor with servo-controlled swashplate. Most of the exotic material (bronze?) seems to have worn off mine. Pins retained the bearings, and were unlikely to back out. The maker is Bondioli Pavesi, if anyone cares.

Thx for posting. Any pix of the worn out parts?

Hi Dave - great to hear your voice ! I was hoping you'd see this thread and comment, maybe curate a bit with you knowledge of Kubota HSTs. btw I almost bought a Landini orchard tractor years ago - big power in a small package.

Here are a couple pics of the worn parts. The bearing still in place shows a narrow edge of original thickness and remnants of bronze facing. Looking close there are scuff marks on the head of its screw - and it is loose. The other bearing is trashed, screw head gone - threaded end is in the hole and turns freely.

The HST in the L48 is the same one used in Lxx10 models. The L48 is over twice as heavy as a L3010. Think about the pressure peaks reversing a 4-ton machine vs one only 2 tons. I'd say the cradle bearings have an easier life in the L3010.