B7800 front axle ball bearing question

[Neil]

After recently buying a 20ish year old B7800 I pulled the mower deck only to discover bad sounds coming from the left side of the front axle. When I pulled the cover, I noticed overflowing sealant that wasn't present on the other side, so maybe this has been opened before.

The bottom bearing had completely disintegrated and bits of the ball cage were going through the bottom gears. With the balls out of the race, the pinion shaft could move plenty far enough to open up clearance for race bits to go through the gears. The pinion was damaged enough I chose to order a new one, but the ring gear looks OK. The pinion shaft had dropped down and the pinion teeth were only engaging the outer edges of the ring gear teeth.

Obviously bits of metal have been circulating through the axle. The big bits seem to have mostly stayed in the lower portion of the left side. While waiting for parts, I decided to open the right side to see how things looked over there. What I found is a little grittiness in the bearings, but no chunks. All bearings will be flushed until the spin freely.

The really weird part is that the two sides have different bearings in front of the ring gear. The left side is a NSK pressed onto the hub shaft. The right side is a KOYO that is a close, but free fit on the hub shaft. Both are numbered 6305. One of these must be wrong. Pressed fit seems like that should be right, but I don't know. Maybe it doesn't matter for a low speed application like this.

Advice and comments are welcome.

 

[TheOldHokie]

After recently buying a 20ish year old B7800 I pulled the mower deck only to discover bad sounds coming from the left side of the front axle. When I pulled the cover, I noticed overflowing sealant that wasn't present on the other side, so maybe this has been opened before.

The bottom bearing had completely disintegrated and bits of the ball cage were going through the bottom gears. With the balls out of the race, the pinion shaft could move plenty far enough to open up clearance for race bits to go through the gears. The pinion was damaged enough I chose to order a new one, but the ring gear looks OK. The pinion shaft had dropped down and the pinion teeth were only engaging the outer edges of the ring gear teeth.

Obviously bits of metal have been circulating through the axle. The big bits seem to have mostly stayed in the lower portion of the left side. While waiting for parts, I decided to open the right side to see how things looked over there. What I found is a little grittiness in the bearings, but no chunks. All bearings will be flushed until the spin freely.

The really weird part is that the two sides have different bearings in front of the ring gear. The left side is a NSK pressed onto the hub shaft. The right side is a KOYO that is a close, but free fit on the hub shaft. Both are numbered 6305. One of these must be wrong. Pressed fit seems like that should be right, but I don't know. Maybe it doesn't matter for a low speed application like this.

Advice and comments are welcome.

Dimensionally a 6305 is a 6305- 25 x 62 x 19

Koyo and NSK are both quality bearings.

Your slip versus press fit is someplace other than the bearing. There are applications where that may be necessary for assembly purposes.

Dan

 

[GeoHorn]

Neil…. How do you know your 20-ish year old machine hasn’t had someone in there before You that substituted non-OEM bearings on one side..?

One of the nice things about owning a Kubota is….. they still support their old stuff. You may not like to pay shipping for the correct OEM part…. but it’ll be the correct part!

Otherwise… the actual brand of the bearing usually is insignificant…NSK…KOYO…. etc etc….as long as the Part No. is the same.

Another issue is: An old worn out bearing can begin to rotate it’s race inside the casting…..and wear-away the casting…. leading one to believe it is a “slip fit”…when originally it was a “Press-fit”.

 

[North Idaho Wolfman]

Kubota parts are never press fitted, except in some very rare instances and this is not one of them.
They are normally held in place with e rings.
If you have something that is out of whack then I would investigate further.

 

[torch]

Dimensionally a 6305 is a 6305- 25 x 62 x 19

I think that might be a typo? I believe the 6205 is 17mm wide, not 19mm.

A minor point, since width has no bearing on the slip fit issue (sorry, couldn't resist the pun).

I'd be looking very closely at the left side shaft surface finish. Since that is the side that crumbled and assuming the right side slip fit is representative of the original state of the left side, it is possible that the bearing jammed, forcing the shaft to rotate within the bearing's bore. Galling of the shaft or trapped debris may now be causing the stiction. If damage is minor, careful cleanup with a pair of toolroom stones (NOT a honing stone) may be required.

GeoHorn also raises the very real possibility that the outer race has spun within the casting, so a careful examination of the casting bore is also in order.

If damage is minor, you might consider one of the Locktite bearing retainer products (https://www.henkel-adhesives.com/uk...taining-with-loctite-retaining-compounds.html)

If damage to the bore is substantial, a machine shop can clean it up to a specific oversize and fit a bearing tolerance ring (basically a corrugated shim that takes up the space between the bearing and the oversized bore -- EG: https://www.mcmaster.com/products/bearing-shim-rings/for-bearing-od~62-mm/)

 

[TheOldHokie]

I think that might be a typo? I believe the 6205 is 17mm wide, not 19mm.

A minor point, since width has no bearing on the slip fit issue (sorry, couldn't resist the pun).

I'd be looking very closely at the left side shaft surface finish. Since that is the side that crumbled and assuming the right side slip fit is representative of the original state of the left side, it is possible that the bearing jammed, forcing the shaft to rotate within the bearing's bore. Galling of the shaft or trapped debris may now be causing the stiction. If damage is minor, careful cleanup with a pair of toolroom stones (NOT a honing stone) may be required.

GeoHorn also raises the very real possibility that the outer race has spun within the casting, so a careful examination of the casting bore is also in order.

If damage is minor, you might consider one of the Locktite bearing retainer products (https://www.henkel-adhesives.com/uk...taining-with-loctite-retaining-compounds.html)

If damage to the bore is substantial, a machine shop can clean it up to a specific oversize and fit a bearing tolerance ring (basically a corrugated shim that takes up the space between the bearing and the oversized bore -- EG: https://www.mcmaster.com/products/bearing-shim-rings/for-bearing-od~62-mm/)

Yes thats a typo and I think its safe to say something more than the bearing is damaged.

Dissasemble, inspect, and measure the parts.

Dan

 

[GeoHorn]

You know it’s a bad day when the Wolfman posts things like “Kubota bearings are never press fitted….. (except when they are)….” and ”…when something is out of whack then …investigate further.”

I’m only Ribbing-you, Wolfman! You KNOW I admire/appreciate you!

 

[Neil]

Neil…. How do you know your 20-ish year old machine hasn’t had someone in there before You that substituted non-OEM bearings on one side..?

One of the nice things about owning a Kubota is….. they still support their old stuff. You may not like to pay shipping for the correct OEM part…. but it’ll be the correct part!

Otherwise… the actual brand of the bearing usually is insignificant…NSK…KOYO…. etc etc….as long as the Part No. is the same.

Another issue is: An old worn out bearing can begin to rotate it’s race inside the casting…..and wear-away the casting…. leading one to believe it is a “slip fit”…when originally it was a “Press-fit”.

In this case, the slip fit is on the good side and the interference fit is on the trashed side.
To be clear, I'm referring to the fits on the hub shafts, not in the cast housings.

 

[TheOldHokie]

In this case, the slip fit is on the good side and the interference fit is on the trashed side.
To be clear, I'm referring to the fits on the hub shafts, not in the cast housings.

Jaw boning isn't going to resolve your issue. Pull and inspect the shaft. A typical press fit on a C3 bearing would be a scant ,001 interference. You know the bearing bore so all you have to do is measure the shaft. You will need a 1" micrometer.

Dan

 

[torch]

In this case, the slip fit is on the good side and the interference fit is on the trashed side.
To be clear, I'm referring to the fits on the hub shafts, not in the cast housings.

As Dan points out, the difference may be little more than one or two thou. A bit of trapped debris from hardened bearing components can easily raise a divot or ridge that high, making a slip fit feel like a press fit.

Since it appears you have already disassembled both sides, you have a go-no go gauge in your hands: Have you tried the known good right bearing over the left side shaft? Or vise-versa?

 

[torch]

Also: when you are inspecting the suspect shaft, make sure it's not bent. Roll it on the flattest surface you can find -- a surface plate if possible -- or chuck it up on a lathe between centres, or set it on a pair of V-blocks, etc.

 

[Neil]

Today I took the tight side to a friend with a press. No joy, so we cut the bearing off. The shaft underneath was partially galled. After a stop at the local Kubota dealer for a new bearing, I chucked the hub up in my lathe and polished the bearing surface down to achieve the same barely free fit with the new bearing as the other side had.

FWIW, the new bearing is an NSK, and it wouldn't fit on the same shaft where the Koyo was free. After polishing the galled shaft, to achieve the free fit with the NSK bearing, it mic'd a couple tenths smaller than the Koyo shaft.

 

[TheOldHokie]

Today I took the tight side to a friend with a press. No joy, so we cut the bearing off. The shaft underneath was partially galled. After a stop at the local Kubota dealer for a new bearing, I chucked the hub up in my lathe and polished the bearing surface down to achieve the same barely free fit with the new bearing as the other side had.

FWIW, the new bearing is an NSK, and it wouldn't fit on the same shaft where the Koyo was free. After polishing the galled shaft, to achieve the free fit with the NSK bearing, it mic'd a couple tenths smaller than the Koyo shaft.

What were the neasuremenths to tenths. of both shafts. Is your mic reliably calibrated?

Assuming both bearings are normal precision the manufacturing tolerances for boundary dimensions are the same. The diameter tolerance for the inner ring is strictly undersize (.00000/- .00039). In other words worst case deviation is ,00039 and statistically half that.

Dan

 

[torch]

FWIW, the new bearing is an NSK, and it wouldn't fit on the same shaft where the Koyo was free. After polishing the galled shaft, to achieve the free fit with the NSK bearing, it mic'd a couple tenths smaller than the Koyo shaft.

That is, ummm, unexpected !

I wonder if you ended up with a counterfeit NSK. Apparently counterfeit bearings are a thing and infiltrating supply chains. Where did you get it?

 

[TheOldHokie]

That is, ummm, unexpected !

I wonder if you ended up with a counterfeit NSK. Apparently counterfeit bearings are a thing and infiltrating supply chains. Where did you get it?

"After a stop at the local Kubota dealer for a new bearing,"

I expect Kubota buys bearings direct from NSK. What surprises me is its not a Koyo bearing.

I would be wondering far more about the actusl dimensions of the two used shafts one of which we know for sure is no longer factory.. i would also be looking at a WSM to see what it says about fit.

Dan

 

[Neil]

My measurements were made with a 2 year old Mitutoyo 293-340-30 Digital Micrometer, which has never been dropped or recalibrated. I'll try to make a more thorough comparison later and report back. For now, I can add that as soon as I opened the NSK bearing I tried it on the shaft where the Koyo was free. The NSK absolutely would not free fit on that shaft.

 

[torch]

"After a stop at the local Kubota dealer for a new bearing,"

I expect Kubota buys bearings direct from NSK. I am more surprised its not a Koyo bearing.

I would be wondering far more about the actusl dimensions of the two used shafts one of which we know for sure is no longer factory.. i would also be looking at a WSM to see what it says about fit.

Dan

I'm not saying it is counterfeit, just looking for a possible explanation. NSK is one of the top bearing manufacturers in the world -- they can spec their products to tenths of tenths for crissakes. I wouldn't hesitate to trust an NSK bearing and Koyo is no slouch either. I'm sure either manufacturer can trace their metrology to one or two steps from NIST.

I can't. I doubt most hobbyists can. And even if the tool is accurate, in the world of tenths just the warmth of holding the micrometer in your hand too long can affect the outcome. So at best any numbers would be comparative of one shaft to the other, but not reliably to marks on a platinum-iridium bar stored in a climate controlled Washington vault.

But we shouldn't need that -- the bearings themselves should be usable as a go-no go gauge. They should be so close to identical as not to matter. That won't tell us what the dimension is, or if the shaft is ovaled or bent but will confirm that the maximum dimension is smaller than the bore of the bearing.

So, when he says the NSK bearing won't fit on the right side shaft, even though the Koyo is a slip fit, I have to believe the NSK bore is smaller than the Koyo bore. And that just doesn't make sense. IF it's a real NSK. (I am making a big assumption here that the Koyo was the original, based on the circumstantial evidence at hand).

I missed the bit about sourcing the bearing from a Kubota dealer and while I'm sure you are right that Kubota probably buys in bulk direct from a manufacturer for production and while it is true Kubota might switch suppliers from time to time, they won't be buying them in little blue cardboard retail boxes. Kubota USA may well be sourcing them locally, so who knows were in the supply chain the NTK bearing was injected.

Counterfeit parts are a huge problem and can creep in the most unlikely of places. I remember a few years back when Air Force One was found to have counterfeit parts fitted (bolts not bearings, IIRC) and it's hard to imagine a harder nut to crack than that.

So I'm not saying that is the only possible explanation, just wondering if it is. Hey, who knows? Maybe the one that failed was a counterfeit too.

 

[TheOldHokie]

My measurements were made with a 2 year old Mitutoyo 293-340-30 Digital Micrometer, which has never been dropped or recalibrated. I'll try to make a more thorough comparison later and report back. For now, I can add that as soon as I opened the NSK bearing I tried nit on the shaft where the Koyo was free. The NSK absolutely would not free fit on that shaft.

Then it needs to be recalibrated if you want to measure absolute values to tenths. At a minimum you need to check it against a 1" gage block. I have the mechanical version of that Mit which I trust more.

I also know that regardless of brand those bearings all have an inner ring manufacturing tolerance of 25.000/25.001. For the metric challenged thats .00039". Bearings are the last thing I would suspect in your situation.

What I dont know is how Kubota engineered that fit abd how tight they held the shaft. Standard design practice for mountIng deep groove radial ball bearings is to fix one but not both of the rings. If the outer ring is a slip fit in the housing the inner ring would be a press fit on the shaft. Depending on the application the interference can be as small as a couple tenths or as much as a thou.

Do you have a WSM?