Craftsman TS Arbor Replacement - Going In Through The Side Works Fine

[DerbyDad03]

The standard procedure for replacing the arbor bearings (or the arbor itself) on a 60's/70's vintage Craftsman table saw is to either flip the entire saw upside down or fight gravity and wrestle the innards out through the back after disconnecting the tilt and height shafts, the rear trunnion and a bunch of other parts.

There's no way I can (or want) to flip the saw over. I'd have to remove the fence rails, the router table extension, etc. Even then I couldn't flip it over on my own, so I wanted to try something different.

I decided to see if I could get the arbor out by going in through the side. To get the arbor out you need to remove the rear bearing retainer ring (3 screws) and a c-clip. (shown below) To get the entire arbor assembly out, all you need to remove is 1 e-clip. Then you can R&R the arbor on the workbench. The problem is, it's really hard to get to the screws or the e-clip from the back or bottom of the TS housing. That's why people flip it over and remove other parts.

So, I broke out the oscillating tool and made an access hole:

After removing the belt pulley, I had access to the back side of the arbor. You can see the 3 screws for the bearing retainer ring. On the right you can see the link plate and the bottom of the e-clip.

I popped the e-clip off of the link, slowly tapped the arbor assembly off of the link pivot rod and height adjustment shaft. Once it was free, I lowered the assembly out through the bottom of the saw and then removed the arbor.

Since I had a spare arbor that already had a new front bearing on it, I decided to use that and just add a new rear bearing. Once the rust was removed from all mating surfaces and pivot shafts, the arbor assembly slid right back in. No tapping required.

The only alignment issue that I ran into is that the blade is about 1/32" off from it's original position. You can see here where it cut into the zero clearance insert.

The insert was getting sloppy anyway, probably from the bad bearing. I have a few blanks, so it's not an issue. I will need to do something about my sleds though. I'll probably just glue a strip into the slot and re-cut it.

The only thing left is to patch the hole in the housing. A piece of plywood to cover the side will give me a place to hang feather boards and other accessories that are currently taking up shelf space.

The saw is purring once again.

[Leon]

Good job! I would suggest insuring that the blade is parallel to the fence before cutting another zero clearance slot.

[DerbyDad03]

It's parallel to fence and parallel to the miter slots. None of that changed. The only thing that changed is the location of the blade from a left-to-right perspective and I think I have figured out why.

While cleaning up, I look a closer look at the old arbor. I noticed that the c-clip behind the front bearing was not in its slot. It was still around the shaft and the bearing was still tight against the flange that contacts the blade. Somehow, all three of those parts had slid down the shaft ~1/32" inch - basically the width of the c-clip.

Obviously, it's not anything that I could have seen prior to removing the arbor so maybe it happened during disassembly. However, if I retrace my steps, I can't think of anything I would have done to pop the c-clip out of it's slot and move the flange and bearing towards the rear of the arbor.

When I hung the arbor by the flange in my vice and pounded the shaft down to expose the c-clip slot, it took a few decent whacks to get the shaft to move that small about. It wasn't like the flange was flopping around so I don't see how disassembly could have moved it.

A flange that had slipped backwards would have taken the blade to the user's left, which would explain why the new arbor placed it just a little bit to the right of where it was.

A curious situation to say the least. In any case, I've adjusted the little red line on my fence to compensate. ;-)

Inserts and sleds can wait another day.

[DerbyDad03]

I've been thinking about the blade location a bit more and may have narrowed it down to the root cause.

A major difference between the new arbor and the old is that the new arbor has a rather hefty e-clip behind the front bearing...

...while the old one had a wimpy c-clip:

This saw is ~40 years old. I'm guessing that at some point the blade was so over-tightened that the flange pushed against the bearing, the bearing pushed against the wimpy c-clip, flexed it and popped it out of the groove.

If that is the case, then maybe Craftsman/Emerson Electric became aware of this issue and beefed up the bearing support with the heftier e-clip.

I've had the saw for about 30 years and I'll admit that early on I used to tighten the blade more than it needed to be. I may have been one of the many reasons that the front bearing clip was beefed up.

[Leon]

I think the fact that you R&R'd parts caused this issue. It is not unusual for the blade to be somewhere else after a major parts replacement.

[DerbyDad03]

Did you see my post where I said that the original clip was out of the slot and that I had to move the shaft backwards/flange forward to get it back in?

The flange that supports the left side of the blade had moved to the left from the user's perspective. If I put that arbor back in the saw now that I've fixed it, the blade would move to the right, just like it is with the new arbor.

I'm not saying the *nothing* would have moved had the original arbor been "OK", but the fact the flange was physically out of position, in the direction that fits the exact issue that I have, has to have contributed to it.

In any case, a new zero clearance insert has been cut and the strip is drying in the big sled. Once I re-cut that slot, I'll move on to the smaller sled. I don't think that I have any other jigs will be impacted by the new blade position.

[Clare Snyder]

Since the days of assembly line and interchangeable parts that should NOT happen.

OP likely pushed the C clip out, as per his explanation, WAY back before making his zero clearance inserts and sleds - before he got to the point he knew his own strength and how tight things did NOT have to be.

[DerbyDad03]

In a perfect world, yes. In a world where the saw was built in ~1978 (by Emerson Electric) and the new arbor was bought in 2015, things could have changed. In fact, things *did* change, as I showed with the old c-clip version vs. new the e-clip version. The e-clip is thicker than the c-clip, so something had to change to make it fit. I have no idea when that change was made, but if it was changed because other users did the same thing that I did, that may have taken a long time to be discovered. (I said *may*) It's not inconceivable that by the time they made that change, the original assembly line and process may have been long gone. (that's my point, BTW)

That said, I do think that the clip being pushed out caused some, if not all, of the leftward movement, but I'm not willing to totally discount some change due to the different manufacturing method.

Since we can't really retrace all steps, some part of this will remain a mystery. In any case, all inserts and sleds are fixed, the hole in the side of the housing has been covered and the saw purred through the rip cuts required for the "filler strips" for the sleds.

[Clare Snyder]

The "replacement parts" are "direct replacement" "upgraded" parts - even from Emerson. 1978 is not "old" equipment in the woodworking world. My last table saw was a Beaver made back in the late 40s or early 50s. Any replacement parts I was able to source were identical, bolt in parts. Even parts supplied by rockwell 50 years later.

I'd bet on 99.9+% of the difference.

[DerbyDad03]

Upgraded means "different". Doesn't mean it doesn't fit/work, but it also means that it's not identical. I guess you are trying to help me prove my point.

Who said it was? I simply mentioned when the saw was built, when the arbor was bought and the fact that they were different.

That's nice. My part was not identical, even though it was sourced from Sears, the same vendor that sourced the saw.

If you don't think it was 100% of the difference, maybe, just maybe, the remainder was caused by the replacement part. Thanks for helping me prove my point again.

[J. Clarke]

Google "manufacturing tolerances".

There's a reason that saws don't come with pre-cut zero clearance inserts.

[Clare Snyder]

Thats why I didn't say 100% of the difference was caused by the slippage of the old c clip. I allowed for a few thousandths of an inch for manufacturing tolerances.

The difference between the proper "hand tight" and30 ft lbs will make more difference than the few thousandths of an inch of manufacturing tolerance. You can walk the blade edge significantly by "dishing" the blade by overtorquing the blade.

[Leon]

When humans are involved in the assembly process there are issues that can come up. Take two off of the assembly line and see of they are precisely identical.

[Leon]

OMG did we just agree on something? Although SawStop TSs come with precut inserts.

[Leon]

Just another thought on this. He had his TS blade aligned to the fence with a worn out bearing. A new bearing will place the blade in a new position.

Now Derby did have a reasonable explanation but gosh, why realign the front end on a car after replacing ball joints or tie rod ends?

[DerbyDad03]

I think you're on my side, but even I have to question your "new bearing will move the blade" theory. Why?

In my case, the front bearing (good) was sandwiched between the flange and a c-clip. The rear bearing (bad) was sandwiched between an e-clip and a c-clip.

Since the bearings had to end up in the same position on the shaft, why would a new bearing (either one), in and of itself, reposition the blade?

[Clare Snyder]

That depends on your definition of precisely, and identical. You might find enough difference from the farthest off-spec one way and the farthest off the other direction to be detectable with an engineers rule

[Clare Snyder]

Well, the tie rod ends are threaded in and adjustable - but I have on MANY occaisions replaced 4 tie rod ends on a vehicle and ended up with the toe-in(the only thing affected by the tie rods) accurate to within

1/16 of an inch - less than half the allowable tollerance.Even with third party (non-oem) parts.

I've replaced ball joints and had them well within spec too. Depends if the alignmenthad been screwed with trying to correct for worn parts or not. If the front end was aligned with good parts, then the parts ware, and the worn parts were replaced without re-aligning first, there is no reason it can't go back together and be within spec.

If Derby made the zero clearance inserts after the e-clip was knocked out, it can't fit after replacement with the clip in place - that's obvious. However, if he got a zero clearance insert from another saw of the same model, without a shifted arbor, the insert would in all likelihood be so close to a perfect fit you couldn't see the difference.

[Clare Snyder]

Wouldn't unless it was REALLY wallered out. - way past the popint of being "unuseable"

[Leon]

Worn bearings introduce play, especially with a load from the material being feed. You adjust with the blade turning and work from there, if the bearing is shot.

A New bearing will take out the play. Time to readjust.

I used to have the same saw that you have. I could adjust the blade to the miter slot and then the fence to the slot. Cut results were unpredictable. I finally got to where I would simply sneak up on the right adjustment for the fence after making a test cut.