Bearing Interference fit problem on shaft in wheel - how to solve using only temperature differential if possible - already tried and failed.

[sdelivery]

Two thumb press fit means exactly that.
With your bearing started straight it should require your two thumbs 180 apart to push the bearing in place, if it slides right on then that is a slip fit.
Most importantly the area where the bearing sits when fully home needs to be as close to a two thumb or light press, if you debur the leading edge to ease installation it shouldn't hurt anything.
Bearing retaining compound is a locktite liquid that chemically locks the bearing in place. Most require some heat to ease release.

 

[FOMOGO]

I wouldn't proceed until you have good measurements. A quality caliper should get you close enough for what you are doing. Do you have a press? Press fit is usually .001 per inch of diameter. Edit: If dry ice+ heat won't allow the bearings to slip in I'm guessing the bore is too small or bearing o.d. is too large. Again, get a good caliper and verify. Mike

 

[Just for fun]

I agree with FOMOGO, you need better measuring devices. It looks like a straightforward operation, maybe a little heat and cooling to get the bearings installed in the wheel and you have that done. At this point the wheel assembly it should just slide on the shaft. I'm thinking you may have the wrong bearings installed. And since that maybe the case you could turn the shaft down a bit. But you better get good measurements first.

Or replace the bearings with the correct ones.

 

[Jake M]

Hello,
sorry about the wall of words, I don't think I can make it smaller, although not complicated to understand.

Background:
I have a bandsaw that needs bearing replacement. the guy I bought it from told me it was replaced only 5 years ago, but Both bearings are bad when I brought it home suspecting poor quality bearings or poor installation. They are 6003 bearings. I want to upgrade to seal bearings from shielded. I have tried a couple of trial and error methods to get this to work Without success.

parts
The wheel ( part 23) is made of aluminum alloy. Bearings ( part 19 and 27 A) are 6003. Parts 25 is a shaft ( not keyed, but friction fit) that is 17 mm in diameter ( unverified). As you can see from the picture, the shaft has a stop in the back for the bearing. I also don't have an inside micrometre to measure the Exact bearing diameter. The shaft is 17.05 mm I think, But this could be wrong, as my digital CalPERS are cheap.)

reassembly installation procedure, deduced:

1. install bearings in wheel. Achieved with a small toaster oven at 200°F, as wheels are made of aluminum. Maximum recommendation for bearings. shaft at room temperature.
2. part 25 must be slipped in through the back and not the front due to the bearing stop near the threads.

goal

I want to achieve an easy slip fit Using temperature differential for two reasons. The first has to do with Not able to swing a hammer anymore With a lot of force. The second, is that it seems to be a two-step process as the shaft if driven last with a hydraulicpress or mallet will be pushing on the inner race while the outer race is also under force, damaging the bearing. My hypothesis is this is what happens five years ago.

My current attempts and/or strategies.

1. ( failed Shaft/ success bearings in wheel, 310°F differential). the shaft and wheel are inserted in the oven together heated to 200°F. the shaft was cooled using dry ice is 109°F. I did not attempt to hammer it in, but tried placing. this may have been a mistake. two step process.
2. ( Not attempted, 560°F differential, unknown if possible). I can replace the two bearings with high temperature steel shielded bearings which are much worse in performance. more expensive bearings at $15 a piece. I can heat to 450°F and drop in a dry ice shaft. I don't like this as the tire will need replacing every time this is performed. this is very difficult to remove without destroying. I'm assuming the bearing can take the 500°F and not just the grease which might be wrong. Making this an expensive bearing placement, Done often). I can't find a seal bearing for 6003 high-temperature steel[ opinion: bad solution, But may B only option]
3. [not attempted, 360°F differential. bring the wheel my local nitrogen supplier in Toronto Canada. I called a guy and ask if I bring the one wheel. I told him it was 15.5 mm rather than 17 mm by mistake. He said this would fail and said don't bother coming.
4. [ not attempted, 510°F differential] I could go back to my same nitrogen supplier and bring my small toaster oven ha ha. Or my heat gun. He might say no. I don't know if this would work
5. different installation method not considered.

Please let me know how you think I should proceed. this is a real pain.
thanks

Step one is you need GOOD measurements. You've got to know how much interference fit you've got before you even consider how much temperature differential is going to achieve what results.

I don't see any spacer between the bearings. That gives you some options. They're a "flat" design, which probably has a proper name, but the inner and outer race are the same thickness. In other words, you can press one bearing with the other. That's a good degree of freedom if you're having to invent ways to do this.

I agree with others, there is no need at all to have both the inner and outer race pressed to an interference level. On the other hand, I've seen a lot of cases where this was done anyhow, for reasons outside of my pay grade. Or related to my price point. One of the two probably.

Unless this was high precision stuff (band saws are not), while you "might" find one, usually the tolerances in manufacture, even quality stuff, means that one machine might have a gentle tap fit with the bearings, and another might take all afternoon to beat together with a post maul. Make no assumptions based on what the fit "should" be. It "should" be what the bearing world would call "sloppy". If you could -count- on heating to a drop in fit, you would not be buying bearings for ten bucks.

My first inclination is that with an awkward setup like this, more than likely something unfortunate happened to the replacement bearings. And/or the replacement was the variety that can be had for two bucks and change, INCLUDING paying the postage right to your front door.... You can literally buy bearings that are made of unhardened steel, presumably mild, as they rust real fast as well, if you try to save enough money.....

I see two (plausible) options really, giving that I'm extrapolting that you have minimal tools for this, so there's my "overview" of what's got to happen. First, as I and other said, you need a good measurement of the shaft and the bearing. Verify that the fits are plausible (0.05mm or 0.002 inches is "probably" doable reasonably, if that's accurate and the bearings are dead on..., but you're at the far end of what that size ofmaterial is going to tolerate). This will NOT slip together with heat/cold levels that are plausible due to changes in metulurgy of the parts. It's gonna need pressing. Maybe a swing press, or maybe a proper press..... So the overview is this- I would start by pressing the two bearings onto the shaft, since that's going to take fixturing and force and what not. You'd need a driver (pipe?) that could press directly on the inner race. One race can press against the other with no issues. Second, I would take that assembly, and as you've already discovered, aluminum expands a LOT, so that assembly, and the easier fit in the wheel makes a good "second operation", since the press/interference is a lot easier to work with.


OR......... As has been mentioned, but I'll second because it just sounds so "wrong" for something as precise as a bearing- This is not a high precision thing. (Obviously some level of care is needed, don't "try" to screw it up and you'll do fine...). If you can spin that little shaft in a lathe, drill press, etc...... Once you've got the measurement situation figured out, spin the shaft and dust it off carefully and systematically with emery tape until you achieve a more plausible press/friction/ fit. Even a slip fit would be fine, given the nut that tightens the whole thing, although I'd want on the "just barely" slip fit range. Although it might well could survive some wiggling, you really don't want it wiggling...

 

[maple syrup]

I agree that something is wrong in that one or the other of the fits should be slip and even the press fit doesn't need to be real tight and need that kind of heat to accomplish it. The bearings revolve slowly and the wheel is under tension so the bearing is unlikely to turn on either the shaft or the wheel bore.

While sealed bearings are fine in this application, they are not an upgrade. Usually it is the other way around.

Dave

Interesting points About the fits. Gives confidence that they can be made smaller. I was reading on the inter-webs that slip fits should be 5/10 to 2 thou larger than the shaft.

No, but that type of bearing is for an entirely different application, not for a bandsaw.

good to know. saved me a bunch of time and money.

If you decide that the inner race just has to be a press fit (I don't really agree with this) take the shaft and freeze it. By putting a container in another pan and surrounding it with dry ice. Put the shaft into the inner can and pour acetone over it. It'll get cold enough; believe me. After waiting for the temp of the shaft to get down, remove it and quickly clamp it in a vise or something. Do NOT touch that shaft with bare skin! While all this is happening get the bearing up to 200-220 degrees F. Bearing should slide on.
I have seen an old tool and die guy do this. He claimed it got the frozen part to below 300.

I like the idea of freezing it, which was my first attempt.

Two thumb press fit means exactly that.
With your bearing started straight it should require your two thumbs 180 apart to push the bearing in place, if it slides right on then that is a slip fit.
Most importantly the area where the bearing sits when fully home needs to be as close to a two thumb or light press, if you debur the leading edge to ease installation it shouldn't hurt anything.
Bearing retaining compound is a locktite liquid that chemically locks the bearing in place. Most require some heat to ease release.

thank you. Good to know

I wouldn't proceed until you have good measurements. A quality caliper should get you close enough for what you are doing. Do you have a press? Press fit is usually .001 per inch of diameter. Edit: If dry ice+ heat won't allow the bearings to slip in I'm guessing the bore is to small or bearing o.d. is too large. Again, get a good caliper and verify. Mike

yeah. I've been putting it off for a long time to get a better one. It just might be time. There's not much I can do without knowing the size difference.

I agree with FOMOGO, you need better measuring devices. It looks like a straightforward operation, maybe a little heat and cooling to get the bearings installed in the wheel and you have that done. At this point the wheel assembly it should just slide on the shaft. I'm thinking you may have the wrong bearings installed. And since that maybe the case you could turn the shaft down a bit. But you better get good measurements first.

Or replace the bearings with the correct ones.

This is an interesting point. It went surprise me if the bearing was slightly different. This comes off the Delta bandsaw that had some very oddball sizes. Like the motor shaft was 5 mm keyway rather than a 3/16 inch. Just so they can sell you their part which they no longer make.
The truth is I would rather use a standard bearing size and have the shaft turn down or sanded. I had to do the same thing with the old motor. It was easier to throw it out then finding a new pulley. a shame

Step one is you need GOOD measurements. You've got to know how much interference fit you've got before you even consider how much temperature differential is going to achieve what results.

I don't see any spacer between the bearings. That gives you some options. They're a "flat" design, which probably has a proper name, but the inner and outer race are the same thickness. In other words, you can press one bearing with the other. That's a good degree of freedom if you're having to invent ways to do this.

I agree with others, there is no need at all to have both the inner and outer race pressed to an interference level. On the other hand, I've seen a lot of cases where this was done anyhow, for reasons outside of my pay grade. Or related to my price point. One of the two probably.

Unless this was high precision stuff (band saws are not), while you "might" find one, usually the tolerances in manufacture, even quality stuff, means that one machine might have a gentle tap fit with the bearings, and another might take all afternoon to beat together with a post maul. Make no assumptions based on what the fit "should" be. It "should" be what the bearing world would call "sloppy". If you could -count- on heating to a drop in fit, you would not be buying bearings for ten bucks.

My first inclination is that with an awkward setup like this, more than likely something unfortunate happened to the replacement bearings. And/or the replacement was the variety that can be had for two bucks and change, INCLUDING paying the postage right to your front door.... You can literally buy bearings that are made of unhardened steel, presumably mild, as they rust real fast as well, if you try to save enough money.....

I see two (plausible) options really, giving that I'm extrapolting that you have minimal tools for this, so there's my "overview" of what's got to happen. First, as I and other said, you need a good measurement of the shaft and the bearing. Verify that the fits are plausible (0.05mm or 0.002 inches is "probably" doable reasonably, if that's accurate and the bearings are dead on..., but you're at the far end of what that size ofmaterial is going to tolerate). This will NOT slip together with heat/cold levels that are plausible due to changes in metulurgy of the parts. It's gonna need pressing. Maybe a swing press, or maybe a proper press..... So the overview is this- I would start by pressing the two bearings onto the shaft, since that's going to take fixturing and force and what not. You'd need a driver (pipe?) that could press directly on the inner race. One race can press against the other with no issues. Second, I would take that assembly, and as you've already discovered, aluminum expands a LOT, so that assembly, and the easier fit in the wheel makes a good "second operation", since the press/interference is a lot easier to work with.


OR......... As has been mentioned, but I'll second because it just sounds so "wrong" for something as precise as a bearing- This is not a high precision thing. (Obviously some level of care is needed, don't "try" to screw it up and you'll do fine...). If you can spin that little shaft in a lathe, drill press, etc...... Once you've got the measurement situation figured out, spin the shaft and dust it off carefully and systematically with emery tape until you achieve a more plausible press/friction/ fit. Even a slip fit would be fine, given the nut that tightens the whole thing, although I'd want on the "just barely" slip fit range. Although it might well could survive some wiggling, you really don't want it wiggling...

The outer race is definitely wider for a 6003. That is what I measured on the old bearings I removed and the new ones I bought.
from what I can tell, the bearing is not flat And correct me if I'm wrong but that means the first option is not possible as both races cannot be banged together at the same time.

I didn't realize the differences between the tolerances. I will call Nachi customer support. That is the brand my local supplier has. My suspicions to do about the cheap bearings. When you saying this need to press, are you saying that making the shaft a bit smaller is not recommended or it's just another option?

So after verifying with a good measurement, how much of the difference should it be in terms of sizing. I've read online somewhere that a slip fit should besomething approximately 0005” to .002” . two thousandths sounds quite large.

 

[tq60]

Take the shaft and wheel to a GOOD bearing store.

From that, they can take correct measurements and provide the correct bearings.

Sent from my SM-G781V using Tapatalk

 

[markba633csi]

Slip fit/light interference fit should be something like .0002 to .0005
Just take the shaft diameter down a bit
Easy Peasy

 

[Jake M]

This comes off the Delta bandsaw that had some very oddball sizes. Like the motor shaft was 5 mm keyway rather than a 3/16 inch. Just so they can sell you their part which they no longer make.
The truth is I would rather use a standard bearing size and have the shaft turn down or sanded. I had to do the same thing with the old motor. It was easier to throw it out then finding a new pulley. a shame

You might be giving them too much credit. If you talk with them, it doesn't seem that they care at all about selling you parts.... They might have just thrown together "whatever was easiest to get in the local economy" because they didn't really care about standardization or down the road parts sales.

The outer race is definitely wider for a 6003. That is what I measured on the old bearings I removed and the new ones I bought.
from what I can tell, the bearing is not flat And correct me if I'm wrong but that means the first option is not possible as both races cannot be banged together at the same time.

My apologies, and I'm glad you caught that. My sample 100 percent does have equal widths, so I assumed, and gave you bad advice on that part. I just looked at a couple of big name sites, and none of them call out any unique dimension for the inner race thickness. My guess, since these are most typically "shaft bearings", is that dimension is not functional, and therefore just not defined.

EDIT- Removed erroneous content.

 

[Asm109]

The bearings are retained on the shaft by a circlip in the groove at the end of the shaft. It may or may not have a wave spring washer to apply a bit of preload. In either case, there is no danger of reefing down on the bearings with a wrench.

 

[Jake M]

The bearings are retained on the shaft by a circlip in the groove at the end of the shaft. It may or may not have a wave spring washer to apply a bit of preload. In either case, there is no danger of reefing down on the bearings with a wrench.

You're absolutely right. MY saw uses a nut, I went and looked to see if by chance there were similarities. Then apparently I smoked a lot of crack and got all confused. I guess I'm really on a roll here.... Good catch.