[Mac.CNC]

Last night, wile cutting more parts, I suddenly heard a clunking noise on the y axis and wasn't sure what it was at first.
Turns out that the upper right bearing cracked!
I'm not sure what caused this yet. I haven't had any serious crashes lately, perhaps I just tightened things up too much in that area - still then this is a a little strange.
Mine are all ABEC7 bearings throughout the machine.
Just wondering if anyone has seen this before and/or tried to upgrade to more expensive bearings yet (that last longer, etc).


WP_20140211_004 by mkloberg, on Flickr

Technically, this should go in the "wear and tear" drawer, but I think that's a little early for that kind of thing.
--
Mac

[Mzones]

MacCNC

perhaps I just tightened things up too much in that area

Agee with you on this point

still then this is a a little strange

.

Nothing strange about it at all.

Mike

[DY123]

Wow, it would seem that something was tightened too hard. When you set up your carriages how did you adjust the "roll" of the bearings on the rails?

[Levictus]

I like to use the bearings with the metal seals as opposed to the rubber seals. They seem to handle lateral loading better due to increased rigidity. At least that's what I have found in my long distance bike riding days.

[Mac.CNC]

I've done it several times now due to upgrades. No idea how this got so tight. Also this particular bearing shouldn't even see any load most of the time.
My approach was always to completely loosen the non load bearings (top front, bottom rear and bottom), then I would adjust the other ones until everything is aligned and the tramming bar turned true (I don't use shims). Then I slightly tightened the non load bearings just so much that the set screws could still move the bearing.
At this point I slowly tightened up the set screws while feeling if the bearing would still turn. I just kept putting more load on the set screws until all bearings had so much resistance that I could hardly still turn them by hand. Only then, I'd tighten up the bolts to lock everything in place. Worked pretty good so far.
--
Mac

[the_tool_man]

Mac:

You may have just gotten a defective bearing. With the rolling-contact load that outer race is seeing, it wouldn't take much of a surface or material defect to start a fatigue crack. Keep in mind that ball bearings aren't really intended to be used in this way, with rolling contact directly applied to the outer race. Ideally, they'd be fitted into rollers (like a cam follower) that take the contact loading. Such rollers typically aren't hardened as much as bearing races, and aren't as brittle. Not criticizing the design. Just pointing out a possible explanation. Going to a higher ABEC rating won't change this, as that only has to do with the tolerances of the bearing, not it's alloy, heat treat, etc.

You might check your rail to make sure there isn't a burr or piece of foreign material on it. Assuming the bearing didn't slip, the same point on the bearing's outer race would have to roll over that spot repeatedly, which might have caused the crack.

[mactec54]

I like to use the bearings with the metal seals as opposed to the rubber seals. They seem to handle lateral loading better due to increased rigidity. At least that's what I have found in my long distance bike riding days.

There is no difference in the bearings, just how they are bearing sealed, the rubber sealed is by far the best for dusty conditions, shielded is just that

Why has his bearing broken/failed, These bearings were not designed to be used how they are being used, even though there are hundreds of people using them this way there will always be failures

These Bearings were designed to be used in a housing, which protects the outer race from distortion, which is what is happening to cause them to break

[Mac.CNC]

Thanks, ToolMan, your post was very insightfully.
At the moment, I can probably rule out any contaminants on the y-rail. It has been as clean as it can get for teh past few weeks, I've been wiping down the rails with WD40 just about every other day and they are precision ground, so the variations in thickness can be ruled out as opposed to cold rolled rails.
I still can't put my finger on this, perhaps it was just a defective bearing as you suggested earlier. Hopefully :-)
It was quite different before I got the dust collection thing going, the rails were always covered in grime and stuff after each job. Perhaps it got its inital crack way back then.

@mactec54, I totally buy your theory, failures like this are to be expected because we're doing somthing with cheap bearings that were not designed to do in the first place.
On the other end, it's a good machine and exellent design, perhaps it was just a fatigue crack never to be seen again, hopefully.
--
Mac

[the_tool_man]

...@mactec54, I totally buy your theory, failures like this are to be expected because we're doing somthing with cheap bearings that were not designed to do in the first place...Mac

Mac:

I just want to reinforce the point mactec54 and I made. It has little (probably nothing) to do with how cheap the bearings are, or their ABEC rating, or whether they are sealed or shielded. Ball bearings are meant to be inserted into a housing, and are designed accordingly. When mounted that way, the bearing load is distributed over a large portion of the outer race. That is a very different stress condition then having the outer race roll directly on another surface, which concentrates the load on a very small area of contact. That's not to say they won't live that way. Clearly, nearly all of them do. But any defect, or other circumstance, that can add to the stress in the outer race has a larger chance of reducing the bearing's life. This holds true even with more expensive, high-precision bearings, as well.

I have not finished my Momus yet. But I'm considering using stainless steel bearings. They have a lower load rating, because stainless races can't be made as hard as bearing steel. But the stainless races aren't nearly as brittle, either. I wonder if anyone else has tried this?

[Mac.CNC]

I want to really thank you tool_man (and all of you) for all the amount of detail in explaining the loads and stresses on these bearings.
At the moment, I tend to conclude that that was just a bad bearing, trying to expose a fatigue fault it had in the first place.
I'm pretty sure there aren't too many variables to consider for additional stress coming from the thickness of my rails - they are precision ground.

I'm actually very interested in the idea to try out the different bearings that you suggest, cost to upgrade will be a little bit of an issue though.
Last I counted, there are 44 of these things in my machine now (this is with the xslave motor upgrade), so unless the new bearings are less than 1 to 4 $ a piece, it'll probably blow my CNC/girlfriend set budget for this month out of the water ;-)

Just kidding, but did you find anything that matches our bearings yet? I've been looking at INA and SKF, but didn't find something that matches yet.
--
Mac

[the_tool_man]

Mac:

Ebay has a bunch of S608 bearing listings from China. Some, if you buy in bulk, are as low as $1 each. But I don't trust their ABEC ratings. And if you read carefully, most are either shielded, or only have a single seal (RS, not 2RS). Outside of that, the cheapest I've found is at Bearings Direct:

S608-2RS Stainless Steel Miniature Ball Bearing 8x22x7 Sealed

Their webpage doesn't list the ABEC rating for these, either. I've inquired, but gotten no response yet. Even so, at nearly $5 each, they're not cheap.

[mactec54]

Mac.CNC

Just replace the broken one, the Stainless steel bearings may last longer, but if they cost more, they will not perform any better, the heat treatment of the stainless steel bearings, is not any were near as hard as the regular ABEC 7 bearings you are using

What is happening to make them break, think of the tire on your car, as it roll's it is deforming all the time, your bearings with this type of setup are doing the same thing, they are deforming as they roll, over time,they will break from metal fatigue with the constant rolling action

You may have them adjusted a little tight, try to adjust them so the bearing can just turn/skid with your fingers

The Stainless steel Bearings will only be ABEC3 with lots of clearance, they won't be anything like what you have, ABEC7

The Ideal Bearing is the Cam follower Bearings, they are designed for this type of use/ loading, they have a much thicker outer & run on needle rollers, for you to use something like this your carriages would have to be modified to be able to use this type of bearing, or completely redesigned, plus the cost would be more, than what you are using

That is why I say stick with what you have, Don't load them quite as much against the rails, & they should last for many widgets being made

The photos show the difference from the Ball Bearing & the cam follower Bearing

[Mac.CNC]

Thanks for all the insights and explanations, Mactec and Toolman - makes a lot of sense.
I already replaced it and haven't had any problems since. :-)
Thanks again,
Mac

[jason003]

From the picture, it looks like the bearing might have been subjected to excessive load or stress. Over-tightening could be a factor.