Since my humble post is the object of such adversarial scrutiny I feel obliged to respond point by point:
Precision ball bearings are manufactured to standards established by the Annual Bearing Engineers Committee (ABEC) of the Antifriction Bearing Manufacturers Association (AFBMA).
Uuuuuhh, actually that's Annular Bearing Engineers Committee. As in "shaped like a ring", not as in "meeting once a year".
They have been accepted by the American National Standards Institute (ANSI) and conform essentially with the standards set by the International Standards Organization (ISO). Doesn't tell the average person (myself included) a whole lot, but gives an idea as to what the term "ABEC" means. Basically ABEC is a grading set up to identify and set dimension tolerences in the manufacturing of bearings.
Sorry, it's easy to drop terms one is used to using without thinking twice. I'll try to do better from now on, but I may occasionally err. At least I can depend on there being good folks here who will let me know when I slip up. (Whew, . . . I was worried about that!)
ABEC does not cover: Radial Play, Raceway Curvature, Roundness or Surface Finish, Material, Ball Complement, Number, Size or Precision Level, Lubrication, Cleanliness at Assembly or Packaging, Retainer Design. If you're looking for useful information, it doesn't tell you much. Made-in-PRC motors usually come with iso-compliance stickers attached. More puffery than guarantee. It's still buyer beware.
Everything is still buyer beware, but since you brought it up, ABEC ratings do address most pertinent bearing tolerance issues that I face with any regularity when specifying bearings, and I think that you give the wrong impression when you start quoting a punch list of things that are not covered by the ABEC. For instance, "Radial Runout" is a value specified by the ABEC, "Radial Play" is not, and they are not the same thing, at least not exactly. But if you build a bearing that meets rigid Radial Runout specs, by default the Radial Play is drastically minimized. You can't have one without the other, and it's that way with most specs. When manufacturing an item you don't typically get to select the specifications that you want to meet, . . . they are defined by your industry, and you use the testing methods and standards that are recognized and approved by your industry (like the ABEC scale), or your competition will certainly point out to all of your customers that you don't.
Re: ISO ratings, from what I can see, all that ISO guarantees is that if you make junk you will have a record of how you did it and make it the same way every time. But please don't tell my corporate ISO officer that I said that. <HA>
Dynamic load ratings can be as much as 250% of static load rating and are "stated" in multiple ways. I consider them meaningless. Those NTNs top dynamic rating is 2770 - pure b/s, IMO.
Static and dynamic loading are vastly different. I work with engineers who specialize in static loading and with engineers who specialize in dynamic loading, and often when hashing out new designs we have to put our heads together to insure that we don't focus on one loading problem to the neglect of the other. (Or other
s, as the case may be.) The static load ratings and dynamic load and RPM ratings have enormous meaning to the designers who develop them, . . . and to the attorneys who represent them if, in fact, their ratings turn out to be pure b/s.
FWIW, "ABEC-1" is the lowest grade (9 being the top).
Stated that way, I think it gives the wrong impression. The ABEC scale is a rating of degrees of precision. Many radial ball bearings do not meet any of the ABEC scale ratings. ABEC-1 is the scale with the "loosest" specifications, and is the specification that the vast majority of ABEC rated bearings meet, but you can be sure that an ABEC-1 bearing is superior to any non-ABEC bearing, unless, perhaps, the non-ABEC bearing is some sort of specialized, one off part. Also, as the ABEC rating goes up, the price goes up dramatically, and, typically, either the load rating or the max RPM rating will go down.
Consider this, . . . in bearings sized for a 12mm axle, like our minis have, the allowable radial runout in the ABEC-1 rating is 0.0003", or roughly 1/10th the thickness of a human hair. I suspect that the runout in hubs, wheels, and tires of most of our 30+ year old minis is enough to make 0.0003" pretty darned insignificant. So for wheel bearing applications on Honda MiniTrails an ABEC-1 bearing with a 2200 pound dynamic load rating, and a 13000 max RPM rating, is clearly overkill, but you've made a strong case for a little overkill being a good thing, and I'd have to agree.
(Also, FWIW, I think that the ABEC-1 rating is part of the specification for the 6301 bearings used in our minis, since I cannot find an unrated 6301 in any of my catalogs.)
All of which is confusing information overload.
Ya think??
You don't see many failed bearings on vintage CT70s. OEM bearings came from Koyo & NTN. Going with a known supplier covers the "buyer beware" side of things and simplifies the entire process. Even 900lbs static load + 10000rpm is massive overkill for any bike, let alone one with 10" wheels. Odds are, that for anything other than a 50mph+ road bike, most bearings available will be okay. Worst case, a wobbly wheel at 25mph allows plenty of time to stop safely. On the other hand a roadgoing bike, with more than 110cc of motorvation in the frame, potentially puts the rider in very different scenarios...like being stuck on the road (between "No" and "Where") or going down at freeway speed. Whether spending an extra 35 bucks is a large outlay is an individual call. Could turn out less costly, at the bottom line, than two sets of cheapies after the first one goes away after 620 miles.
Look, all I hoped to do with my post was offer a source of excellent quality bearings that haven't had their prices jacked by a seller who wants you to think that they are something more than what they are. All of the bearings that I have purchased from McMaster-Carr have been from respected manufacturers, like SKF, Fafnir, NTN, FAG (side note - it's actually pronounced "fog"), Timken, NSK, etc. I think that anyone who has hung around and endured this thread is now drowning in other people's opinions, and has probably been able to form their own opinion on the matter.
. . . . but what do I know, . . . I buy new seals and clean and repack the original factory bearings every few years, and they seem to be lasting forever.
-kevin