Zipp Grade 2 Bearings???

Back by popular demand, the general all-things Road forum!

Moderator: robbosmans

eschelar
Posts: 11
Joined: Tue May 29, 2012 11:07 am

by eschelar

So I was reading on Zipp's website the other day and they say that they are using Si3N4 bearings. OK that's fine.

But then they say that they are using Grade 2 bearings.

I've got a little engineering background and while I'm not directly in this field, I do know enough that there is no Grade 2 bearing commercially available. It might be possible to create such bearings in extremely limited quantities but no bearing manufacturer currently offers Grade 2 bearings. I can't find them anywhere from any of the suppliers I normally deal with. I can't find them on Alibaba. I can't find them via Google anywhere. I can't find them with wholesale dealers. I can't find them from aftermarket vendors. Every manufacturer clearly states that the finest grade available is Grade 3.

Now they say they are "Swiss-made", but I can't find any Swiss manufacturers offering Grade 2 bearings. I sent emails to WIB, AST, Bones and RKB.

I've asked around and everyone seems to be saying the same thing. Grade 3 is the highest grade available. Grades 1 and 2 can be made in very small quantities (like one at a time) and because of the naming convention, the roughness is actually known. By contrast, ABEC11 bearing races are apparently only ever used for research and measuring purposes and for super-exotic stuff like secret NASA projects and replacing worn parts in alien technologies being refurbished from crash sites... But they are available! ABEC 13 races though... well no, those don't exist. Nor indeed do ABEC12 if you want to split the difference for something equally nonexistent.

So what's going on here? Does Zipp have access to some obscure scientific research lab who doesn't mind selling one-off custom made balls for their 15-ball bearing packs at competitive market pricing? That's certainly what they are suggesting.
"For years, we've been the only hub manufacturer using Swiss-made bearings that are two and a half times rounder than any others in the bike industry - rounder even than many ceramic bearings. But for the 88 and 188, we added adjustable preload and switched to a slightly larger size that gives us room for a 17mm diameter axle, the stiffest in any road hub. The new bearings also yield a 28% higher load capacity, improve durability, and will save you a watt or two relative to the competition, regardless of conditions. If you upgrade to Si3N4 Ceramic Bearings, know that you'll be getting Grade 2 balls, the finest available, with tolerances less than 2 millionths of an inch per set of 15 bearings." - Zipp

"Rounder even than many ceramic bearings" Uhm... Si3N4 is a ceramic bearing... why the implication that it is better than ceramic bearings when it actually is a ceramic bearing? When I read stuff like this, I assume that the person who wrote it didn't really know what he was talking about - ie a marketing guy rather than an engineer. I did find one mention of RA=0,2 roughness grade bearings from RKB. That has a "2" in it... but RA=0,2 is not specific to whether this is a metric measurement or an inch measurement. Since everything else on the page is in metric, it could be metric. If so, it's Ra-μm = 0,2... that's equal to 8 millionths of an inch, or about Grade 8. RKB also claim to make Grade 6 and 7 steel balls with special steel, so it's likely then that this is actually what they are referring to on that page.

An uneducated reader could see that and assume that this was a Grade 2 bearing.

Or is this a blatant lie? Since they are using ABEC 7 bearing races, the facts don't seem to add up. Grade 3 balls on ABEC7 races is pretty common and is widely available at price points that seem to make sense for the cost of a Zipp wheel. Even more confusing is this page: http://www.zipp.com/accessories/detail.php?ID=30012
So it seems like they are saying that they use CeramicSpeed bearings and appear to be using these for the 88 and 188v9 hubs as described previously. CeramicSpeed uses Grade 3 Si3N4 balls with a variety of ABEC rated races (typically ABEC5, but it's not impossible that Zipp gets ABEC7)... Oh and they're not Swiss... Danish. In fact, they state Grade 3 all over the place on their site. So why would Zipp partner with them and call this an upgrade from their own 61803 Grade 2? Fishy... I could believe that this page has outdated information since it doesn't appear to reflect the redacted statement about saving up to 9 Watts, which is utter poppycock (even the currently claimed 1-2 Watts is debatable).

All my research so far points to them using Grade 3 silicon nitride bearings, *not* Grade 2.

But this also means that they are publishing and repeating a bold-faced lie. That's kind of uncomfortable...

I know there are engineering types out here on WW forums. Love to hear what you think of this.
Last edited by eschelar on Fri Oct 02, 2015 8:35 am, edited 4 times in total.

by Weenie


Visit starbike.com Online Retailer for HighEnd cycling components
Great Prices ✓    Broad Selection ✓    Worldwide Delivery ✓

www.starbike.com



goodboyr
Posts: 1487
Joined: Wed Jan 10, 2007 10:56 pm
Location: Canada

by goodboyr

I've got nothing to contribute, but I've got myself a big bag of popcorn, ready to go......

mattyNorm
Posts: 126
Joined: Fri Jan 02, 2015 12:29 pm

by mattyNorm

I can't wait till I finish my degree and can fill my sleepless nights with stuff like this instead of studying!

wingguy
Posts: 4318
Joined: Thu Mar 08, 2012 11:43 pm

by wingguy

eschelar wrote:"Rounder even than many ceramic bearings" Uhm... Si3N4 is a ceramic bearing... why the implication that it is better than ceramic bearings when it actually is a ceramic bearing?


I can't speak to anything else in your post, but you've read that bit wrong.

It's pretty clear that Zipp is first talking about their standard steel bearings being better than many ceramics (which isn't unusual) and then talking about their Si3N4 Ceramic upgrade option being even better again.

eschelar
Posts: 11
Joined: Tue May 29, 2012 11:07 am

by eschelar

wingguy wrote:
eschelar wrote:"Rounder even than many ceramic bearings" Uhm... Si3N4 is a ceramic bearing... why the implication that it is better than ceramic bearings when it actually is a ceramic bearing?


I can't speak to anything else in your post, but you've read that bit wrong.

It's pretty clear that Zipp is first talking about their standard steel bearings being better than many ceramics (which isn't unusual) and then talking about their Si3N4 Ceramic upgrade option being even better again.

Can't agree with you there man. They are not comparing their standard steel bearings to many ceramics - nor would that make any sense. Most bicycle wheels using steel use either use Grade 10 or Grade 25 (or really low end stuff with Grade 100 etc) bearings. In actual fact, the highest grade I have found for steel bearings is around a 6. Zipp themselves state that they use Grade 10 though - on the pages I referenced too.

However, the standard for ceramics is either Grade 3 or Grade 5. Even the cheap stuff. I can't think of a reason why Zipp would be claiming that a Grade 10 (their number) steel ball is higher roundness than a Grade 3 or Grade 5 ceramic ball - even a cheap Chinese one. The Grade is a literal descriptor of roundness in millionths of an inch, so by definition, a Grade 3 (3 millionths of an inch) is rounder than a Grade 5 (5 millionths of an inch), which in turn is rounder than a Grade 10 (10 millionths of an inch). To say that Grade 10 is rounder than Grade 3 would be like saying 10 is a smaller number than 3. Or that a person that is 10 feet tall is smaller than a person that is 3 feet high. Doesn't make sense.

Also, that page is specifically referring to their Si3N4 (silicon nitride) bearings and comparing them with other ceramic bearings. Although the vast majority of Grade 3 and Grade 5 ceramic bearings are Si3N4. There's no benefit to using ultra-elite Grade 6 steel bearings though when they also lack the hardness benefits of using ceramic and are probably just as expensive.

It's like saying that you use bananas for your banana split which are high in potassium, therefore your banana split is better than other banana splits that use bananas... which are also high in potassium.
Last edited by eschelar on Mon Oct 05, 2015 2:18 am, edited 1 time in total.

crankinstein
Posts: 15
Joined: Wed Jun 23, 2010 7:25 pm

by crankinstein

I think I now know more about bearings than I ever though possible, and less at the same time.

Sent from my SM-N910W8 using Tapatalk

Jmdesignz2
Posts: 272
Joined: Tue Jan 13, 2015 2:27 am

by Jmdesignz2

good stuff- let us know what they tell you when you ask them to clarify?

eschelar
Posts: 11
Joined: Tue May 29, 2012 11:07 am

by eschelar

Jmdesignz2 wrote:good stuff- let us know what they tell you when you ask them to clarify?
I'd love to, but Zipp isn't likely to respond and these bearing companies haven't responded in over a week. There's no real reason to. I'm not a customer of theirs.

That's why I posted here because I'm curious to find other engineers who came across this oddity. When I was googling the subject, I did see in another post from a few years back where a girl posted something about her boyfriend being an engineer and telling her that there's no such thing as Grade 2 bearings because the max available is Grade 3. But even though this was an engineer basically saying that Zipp is straight up lying, nothing came of it and I don't like to necropost.

There has got to be someone on here with an engineering background though. If ever there was a crowd of people who were "obsessively interested" in details, it's got to be weightweenies (oh and please understand that's not a pejorative, I'm also obsessively interested in details and I consider it a good thing).

goodboyr
Posts: 1487
Joined: Wed Jan 10, 2007 10:56 pm
Location: Canada

by goodboyr

I swore I wouldn't......but here goes. The grade of a bearing ball relates to the deviation from spherical in millionths of an inch. Grade 3 deviates max of 3 millionths of an inch. So what zipp is saying is that they get balls (probably pick and measure from lots of grade 3 balls) that deviate from spherical by no more than 2 millionths of an inch.

eschelar
Posts: 11
Joined: Tue May 29, 2012 11:07 am

by eschelar

goodboyr wrote:I swore I wouldn't......but here goes. The grade of a bearing ball relates to the deviation from spherical in millionths of an inch. Grade 3 deviates max of 3 millionths of an inch. So what zipp is saying is that they get balls (probably pick and measure from lots) that deviate from spherical by no more than 2 millionths of an inch.


Thanks Goodboyr. I think that much is already established by the definition of ball grades and in my first post where I was talking about millionths of an inch as a defining feature of ball grades (like the fact that RA=0,2 appears to be Grade 8 since it has a roundness tolerance of 8 millionths of an inch).

Every source I can find is saying that the highest grade available on the market from any supplier anywhere is Grade 3. Ball manufacturers can potentially make Grade 2 balls, but because of the extremely tiny tolerance, they are not mass produced and a single ball might cost you as much as a small car... from Italy... with a V12...

OK, so that last part is a bit of exaggeration, but seriously, nobody makes Grade 2 and every engineer and bearing manufacturer that I've talked to has told me the same thing: Grades 1 and 2 are theoretical grades because they can be defined by the roundness deviation in millionths of an inch and could presumably be made in a laboratory. But just because something can be made in a laboratory, doesn't mean that it is made for mass production.

For mass production, Grade 3 is the highest available. It doesn't matter if you're buying from Switzerland, Denmark, France, the US or China. Nobody makes grade 2.

So why is Zipp saying that's what they use? And why are they saying that CeramicSpeed bearings are an upgrade if they are ball Grade 3 on ABEC7 races? Does this mean that Zipp is buying from CeramicSpeed? If so, they are using Grade 3 balls. Because CeramicSpeed doesn't make Grade 2.

Zipp is claiming to use Grade 2. And I understand the idea behind trying to obtain Grade 2 by pick and measure on Grade 3, but they have not said this is the case, nor indeed is this a practically plausible thing for Zipp to be doing. The instrumentation to do pick and measure on ball bearings to differentiate between Grade 3 and Grade 2 is rather beyond Zipp's technology level and would result in some rather shocking price jumps. Grade 3 are already fairly expensive and if ball manufacturers aren't claiming to be able to make a Grade 2, it's safe to say that there is a very, very small percentage that would make it to Grade 2 (or else the ball manufacturers would be offering them). Ball manufacturers make *billions* of balls per year. If they have a production of 100 million Grade 3 Ceramics and there's even 1% of their batch that makes it to Grade 2, then they would have 1 million per year at Grade 2. That's enough to be marketable. However, if the number is much lower than that (ie 0.01%), you would end up with 10,000 Grade 2. While that would in fact be enough to be relevant in comparison with Zipp's annual consumption and low enough that the ball manufacturer isn't making them available to others, the cost of those balls would probably be way, way, way higher than a Grade 3. And I can't believe that Zipp would be doing this on their own because the cost would then be literally increased by 10,000 times. Yeah, 10,000 times higher cost than regular ceramics... If all of Zipp's wheels now use ceramic bearings and all of Zipp's bearings use Grade 2 as their literature suggests, they could hypothetically be buying 10,000 Grade 3 bearings just to get 1 at Grade 2. And with a basic run of 1000 wheels per year, Zipp would be one of the largest consumers of ceramic bearings in the world.

Contrary to popular belief, the cycling industry does not get first dibs on the highest quality of manufacturing materials available. The cycling industry usually has to queue behind the aerospace industry for carbon fiber and Nomex because the demand is so high. And then there's the whole scientific instrument community who would probably be glad of some Grade2 Balls, but is quite content with currently available grades.

Also, if they are using Grade 2 balls via this extremely expensive method of check and choose, why would they recommend CeramicSpeed with its Grade 3 balls?

It just doesn't add up in practical terms.

Occam's telling me that the most likely reality here is that Zipp is just simply lying about their ball grade. Or that they made a claim based on an error (I think the most likely here is the RA=0,2).

goodboyr
Posts: 1487
Joined: Wed Jan 10, 2007 10:56 pm
Location: Canada

by goodboyr

Sorry...missed that in your first post. As I originally said, I had nothing to add. [emoji4]. Should have listened to myself.

I do own two sets of zipps with regular non ceramic hubs. They spin just fine. And I'm sure I wouldn't be able to feel the difference between grade 2, 3,5 or even 10 balls.

eschelar
Posts: 11
Joined: Tue May 29, 2012 11:07 am

by eschelar

goodboyr wrote:Sorry...missed that in your first post. As I originally said, I had nothing to add. [emoji4]. Should have listened to myself.

I do own two sets of zipps with regular non ceramic hubs. They spin just fine. And I'm sure I wouldn't be able to feel the difference between grade 2, 3,5 or even 10 balls.

I wasn't sure what you meant by "I swore I wouldn't".

As to the difference between the grades, no, conventional wisdom says that you probably wouldn't even be able to tell the difference between grade 3 and grade 25 balls. Probably not even grade 100 balls to be honest if they are properly lubricated.

Even grade 3 is considered snake oil by most. This issue here though is not whether you would be able to tell the difference, it's whether or not Zipp is blatantly lying.

Every company says that you will go faster with their stuff. And nobody cares because the interpretation of that is pretty broad. However, when a company says "we use this specific item" and they aren't, that's different. That's called false advertising. I expect that kind of BSing from a low end Chinese company, but Zipp purports to be a US company. I expect a higher standard. This is especially true when Zipp is making a claim that they are better than other companies because they use this specific item. But if this item doesn't exist, then there's a problem.

If you buy a computer that says it has an NVidia GTX 1185 in it that nobody else can even buy, but it actually contains a 980, the same as everyone else on the market, this is a problem. Not only that, no other company *could* genuinely have a GTX 1185, since that model does not actually exist at this time. It's no different here.

The value associated with a ball grade of 2 is probably a few thousand bucks per ball, if not tens of thousands... For Zipp to claim that they are using this grade with 15 balls per bearing and 4-6 bearings per wheel - if it genuinely does not exist - is some damned dirty business. Kinda like how most people feel about Lance Armstrong, just straight up lying...

goodboyr
Posts: 1487
Joined: Wed Jan 10, 2007 10:56 pm
Location: Canada

by goodboyr

I swore I wouldn't get sucked in to this conversation......but I did.
You clearly feel strongly about this....and by the lack of response it seems you are in the minority. No one else cares. So probably best to pursue this through official channels.

eschelar
Posts: 11
Joined: Tue May 29, 2012 11:07 am

by eschelar

I don't understand why you - especially since you own two pairs of Zipp wheels - are not concerned about a company flat out lying to you about the stuff they use to make the products you ride on? You paid a premium to own Zipp wheels.

Why wouldn't it concern you if you had made that purchase decision based on information that they were literally just making up as they went along? Why wouldn't it concern you that they were using illegal marketing practices to secure a position as a world leader name brand without actually matching the quality of the products with their claims.

By the way, I'm considering making a wheel company of my own using brand new helium fiber and using Grade 1 silicon-diamond balls on ABEC 11.8 races. It's guaranteed(tm) to be at least 300% faster. It's so fast that light itself can't interact with the wheels. All you can see is just a skewer, which features a hand-peened custom dimple pattern. (also lowers ride-height by around 12-14 inches for true aero).

$3000/pair.

You wouldn't think that's acceptable by some guy on the forums. Why would that be acceptable by Zipp?

goodboyr
Posts: 1487
Joined: Wed Jan 10, 2007 10:56 pm
Location: Canada

by goodboyr

Sorry. I'm not playing. By by.

by Weenie


Visit starbike.com Online Retailer for HighEnd cycling components
Great Prices ✓    Broad Selection ✓    Worldwide Delivery ✓

www.starbike.com



Post Reply