Superjolly Bearing

[clear knot]

Do you think it has C2 or or printer-class bearings in it now?

My gut says the bearings are far enough apart on a SJ that radial internal clearance of the bearings cannot possibly have a meaningful impact on the stability of the burr carrier.

*stability of alignment, that is.

Because shielded are cheap and these were commodity grinders back in the day!

Do you happen to know the internal clearance on the C3 bearings and the bearing spacing distance on the rotor?

You would need to hit the thermal limit on the grinder before the internal clearance would be taken up...

Yep

That sounds correct. My guess is that they are not specced for temperature, but if they were, it's based on commercial duty cycle.

@jade bolt, if you get a chance, can you verify how the stator is secured into the body?

[exotic wadi]

Just standard bearing clearance, CN would be fine and probably cheapest to source. I doubt C2 is necessary but I wouldn’t buy C3 since they’re usually more expensive than CN for no reason in this application.

[exotic wadi]

Press fit into the cast outer part

[clear knot]

Thought so.

Thanks!

Ah, CN. The bearing with no class

You're down in the 30um range for radial internal clearance.

(Using the chart, not the equation)

Prolly.

[exotic wadi]

fyi my Mazzer Major came with C3 ZZs

[clear knot]

I see your bearing being in here-ish

So yeah, less than 25um, for sure.

[exotic wadi]

It’s all very much a non issue

[clear knot]

Yeah.

[exotic wadi]

C2 CN or C3 will all be completely fine. The radial runout on the shaft + the carrier and the mounting of the burr are all three higher than the C3 clearance

[clear knot]

Worst-case scenario, top bearing is pushed 25 microns this way and bottom is pushed 25 microns that way. So you have 50 microns of absolute tilt across the 7 inch (178mm) span of the bearings.

[clear knot]

So the math for what that means at the edge of a burr is to take the same ratio.

So 50 *64/178 in the worst possible case, which is highly unlikely.

[exotic wadi]

My phone dosnt do arctan

[clear knot]

No need.

They are like triangles.

So x/64 = 50/177.8

In microns.

And x = 18

[exotic wadi]

Would say /32 in that case

[clear knot]

I thought that.

But the other side tips up, right?

But technically you are correct. +/- 9um is the answer.

[exotic wadi]

But I think the issue wouldn’t be the angling, it would be excessive wear and vibrations ?

Honestly for such a low tech application as a coffee grinder I can hardly think of issues

But I like hypothetic scenarios

[clear knot]

This is the maximum allowable shaft wobble that these bearings would provide.

In the worst possible case.

[exotic wadi]

Under x force right?

[clear knot]

For which I can think of no possible use case that grinding would create.

[exotic wadi]

They run extremely true under 0 force

[clear knot]

Yep

And extremely true under a "uniform" grinding force.

Not really. But the forces that would tilt the shaft would have to be radial and located at each bearing and opposing each other for this worst case scenario to actually occur.

Actual shear forces during grinding would not do this.

[exotic wadi]

Yeah I guess coffee grinding doesn’t provide a great force spread over the burr, but since it’s all only on an arm of 32mm, there will barely be any angling force - assuming it’ll all be on one side

[clear knot]

And it wouldn't. Especially not with horizontal burrs.

[exotic wadi]

Which would, due to the nature of two bearings apart come over as a very radial force on the top bearing, but yes still give the 50u play

[clear knot]

There will be a radial and axial component. But really nothing at all acting on the lower bearing.

And even with ssps that have a counter-rotating sweep of tooth interference, the sum of all the forces will be mostly axial.

[exotic wadi]

I don’t get this tho

What would vertical burrs change

[clear knot]

Well, the bottom of the fixed burr sees all the initial action.

Thanks to gravity.

No doubt 90% of the load gets spread out thanks to centrifugal motion, but that initial cracking of those beans will happen on the bottom half.

[exotic wadi]

This is taking it far haha

Oh yeah you have to always just go to a bearing dealer

We should but a load sensor on a bearing carrier

[orchid cedar]

It's less the radial play, what might matter is axial end play. When I've built grinders I've used preloaded dual-row angular contacts, or two deep-groove bearings with some preload against each other

[exotic wadi]

Well, you should always 3x the added cost for money to all extra costs in between (dealer, taxes)

[clear knot]

Accelerometer, for sure.

[clear knot]

100% agree.

[orchid cedar]

Sealed bearings aren't always ideal because under constant use they get hotter etc.

Not an issue for home use ofc

[exotic wadi]

I really think that the axial force is always great enough to lock it in

But play might increase fines ?

[orchid cedar]

Probably the stock ones are CN (normal clearance, between C2 and C3), and just uses the weight of the rotor and shaft and assumes endplay is negligible

Rubber seals have gotten better and cheaper in the last 40 years too

[exotic wadi]

Yeah his SJs were C2 my Major was C3

I have an old one to, very old

[orchid cedar]

The La San Marco grinder I pulled apart a few months ago uses rubber bearing bushings lol, so clearly rigidity and axial play weren't priorities they were thinking about

I press fit a new bushing to get some negative clearance on a a dual-row angular contact bearing for that one, definitely an improvement 😂

[clear knot]

Funny, my manual doesn't say anything about the class of bearing, nor whether it is sealed or shielded.🤷‍♂️

[exotic wadi]

There was coffee in my old bearings, or atleast so thic on the shielding I was worried

So new ones were installed with thick layer of grease over them. So coffee would get stuck in there

[clear knot]

In actuality, it is.

Heck, the Forte just uses bushings and has the bottom of the carrier spinning on the adjustment "spline" like a top. There is an infinite amount of upward axial clearance in that design.

The trick with ball bearings is loading them enough axially that the radial clearance is absorbed.

This is why I'm not nuts about the Kafatek dual row angular bearing. You have a single unloaded bearing trying to support the entire business end of the grinder...

I mean, it's a very nice bearing.

And it will never wear out.

But the carrier is free to wobble about, and that is not entirely ideal.

[orchid cedar]

That's how lathes and machine tools work, those angular contacts are probably preloaded with far greater force than is applied by weight and even transient loading from grinding. So the lower bearing never completely unloads

Actually idk what bearings kafatek uses, that's what I would do though ^

[clear knot]

Me, too. But they use a single unloaded dual row bearing.

5206-2rs

There is no lower bearing. just the one.

[orchid cedar]

Smh no preload, that's pretty surprising

On that short a shaft too

[clear knot]

Yep

Like I said.

Not entirely ideal.