[PROMO] LeverCraft Ultra Grinder - Page 45

Grinders are one of the keys to exceptional espresso. Discuss them here.

#441: Post by zero610 »

The Levercraft edition was a couple years ago now. I wonder if Zhang incorporated any changes/updates from the Levercraft version. Clearly, based on the picture a few posts above, he's still developing/improving the grinder. If anyone in the community speaks Chinese and can get some answers from Zhang, I'm sure it would be appreciated by all...


#442: Post by protossw512 replying to zero610 »

I reached out to Zhang last week trying to order one. He said there is no 110V version available anymore. :(


#443: Post by BaristaBob »

That's true, but can't you use a step down transformer to go from 230v to 110v?
Bob "hello darkness my old friend..I've come to drink you once again"

Supporter ♡

#444: Post by leozava »

protossw512 wrote:I reached out to Zhang last week trying to order one. He said there is no 110V version available anymore. :(
Did he give any details on recent updates? Or if he did anything to fix the lower burr carrier issue

User avatar
Team HB

#445: Post by Jake_G »

Here's a brief synopsis of the issue with Zhang's grinders (as delivered to Eric in 110V trim):

Zhang clamped the burr carrier to the motor shaft through a deep groove ball bearing. What I mean by this is that you take the servo motor and place the lower plate atop it that has a bearing pressed into it. The outer ring of the bearing sits happily in the lower plate and centers the plate on the motor shaft, while the lower plate itself is bolted to the motor by way of the linear bearing posts, which rest in recessed bores in the lower plate. So the outer ring of the bearing is supported by the lower plate, but the inner ring is just floating on the motor shaft. It is centered on the shaft, but not supported by it.

A small washer sits atop the inner ring and then the burr carrier rests on that washer, and a nose cone with a central bolt clamps the burr carrier down. The central bolt is threaded into the motor shaft, so it pulls the burr carrier down by pulling up on the motor shaft.

"How hard does it pull up?" You ask.

Well, the screw is right hand thread (clockwise to tighten) and the burrs turn counterclockwise, so every time you grind, the force of grinding the beans tightens that little screw just a little bit more. The only thing making the burr carrier spin is the screw itself. Here is a little diagram showing the way Zhang shipped the original batch of parts:

The Blue arrows show the downward force applied to the carrier through the nose cone and the screw (not shown).
The Red arrows show the equal and opposite upward force that the shaft experiences, and you can see that the only thing providing that force is the upper bearing in the motor itself. When Eric assembled the first LCU, the no load amps on the motor was 7.0, as measured by a Kill-A-Watt plugged into the wall socket.

We discovered the root cause of the issue by first noting that the carrier was nearly impossible to turn by hand when assembled, but when the nose cone was loosened, the motor spun freely.

The first batch of LCU grinders that corrected this issue did so by supporting the nose cone against the shoulder of the motor shaft, which completely unloaded the motor bearings. Here is the free body diagram showing that initial design:

This design worked great and dropped the no load amps down to almost zero. The grinder worked! And Eric shipped them. You can see that the forces are now entirely contained within the end of the motor shaft. The force to turn the burrs is now translated through the shoulder on the shaft, through the spacer and into the lower carrier directly. Which means the central screw is now only clamping and not trying to twist the lower carrier. As such, it no longer self-tightens as you grind.

However, there was an unintended consequence of unloading the bearings. Motors have bearings that are preloaded from the factory using something called a split wave spring washer. These are mounted somewhere between one of the bearings and the motor housing, and take up the internal clearance in the bearings. They reduce sliding and skidding of the rolling elements in the bearings smd prolong motor life. They also compress under load.

The motors came assembled from the motor shop with 110V windings, ready to go when Eric received them, so it took us a wee bit of time to understand what was going on when an early adopter reported that when they ground hard beans, it seemed like the grinder was set to a coarser setting than with softer beans. We mounted a test indicator to an Ultra on the bench and measured that the lower carrier was in fact moving - a lot, when the funnel was tightened further than what we initially considered "locked". What we found was that this split wave spring washer is installed under the upper bearing in the motor, and when you put hard beans in it, the washer compressed and the lower carrier moved down with the motor shaft. When you were done grinding, the washer expanded and it left initial owners perplexed.

After we identified the issue, Eric and I devised a solution that required measuring the amount of downward movement on each individual grinder and machining a new spacer and bearing combination exactly to length to limit washer compression to no more than 10 microns. We retained all the benefits of unloading the motor bearings, but picked up the rigidity that Zhang had with the original design by eliminating the movement of the upper bearing.

Here's what the end solution looks like:

Red and Blue arrows represent the force vectors clamping the carrier to the shoulder on the motor shaft.

Orange and Green represent much smaller forces limiting the washer compression. This was accomplished by having a lower carrier support spacer that was of a known length, and known to be long enough to allow us to measure the remaining spring compression when the funnel was tightened completely by measuring the lower carrier movement. A new spacer was then cut to the exact length to limit that movement to between 5 and 10 microns.

Having movement of some sort here is key. If we set movement to zero, we could have undershot the spacer length and had the same bound bearing situation as we received from Zhang. Too long and you get excessive movement. The 10 micron limit allows us a window to work with and ensures each grinder would enjoy long bearing life and perform under extreme conditions without issue.

Hope this helps!

LMWDP #704
★★ Quite Helpful

Supporter ♡

#446: Post by leozava »

Thank you for the detailed reply Jake! The diagrams really helped me understand everything. So would it be safe to assume that I wouldn't have any of those issues if I ordered a 220v version straight from Zhang?

Also, is it the motor and the driver that are 240v? I guess my question is, could the original motor be kept and the driver swapped out for a 120v unit?

Team HB

#447: Post by ira »

If the manufacturer makes a replacement supply for the same motor that's 120V then it should be as simple as just swapping controllers.


#448: Post by NelisB »

(never mind my post) sorry

User avatar

#449: Post by ducats »

Jake_G wrote:Here's a brief synopsis of the issue with Zhang's grinders (as delivered to Eric in 110V trim)...
Post of the year!

User avatar
Team HB

#450: Post by Jake_G »

leozava wrote:So would it be safe to assume that I wouldn't have any of those issues if I ordered a 220v version straight from Zhang?
Well, being that all of these issues were a direct result of how Zhang designed the original Ultra, no. He may have made revisions to the carrier mounting, but when I brought up the above issues, his response was "Isn't that what bearings are for?". So, I'm not convinced.

That said, if you run a 220V unit on 220V, the idle current should be about half as it is on 110V, so ~3.5A instead of 7. I'm sure that the 220v models have enough grunt to push through that, but eventually the bearings will cry uncle.
ira wrote:If the manufacturer makes a replacement supply for the same motor that's 120V then it should be as simple as just swapping controllers.
The drive (or controller) is agnostic to the operating voltage. The motor largely doesn't care, either. The caveat is that 220V motors rated for 2500 rpm will only hit around half that speed on 110V due to the voltage constant (kv) of the motor. Interestingly, the 1500 rpm 110V motors have a very similar kv to the 2500 rpm 220V motors, but do so with a much larger rotor and higher stall torque. That said, there are a few internal parameters on the drive that need to be tweaked if you intend to run it on a non-native voltage. Namely the minimum voltage if going from 220V to 110V, and the maximum voltage if going the other way.


- Jake
LMWDP #704