Close

The $65 Chinese Rotation Mount Scam

A project log for Step-and-Scan Digital Microlithography

submicron-scale patterning of silicon and dielectrics

helgehelge 05/01/2018 at 16:320 Comments

There are some quality metal parts from China. There are some acceptable parts (like the kinematic mount I analyzed here https://hackaday.io/page/3711-29-chinese-4-axis-kinematic-mount-what-to-expect ) and then... there is this "R Axis Manual Rotating Platform Precision Bearing Linear Stage Load 29.4N 60mm":

Let's start with the few things that are actually right with this interpretation of a rotating stage. There are dimensional drawings:

While the micrometer adjuster is tiny and too close to the plate, it works ok. It even has a ball nose as you'd expect.

And that's about where it ends with the rainbows and the unicorns.

The thumb screws are flimsy and their fit is horrible. Since at least the center one that locks the precision rotator to the turntable seems to be turned on a lathe I assume it just came out way undersize.

Next obvious thing: the top hole pattern is off by 5-ish degrees. While this might be deliberate, you'll have to create a rotated hole pattern if you want to bolt something to it that lines up with the scale. Plus I'm not sure if the scale is just scribed in a random orientation and every part is different. Good luck with that.

But don't worry about the scale marks. You won't be using them anyway.

Seems ok?

No.. it's not. While this was supposed to come out looking like a vernier scale I wouldn't know where to put the zero on that thing.

While these issues are just superficial, let's have a look at how "precision bearing" is being interpreted. And nothing speaks precision like these butchered, rusted  countersinks:

Let's compare this to the auction picture:

Something undoubtedly changed anlong the way. But wait, there is more. I'm suspecting that what makes this thing have a precision feel is nothing but a generous helping of this glorious stuff (oh so Very Press):

Undoing the four PH0 screws makes the steel plate come off.

This doesn't look much like a precision bearing, does it? It may be a "bearing surface" but it sure isn't lapped and neither is it precision (such tool marks are usually 10+ µm in roughness). Then again, why would it need to be lapped if it rubs against the paint of the steel disc, getting loaded with paint particles during use. The center is threaded and four set screws lock it in place, also annoyingly backing off the back disc as you tighten them which makes the assembly loose once more. Fun fact: the T-shaped part attached to the ring is a powder metal part. Maybe that's why the thread is so sloppy. Back to topic. Below you can see the bottom reference surface of the turntable.

I then took the rest of it apart, revealing the whole mess. I'm not sure if I'd call the aluminium anodized judging by how it chips away at the edges. Here's the last positive thing of note: they actually relieved the contact surface. I didn't measure the inside and outside residual ring heights but I'm guessing it'll want to make contact on the inner one as well. *sigh*

The conical steel? ring inside is slightly raised and pivots a bit and when the center thumb screw is tightened, it rotates against the main body to translate the 3-4mm of usable travel of the adjuster into some +/-2° of fine rotation while the turntable still slides on the face of the main plate.

The cone also allows it to self adjust or compensate for coplanarity error of the whole clamped stack as the stage rotates (doesn't work in practice once you adjust for zero backlash). The plate looks like it had already been mounted to something so my guess is that it's being screwed onto a face plate and subsequently the conical inner surface is turned to match the ring.

Now whitness the craftsmanship.

Uh yeah, dremel away at those burrs. And that's no MoS2 or graphite grease, mind you. That's aluminium particles mixed with grease for some extra crunch.

I sanded and cleaned the center thumbscrew so it wouldn't bite into the aluminium surface of the turntable.

After that I wiped down all the grit and junk with ethanol, then with some light oil and finally applied a controlled amount of fresh grease and put everything back together. Getting the bottom steel plate tightened to just the right amount so it comes out convincingly after the setscrews are finger tightened again is not easy.

With everything back together I'm going to leave it like that for now. I'm hoping static friction will hold the turntable in place so it doesn't move around on me during operation when all mechanical adjustments are made.

Setting the rotation will need to be the first step in the alignment procedure and cumulative tilt errors will be compensated with the 2-axis kinematic mount not shown here. The height will be left to the objective z stage and offsets are irrelevant thanks to the XY stage.

Conclusion? Next time I'll grudgingly fork out $100+ for a quality surplus part and move on.

ps.: Constructive Citicism:

It is what it is. The base plate top - to - turntable bottom bearing contact is the main locating feature in this assembly. The conical ring holds the turntable down and locates it laterally, so clean surfaces and a consistent preload are key.

* take the whole thing apart, degrease and inspect. Stone raised burrs if present, don't sand the aluminium bearing surface (alumina or carbide particles will get embedded into soft metals and cause scuffing)

* add a wave spring washer between the inner ring and the bottom plate when you put it back together. The threaded stud has approx. 30mm diameter, the conical ring comes in at 45mm and the pocket is around 50mm in diameter. Get multiple thin spring washers with >=32mm ID and < 50mm OD - they must contact the ring which is between ~35 and 45mm . The washers can be stacked to empirically determine the ideal preload. You don't want more than 10 Newton and there are only a few tenths of a mm to work with anyway. Keep in mind that this preload must not translate into friction larger than the torque caused by the spring opposing the micrometer or the turntable will get stuck, messing up the fine adjustment capability.

Discussions