A kinematically coupled, nanometer-resolution piezo focus stage | Hacker News Reader

A kinematically coupled, nanometer-resolution piezo focus stage

bunniestudios.com

71 pointstodsacerdoti16 comments

Related items of interest:

Dan Gelbart on Flexures: https://www.youtube.com/watch?v=PaypcVFPs48

Design of a kinematic mount: https://practicalprecision.com/kinematic-mount-design/

Dan Gelbart, a great. Adding MIT Alex Slocum’s kinematic and flexure design resources: http://pergatory.mit.edu/resources/fundamentals.html

Also this video of a JWST-inspired positioning flexure: https://www.youtube.com/watch?v=5MxH1sfJLBQ

I had to immidiately think of the JWST as well, the idea being to use basically a compliant mechanism as a lever that turns a big weak movement into a tiny stronger one.

There was also this paper [0] making the rounds in 2022. Ultrahigh precision equipment for hobby level prices makes me super excited to be alive right now.

[0] https://www.sciencedirect.com/science/article/pii/S246806722...

I've been enjoying these posts about the IRIS project. I would love to see some quantitative measurements of its performance. If it works for the autofocus purpose, that is great. But I'm skeptical of calling this kinematically coupled or nanometer resolution with vhb taped non spherical balls and no preload. I wouldn't expect that to repeat very well. Also, I'm having a hard time thinking about whether you can actually use the maxwell as a trim/tram table like this, is this known? The piezo's are going to be moving in 3 slightly different axes and surely with some parasitic motion so the ball's position won't stay in radial symmetery. And they have to slide in the groove so there will be stiction. I think a linear stage at 1um would suffice because the correlation between images should be higher

Totally, it would be interesting to put an interferometer on this thing

To give an idea how accurately these things can be used, the quantum computer group at the local university use piezo actuators to move their ions around with sub-100nm resolution (to create fast interconnects between quantum chips, for reasons of modularity). Got lucky enough to have a demo/playtime (which involved pushing rubidium ions around a few nanometres at a time) before they went commercial [0] with it. As Feynman said, "plenty of room at the bottom".

[0] https://thequantuminsider.com/2022/11/02/universal-quantum-s...

I like that it's not a product for regulatory reasons. Now I wonder about the boundaries on "productionizing" open hardware without selling it.

Let's suppose the creator makes it dead simple to make. Solves problems that would tie up any replication effort for months. Replication now takes a week of effort for a pair of hobbyists with a 3d printer, a CNC machine, and a soldering iron. Setup for mass production setup now takes a month for a well appointed electronics factory.

The creator has eliminated all production problems, while skipping all the steps required to earn a revenue.

Does this increase the risk of a cease and desist? After all, G-men might be worried that a foreign power will replicate the tech without paying an American corporation!

> Does this increase the risk of a cease and desist? After all, G-men might be worried that a foreign power will replicate the tech without paying an American corporation!

Many of the regulations he talks about aren’t related to the United States or corporations at all. We like to poke fun at America, but in my experience it has been much easier to launch physical products in America than many other countries. The blog gives one example of requiring “UKCA” stamped somewhere to deal with some post Brexit thing. Then there’s the matter of customs and import issues, which are a nightmare to navigate in many of the various countries you might ship to other than the US.

> The creator has eliminated all production problems, while skipping all the steps required to earn a revenue.

Even with full instructions, production problems would actually be the majority of the work. It’s still quite difficult to set up manufacturing for something like this. Furthermore, it’s nearly impossible for someone to perfect a manufacturing process and document it without having done the manufacturing and learned from experience.

Finally, the product described isn’t actually that novel. Anyone sufficiently motivated and capitalized to set up a manufacturing line would also have the resources to have developed something like this.

The comments in the blog read more like a commentary on the difficulties of low volume product logistics, not a suggestion that “G-men” are trying to crush someone’s hobby project so that an American corporation can have more profit.

I'm an industrial designer and manufacturing engineer who founded a reverse engineering firm. We use some nice techniques to virtually eliminate defects in our client production streams, even though we have basically no tacit knowledge of their production back ends. What we do understand is our design geometry and its quality metrics.

In our case, simply controlling the quality of our geometry is enough to stamp out defects and rework, which is why our clients use us as a turnkey solution for eliminating design risks.

Those kinds of quality controls and universal productions sims don't get developed for simple open source designs.

So yes I understand that productionizing a design traditionally depends mainly on accruing tacit knowledge.

At the same time I understand also how important quality controls in the design phase are for eliminating rework in the production phase. And we are increasingly able to plan designs that will enter into production flawlessly.

The scenario I'm asking you to envision, therefore, is not this project, but some hypothetical other scenario where the creator really supports larger scale efforts with novel quality controls that have traditionally been too expensive for open source.

I'm also interested in a political scenario where we're designing a purely software weapon, a new drone munition, or other dangerous tech. Let's even suppose we're doing this because we hope it will disrupt for corporations and governments.

The question then remains: when due this kind of work cross into dangerous territory, and incur outside risks. Where are the limits? How much can we expect to "get away with"?

Obviously the answer is "it depends' and "talk to a lawyer". But I'm curious about insights and stories folks might have here.

I’m doing this process for my open source farming robot [1], though not to dodge any patents I simply want people all over the world to have access to this technology for as cheaply as possible, and someone in Bangalore or Nairobi will be able to get them more cheaply from a local fabricator than from some California company shipping overseas.

https://community.twistedfields.com/t/join-the-solar-farming...

This is awesome, and I'm going to look into your project!

I think you're doing it for the right reason. I hope we can get many more of the world's engineers thinking that way in the next decade. And I feel like humanity's future might ride on it!

Thank you! Yeah I think that “proprietary thinking” is really shooting ourselves in the foot as far as cumulative development of humanity. Community development of 3D printers after the patents expired showed me that proprietary single-organization patent encumbered engineering development is only one way to get critical engineering done, and it has many downsides. A major advantage of community oriented engineering, where you share your designs without restriction, is that people all over the world find ways to contribute to your design. You end up with a better result in the end, and in the process you’re lifting potentially billions of people out of poverty. To me it seems like it’s just the right thing to do aside from the fact that it provides real material benefit to the people that may have originally produced the ideas. But these concepts are not taught in schools, and there is a minor problem with financial allocation that needs to be solved with for example donations. I think it’s an extremely important mode of engineering for humanity to explore!

I had always thought that kinematic mounts consisted of the three balls on one side and then on the other side, each ball would rest on a flat, a v-groove, and a cone, respectively.

This design seems to have each ball in its own v-groove. Doesn't that require extraordinary machining precision with each of the three grooves?

I grew up in a world with fewer trade barriers, and without a DMCA. I remember when we were free to think, travel, and trade, and I liked it. I feel kind of sorry for the youth today, growing up in a world shaped by narratives of safety, fear, and ultimately, control. Also, climate change.

Reply on news.ycombinator.com