This guy is a legend. His whole series on building this along with other videos is worth watching on YouTube. It amazes me how some random guy simply drops a sub um manipulator diy on YouTube and disappears into the ether. No promotions nothing.
I was surprised to see this here and was wondering if he updated anything from a year ago when I first watched it. Doesn't seem like it but I haven't dug into it. Really nice stuff and I'm hoping he got a job from it or got really busy doing stuff around it that pays the bills.
Curious how temperature and so forth are managed --- that was a notable effect back in the days when folks were doing things such as engraving The Lord's Prayer on a piece of metal type.
I'm not an expert but the video is very comprehensive. There is closed loop control that is extremely precise in a relative way. In this domain it's all relative anyway because there's no GPS so to speak. It's all reference frames and 3-2-1 locality. [edit: ahhh at 21:47 in the video he says (paraphrasing) "there is offset due to thermal drift of the plastic parts as they get warmed by the motors". ]
In general, platforms are placed in a temperature-controlled/heated draft-guard chamber, on a vibration isolation table, and use uniform materials that do not deform much in the narrow operating temperature range (+-0.5'C).
The stage is not the primary design constraint, but rather it is the position sensors that still cost. At a certain point, some realize the pi-usa stages lofty price was actually fairly reasonable in retrospect.
Interesting. I researched what you referenced. This is a fantastic "hobby project" level of engineering but if one is truly tasked with this level of metrology there are still many aspects that need to be addressed and corrected for. Usually that means there is a lot of money on the line and one can't mess it up. Hence the high price tag of established industry accepted solutions. -whew- I wrote that myself.
They also have a manual, non motorized version. I am building this currently for eventual use, with modifications, in a point diffraction interferometer.
I wonder how it compares to a deltabot, since those are often used for pick-and-place which requires pretty fine resolution. Seems like a delta would have way more range of motion than this.
This guy is a legend. His whole series on building this along with other videos is worth watching on YouTube. It amazes me how some random guy simply drops a sub um manipulator diy on YouTube and disappears into the ether. No promotions nothing.
I was surprised to see this here and was wondering if he updated anything from a year ago when I first watched it. Doesn't seem like it but I haven't dug into it. Really nice stuff and I'm hoping he got a job from it or got really busy doing stuff around it that pays the bills.
edit: direct link to the video: https://www.youtube.com/watch?v=MgQbPdiuUTw
Curious how temperature and so forth are managed --- that was a notable effect back in the days when folks were doing things such as engraving The Lord's Prayer on a piece of metal type.
I'm not an expert but the video is very comprehensive. There is closed loop control that is extremely precise in a relative way. In this domain it's all relative anyway because there's no GPS so to speak. It's all reference frames and 3-2-1 locality. [edit: ahhh at 21:47 in the video he says (paraphrasing) "there is offset due to thermal drift of the plastic parts as they get warmed by the motors". ]
In general, platforms are placed in a temperature-controlled/heated draft-guard chamber, on a vibration isolation table, and use uniform materials that do not deform much in the narrow operating temperature range (+-0.5'C).
The stage is not the primary design constraint, but rather it is the position sensors that still cost. At a certain point, some realize the pi-usa stages lofty price was actually fairly reasonable in retrospect.
Best regards, =3
Interesting. I researched what you referenced. This is a fantastic "hobby project" level of engineering but if one is truly tasked with this level of metrology there are still many aspects that need to be addressed and corrected for. Usually that means there is a lot of money on the line and one can't mess it up. Hence the high price tag of established industry accepted solutions. -whew- I wrote that myself.
Reprap micron is doing something similar https://github.com/VikOlliver/RepRapMicron
Someone posted that today: https://news.ycombinator.com/item?id=50141076
That was me, I realised after I commented that I should make a post about it.
It's also possible to DIY piezoelectric stick-slip actuators with potentially sub-nanometer(!) resolution, albeit at greater expense (~$1k) [1].
[1] "Low-cost, open-source XYZ nanopositioner for high-precision analytical applications", https://doi.org/10.1016/j.ohx.2022.e00317
They also have a manual, non motorized version. I am building this currently for eventual use, with modifications, in a point diffraction interferometer.
https://youtu.be/-mgqq0qdDRM?is=nFvtZprNsjMn3D_B
Wow strange to see this on here. I just spent the weekend working on my build of this.
I wonder how it compares to a deltabot, since those are often used for pick-and-place which requires pretty fine resolution. Seems like a delta would have way more range of motion than this.