Went through it looking for some mention of the blocked coil test setup. I am bugged about that resonance.
Why was there a resonance at all? Did they do the test with a magnet, or did they substitute a demagnetized magnet.
I would have thought that the test would be more appropriate with no gap field, and I suspect that would have made any resonance moot.
They did not specify, I wish they had.
The IMD writeup in 2.1 was spot on.
Jn
Why was there a resonance at all? Did they do the test with a magnet, or did they substitute a demagnetized magnet.
I would have thought that the test would be more appropriate with no gap field, and I suspect that would have made any resonance moot.
They did not specify, I wish they had.
The IMD writeup in 2.1 was spot on.
Jn
Yeah you get some great deals at Ikea 😛
Dumb question time, what is resonating here. I thought I understood until I looked at the paper again and started drooling. Are they saying the precision test jig was resonating?
Naughty question time. This was an actual peer-reviewed paper. Scores out of 10 to the reviewer (given he was likely an audio type who wouldn't know an undulator if he walked into it)? You've picked fewer holes than I would have expected so unless you were short on time it's not too bad. And I've learned something, even if it's that I know even less than I thought I did when I got up 🙂
They said a precision positioning system to hold the coil at different locations in the gap. First sentence in 3.2.
Last sentence in 3.2 they say a mechanical resonance around 4khz. So I suspect they had a magnet producing gap field, and the coil holder was resonating. Which is why they should have demag'd the magnet. I assume they could have just put it in an oven.
There were not many holes, the talent in doing that paper was quite good.
Had I been a reviewer,I would have just pointed out the issues I noted here.
Jn
Argggh..sitting here trying to learn freecad to make parts on my cnc machine. Not there yet, but learning. I can create parts, but don't know how to create a Gcode path yet. Might just call it a day and make some food.
Last sentence in 3.2 they say a mechanical resonance around 4khz. So I suspect they had a magnet producing gap field, and the coil holder was resonating. Which is why they should have demag'd the magnet. I assume they could have just put it in an oven.
There were not many holes, the talent in doing that paper was quite good.
Had I been a reviewer,I would have just pointed out the issues I noted here.
Jn
Argggh..sitting here trying to learn freecad to make parts on my cnc machine. Not there yet, but learning. I can create parts, but don't know how to create a Gcode path yet. Might just call it a day and make some food.
Last edited:
No.
Thank you for the link, it was a page turner..
Lest you think otherwise, I have no life..😀
A foot of snow, my aching back. At least it was light.
This cad program...sometimes I think it would be easier to code my own Gcode generator. I'm still trying to figure out how to cut an escape wheel without chipping out the teeth. I use three layer laminated hardwood blanks, and I have to worry about grain orientation as the outer layers will tend to chip if there isn't wood behind the cut.
Jn
Thank you for the link, it was a page turner..
Lest you think otherwise, I have no life..😀
A foot of snow, my aching back. At least it was light.
This cad program...sometimes I think it would be easier to code my own Gcode generator. I'm still trying to figure out how to cut an escape wheel without chipping out the teeth. I use three layer laminated hardwood blanks, and I have to worry about grain orientation as the outer layers will tend to chip if there isn't wood behind the cut.
Jn
Fusion360? It's gotten a lot more restrictive (like they want to make money or something crazy like that) but the CAM is pretty decent. I haven't started a serious search for a FOSS CAM package yet. CAM is where things start getting interesting in mechanical design.
Twill be interesting. Wooden clock gears are a mix between epicycloidal and modified teeth profiles for minimum friction. Not sure if normal cad/cam programs can do that. I may have to just make an aluminum form for the pin router.
Jn
Jn
I've done a lot of 3d objects with Openscad. I've used an Autocad work-alike for 2D stuff, but it (Autocad and thus the work-alike) is allegedly extra difficult (moreso than other 3d cad progams) for 3d (I learned Autocad 30 years ago, I recall it was a kinda steep learning curve just for 2D, but you can do things several ways, and the interfaces are consistent once you learn them).
But yeah, openscad is mathematical, you specify sizes with numbers and objects with structures that resemble C functions, it's good if you can (just like anything) figure it out, you can array things around a circle, and I bet you can do the math to shape a gear tooth to your liking. One bad thing, the Cube function is badly named, it lets you give it x, y, and z dimensions for sizes, and of course if they're different, it's no longer a cube. And of course much or most of the stuff on Thingiverse is done in Openscad.
Wooden clock gears, are you making a clock like John Harrison's?
The Clock That Changed the World (BBC History of the World) - YouTube
But yeah, openscad is mathematical, you specify sizes with numbers and objects with structures that resemble C functions, it's good if you can (just like anything) figure it out, you can array things around a circle, and I bet you can do the math to shape a gear tooth to your liking. One bad thing, the Cube function is badly named, it lets you give it x, y, and z dimensions for sizes, and of course if they're different, it's no longer a cube. And of course much or most of the stuff on Thingiverse is done in Openscad.
Wooden clock gears, are you making a clock like John Harrison's?
The Clock That Changed the World (BBC History of the World) - YouTube
I use this diy (*) kneeling chair for three months now.
It has saved my back.
The 9 Best Kneeling Chairs of 2021
(*) Made by an Industrial Design student for a project of his studies
George
Attachments
Ed, it’s the former.
My body is 3D shrinking (belly resisting 😀)
From the concept's creator:
http://www.opsvik.no/download.aspx?object_id=F5D972A4201B4D6BA6D4B393D721AF27
George
My body is 3D shrinking (belly resisting 😀)
From the concept's creator:
http://www.opsvik.no/download.aspx?object_id=F5D972A4201B4D6BA6D4B393D721AF27
George
Benb, I am using openscad for all my 3D print designs. I use it to make the STL, then cura to make the Gcode file for the printer.
I haven't seen an option for a cnc file, just slicing.
I've been putting up various tool holders on thingiverse, and finding some neat thinks I've printed.
Clock, kinda like Harrison's, but more ornamental less accurate. I use 1/8th thick 8 by 8 hardwoods epoxied to 1/4 thick. Grains are at 90 degrees, this 3 layer resists warping due to shrinkage and humidity.
The difficulty with the escape wheel is, the very last tooth to be cut will chip out at the tip. I have to make sure the last tooth has the outer grain always aimed so that no grain goes from one side to the other. And I've learned that teeth prefer a climb cut. I end the cut at the tip, but a root finish would be less inclined to tear out, I just don't want to start at a root without a fixture surface to guide, that would be dangerous.
Also, I've learned that my pin routing fixtures must have sufficient mass. Light fixtures can tend to chatter, so I use 1/4 thick aluminum and make the fixtures at least 8 by 12 inches.
Harrison was awesome.
Jn
I haven't seen an option for a cnc file, just slicing.
I've been putting up various tool holders on thingiverse, and finding some neat thinks I've printed.
Clock, kinda like Harrison's, but more ornamental less accurate. I use 1/8th thick 8 by 8 hardwoods epoxied to 1/4 thick. Grains are at 90 degrees, this 3 layer resists warping due to shrinkage and humidity.
The difficulty with the escape wheel is, the very last tooth to be cut will chip out at the tip. I have to make sure the last tooth has the outer grain always aimed so that no grain goes from one side to the other. And I've learned that teeth prefer a climb cut. I end the cut at the tip, but a root finish would be less inclined to tear out, I just don't want to start at a root without a fixture surface to guide, that would be dangerous.
Also, I've learned that my pin routing fixtures must have sufficient mass. Light fixtures can tend to chatter, so I use 1/4 thick aluminum and make the fixtures at least 8 by 12 inches.
Harrison was awesome.
Jn
Last edited:
Tale of yore: in 1978 I worked at TI Stafford, designing NMOS DRAM chips. To characterize and debug our ICs, we built our own active probes (!). They were used to microprobe metal signal wires (3 um linewidth) on the wafer. Circuit was 2 x 2N4416 RF JFETs in metal cans, on a 1x1 inch phenolic perfboard, at the end of an articulated arm home made out of 1/2 inch slab acrylic plastic. A 3 axis micromanipulator purchased from a biology lab, did the fine grain motion. Vernier dials on all three screws.
Source follower + current source + degeneration resistors (one of them a trimpot for offset null)
Source follower + current source + degeneration resistors (one of them a trimpot for offset null)
At UMass I worked as a technician at the Five College Radio Astronomy electronics lab. We created devices for the then SOA 14m millimeter-wavelength telescope. My first project there was fun: a reference black-body controller, it swung the black body reference radiator into the optical path of the telescope for calibration.
Temperature had to be controlled +/- 0.05°C. I designed multiple feedback and feed-forward paths to minimize overshoot and sawtooth variation. I could not stabilize it even in still air (damned convection!), and it ultimately only stabilized in the vacuum of the telescope front-end.
Working on a tracking telescope like that was fascinating; everything to be mounted on the scope had to be weighed and it's center of mass from the telescope's measured and an appropriate counterbalancing performed to keep the huge thing balanced.
The 14 m dish surface consisted of small adjustable panels, and they had to all be within 0.3 mm of the focus since the telescope was receiving down to 3 mm signals. It was the first time I had seen a laser range finder.
I loved that 'scope, and it was torn down in 2011 much to my chagrin...left all my friends at FCRAO when I moved to NC to get a job...and shortly afterward they all left FCRAO and formed Millitech, Inc. and are now rich...dammit...it was not to be the last stupid career choice I made. I chose audio engineering...idiot!
Cheers!
Howie
Somehow I thought STL was a "universal" 3d file format and you could easily use it for CNC, but I suppose it's mainly for additive/3d printing. The local makerspace has a big CNC router (a one-off made of 80/20) that I haven't got around to using, but I recall it runs on CamBam software, which might be worth looking into. I have many physical objects to make in my bucket list (such as "classroom slide rule," would love to take that to a Maker Faire), but haven't done a deep dive into all that's available to do it.
But yeah, Harrison was amazing.
But yeah, Harrison was amazing.
STL are stereolithography files. For transfering 3D cad files between different systems, step (.stp) are the one to use🙂
.iges as well, both are good for importing the model into a CAM software package associated with the CNC machine and generating the appropriate toolpath/work flow.
The learning curve is pretty steep for any sort of interesting parts, so anyone who makes it look easy had gone through a lot of effort to get there.
The learning curve is pretty steep for any sort of interesting parts, so anyone who makes it look easy had gone through a lot of effort to get there.
A carpenter..
I believe his H2 (or H3) met the requirements for the prize, but he refused it twice IIRC. Only after H5, a rather large pocket watch, did he accept the prize money..
Uh, hello King Gustaf, I know you want to give me the Nobel Prize, but ehhh, my work could be better...Maybe next time..😕
sheesh
..I believe he invented the gridiron compensation method and cycloidal cheeks for compensation of circular error. When I was on the tour at the observatory in Greenwich, most of his mechanisms were on display, amazing constructs of brass.
ps..my mini CNC is 300 by 180 by 45mm working volume. I am constructing one roughly a meter square by about 20 cm height. Given the lack of space, I am making it to be vertical and on wheels.
Last time we had a Maker fair, my friend enlisted me as an official to help out. "all I got was this lousy T-shirt.." I was thinking my cnc could be demonstrated at a fair, but the scary aspect of spinning a router bit into wood with bystanders watching, not a good insurance thing.
jn
Last edited: