|
I've also though about making a clock but I have no idea where to find plans. We do wire edm here so making the gears within less than .0002" or so would be no problem. Just need to find some good cad drawings or something. Edit: We have an indexable punch maker thing for our surface grinder. It's pretty heavy so gets tiring to spin manually if you're just grinding round parts. So we hacked together some poo poo and made it automatic and it works pretty fuckin good. 
A Proper Uppercut fucked around with this message at 17:45 on Jun 19, 2018 |
#
?
Jun 19, 2018 17:28
|
|
|
|
| # ? May 9, 2024 21:42 |
|
|
Brekelefuw posted:Everyone should follow and subscribe to this guy on YouTube. His videos are funny and his skills are fantastic. This was awesome, thanks!
|
|
#
?
Jun 19, 2018 18:49
|
|
|
A Proper Uppercut posted:So we hacked together some poo poo and made it automatic and it works pretty fuckin good. What's the brand on that? Does it use a collet or jaws?
|
|
#
?
Jun 19, 2018 20:06
|
|
|
Yooper posted:What's the brand on that? Does it use a collet or jaws? We got it from Gromax, who pretty much just imports Chinese stuff and slaps their name on it. But we've bought grinding and EDM vises from them and it's all been decent quality stuff for the price, this rotary fixture included. It uses a v block actually, that you can adjust up and down (and left and right I think) to grind some weird shapes, though we mostly use it for just grinding round stuff.
|
|
#
?
Jun 19, 2018 21:03
|
|
|
A Proper Uppercut posted:We got it from Gromax, who pretty much just imports Chinese stuff and slaps their name on it. But we've bought grinding and EDM vises from them and it's all been decent quality stuff for the price, this rotary fixture included. Ahh gotcha. We use Harig workheads that are very similar. Ours has a 5C collet setup but they make one with a V block like that. I'm always on the look out for a great solution for holding parts that is high volume and high accuracy. It sucks buying one and spending a shitload of time modifying it.
|
|
#
?
Jun 19, 2018 23:30
|
|
|
MohawkSatan posted:I would suggest watching Clickspring on youtube for an idea of what you'd be getting into. In theory you can manually machine a clock, but it's gonna be a multi year process. On top of that, you're entering a while new world of pain with regards to tolerances. Welcome to thinking of 0.003" as plenty of space. As for books, if you're doing manual machining, I've got nothing to advise. I think I'm a bit more optimistic about the possibility of this. Clocks often have surprising amounts of slop. Not just by machining standards, but by actual play with them by hand you can feel the entire gear train going back and forth standards. This is because their primary concern is very low friction so they can take usable power off a small weight/spring while not depleting it quickly. The backlash of a loose geartrain isn't a problem as they only ever run in one direction. They also have very low actual loads, and move quite slowly. Some of their parts, like the escapement, do have dimensions where the difference between working well and not working at all would only be a few file strokes. The good news (at least for this project) is that file and trial fit is exactly how these parts are typically made. MohawkSatan posted:Plus in either case to make a clock you're gonna need some very good measuring tools, so you can tack another grand. Yes, but also no. Your timing is set by the pendulum, and then though the gear train. The gear train's ratios are fixed by number of teeth, so even if the diameter is slightly undersized or oversized, they still have the same ratios. You'd have to be off by a full tooth to make an error that wouldn't cancel out while it runs. This leaves the pendulum. Making a pendulum to sufficient tolerances to keep good time while being fixed is very difficult, so good designs typically have a means for adjustment. If it's going too fast, you move the weight down, too slow, and you move it up. On a beginner clock you'll just give yourself a wider range of possible adjustments. Low accuracy will mostly waste power and prevent the clock from running as smoothly as it could. In the end requiring a heavier weight/spring, or a lower gear ratio needing more frequent winding. It's not hard to make something that lasts a day. It's hard to make something that lasts a week between windings. All of this is, of course, talking one off/handmade clocks. Factory clocks are a whole different story, as you need interchangeable parts. To put it all another way, people make wooden clocks with wooden gear trains using woodworking tools that will keep time. Woodworking tools have no where near the precision of metalworking tools, and the wood couldn't hold it even if the tools did. You do, I suppose, have to ask yourself what keeping time means. For a wooden clock, it's not hard to do <1min a week and ~1min a month is possible. Dial in the pendulum, and keeping humidity and temperature consistent becomes your biggest problem. That said, I'd never do this without a at least a scroll saw, or similar, otherwise you'll spend forever roughing things out. You'd also want a drill press. Definitely watch clickspring.
|
|
#
?
Jun 20, 2018 07:01
|
|
|
Thanks! I'd be very happy with a minute a week, or even a minute a day. I'm mostly interested in building one to see how they really work, not build a perfectly accurate clock. John Harrison made several very accurate clocks out of wood and invented the marine chronometer without knowing what a thousand of an inch was, so I figured it must be possible. I have a jewellers saw and bandsaw and drill press and am very well acquainted with their use-any books you could recommend on a the math/geometry behind it or even something with some plans?
|
|
#
?
Jun 20, 2018 13:17
|
|
|
Kaiser Schnitzel posted:Thanks! I'd be very happy with a minute a week, or even a minute a day. I'm mostly interested in building one to see how they really work, not build a perfectly accurate clock. John Harrison made several very accurate clocks out of wood and invented the marine chronometer without knowing what a thousand of an inch was, so I figured it must be possible. Going totally off memory here as I’m outside a pub caning down pints during the four minutes of British summer, but in one of Clickspring’s early clock build videos he mentions the books he is using, I think.
|
|
#
?
Jun 20, 2018 13:22
|
|
|
Kaiser Schnitzel posted:Thanks! I'd be very happy with a minute a week, or even a minute a day. I'm mostly interested in building one to see how they really work, not build a perfectly accurate clock. John Harrison made several very accurate clocks out of wood and invented the marine chronometer without knowing what a thousand of an inch was, so I figured it must be possible. Maybe check this page out http://www.woodenclocks.co.uk/downloads.html
|
|
#
?
Jun 20, 2018 13:43
|
|
|
Does anyone know anything about the feasibility of making your own harmonic drives? I've been looking more at making my own robots, and everything seems easy except the gearboxes, which need to be as close to 0 backlash as you can be. The reading I've done online suggests tolerances are really tight and I have read of a couple of failed DIY attempts. Could you do it on a mill/lathe? Failing harmonic drives I suppose you could use pulleys and pre tensioned Gates style belts. Second unrelated question, but someone said a while back Bridgeport J heads didn't have crash hot spindles. How do they compare with, say, a Deckel spindle? In the home machining world Deckels seem to be the gold standard for an accurate machine.
|
|
#
?
Jun 21, 2018 04:08
|
|
|
Mudfly posted:Does anyone know anything about the feasibility of making your own harmonic drives? Whilst you could make your own that's a horrific amount of work, what's the end result you're after? The robots I work with use harmonic drives and once worn they're scrap, refurb would literally be more labor and material intensive than replacement.
|
|
#
?
Jun 21, 2018 11:06
|
|
|
re:harmonic drive you may as well ask about DIY needle bearings.
|
|
#
?
Jun 21, 2018 14:55
|
|
|
What size/power requirements are you needing gearboxes for?
|
|
#
?
Jun 21, 2018 17:22
|
|
|
I thought it might be too hard, but I was after 0 to minimal backlash. I was making small desktop robotic arms for a school I work with and also for a bit of fun and the servos I used had gear trains and the backlash contributed to vibrations. I havent done any calcs but I would guess power requirements would be under 100w. That said, it would be fun someday to make something bigger - I have seen treadmill motors adapted for servo work with rotary encoders. I wanted to do everything on the cheap though, and harmonic drives are expensive.
|
|
#
?
Jun 22, 2018 01:41
|
|
|
If you don't need extreme precision or force, I've seen DIY harmonic drives that use 3D-printed rotors that seem to work pretty well. I imagine they'd work fine for a homemade robot or whatever. https://www.youtube.com/watch?v=5wLx6SDCeyI
|
|
#
?
Jun 22, 2018 02:34
|
|
|
IRT watchmaking: a big benefit of setting your sights so petite is that the machine tools are equally petite (and affordable). IIRC the Taig lathe was originally intended as a watchmaker's lathe and it's basically the smallest and cheapest "real" machine tool you can buy (if we're not including the humble hand file under that umbrella). You can get one brand new + tricked out, such as it is, for like idk $500 maybe, and the lathe + motor have such a small envelope that you can fit the whole assembly as-is in a big toolbox. You can't thread with it without doing some fairly intense modifications and the accessories are non-standard but it'll hold better tolerances than an import mini lathe will and its very modest capabilities shouldn't ever be an issue for watch parts. They can also be reconfigured for freehand wood turning, which may be complementary depending on the specific projects you're tackling.
|
|
#
?
Jun 22, 2018 02:45
|
|
|
I got bored last night so I made a cube from 304 50x5mm flat bar Bit rough but tbh I'm quite happy with it considering I'm only an amateur on the TIG cube
|
|
#
?
Jun 22, 2018 03:00
|
|
|
I need to make a drift punch that's oval in profile to stretch out and shape some off-the-shelf sterling silver beads; they come from the factory slightly undersized for my application and my existing technique of reaming em wider with standard taper punches from both ends and squeezing em into an about-right oval with parallel pliers is a time-consuming pain. It'll be a pretty small punch, the oval part dimensions the beads have to clear are something like .1" x.08", and I'd want the stretched beads to be like +.002 to +0.005ish bigger than that. 1) What's the easiest way to make the punch itself? Can't easily cyl grind or turn an oval (that I know of with our equipment, anyways) and given the scale + tolerances I'm not confident I could forge and/or grind it to spec totally by hand. Maybe I could grind a cylindrical punch slightly oversized from the widest oval dimension and then file + stone the oval section in? Or alternately use a surface grinder and indexing head + barstock and rough out the oval form with a bunch of calculated cuts and then do some hand-finishing to knock all the corners off and make it smoothlike? If I'm making one I'll probably make a couple so I don't have to grapple with this again once a drift is too mashed up to work properly, so something amenable to batch production would be swell. 2) How can I ballpark the springback in the bead once it's off the drift so I can appropriately oversize the punch? I made a blanking punch + die set and worked with a 1 thou assumed springback and that worked pretty well, but this is a much smaller part in a different material and it's 100% deformation with no shearing. I can pick a percent to oversize the punch by and hope I get good results but that ain't pareto-optimal. Ambrose Burnside fucked around with this message at 03:20 on Jun 22, 2018 |
|
#
?
Jun 22, 2018 03:17
|
|
|
McSpergin posted:I got bored last night so I made a cube from 304 50x5mm flat bar good lord the dickweed talking poo poo about your welds always nice to see someone who can only feel good by making other people feel worse
|
|
#
?
Jun 22, 2018 03:17
|
|
|
https://www.youtube.com/watch?v=Qir3QM1M7Nc a really nice video showing how an old-school scissor manufacturer from sheffield makes their scissors. seein em fine-tune the fit with hammer and anvil in a production setting in the year of our lord 2018 brings a tear to my eyes, it does
|
|
#
?
Jun 22, 2018 16:35
|
|
|
There is a beautiful video floating around somewhere of the Holland and Holland gun factory. They still fit everything with lampblack and a file until it fits perfectly.
|
|
#
?
Jun 22, 2018 16:45
|
|
|
Purdey has a good video and they do the same thing. But for $50k for a shotgun they can do whatever they drat well please. On the exact opposite end of things I'm speccing out a 3 jaw grinding chuck that's rated at 0.00002" of TIR. It's a thing of beauty. The only bummer is we need to make new spindles. Oh well. It's going to be beautiful.
|
|
#
?
Jun 22, 2018 16:56
|
|
|
True love is an Index MS-22 holding .025mm part tolerances on a backworking spindle I solve much more interesting problems as a controls tech now but dammit I still miss this beast. https://youtu.be/1ag8r60X_14 shame on an IGA fucked around with this message at 17:05 on Jun 22, 2018 |
|
#
?
Jun 22, 2018 17:01
|
|
|
I guess this is as good as anywhere to post it. But does anyone else here have experience with optical comparators in manufacturing? My job involves repairing them but i'm not well versed in usage.
|
|
#
?
Jun 22, 2018 17:40
|
|
|
frodnonnag posted:I guess this is as good as anywhere to post it. But does anyone else here have experience with optical comparators in manufacturing? My job involves repairing them but i'm not well versed in usage. I use them every ding dong day. They're pretty simple unless you're getting into measuring weird poo poo.
|
|
#
?
Jun 22, 2018 18:23
|
|
|
frodnonnag posted:I guess this is as good as anywhere to post it. But does anyone else here have experience with optical comparators in manufacturing? My job involves repairing them but i'm not well versed in usage. Only a little. ...whats there to repair its a light and some optics?
|
|
#
?
Jun 22, 2018 19:15
|
|
|
CarForumPoster posted:Only a little. you'd think this, and then you see the remarkable things idiot students can do without even trying, even to things generally too robust or simple to be meaningfully gently caress-uppable. anybody can chip or dent a 1-2-3 block, you never expect to see one get bent i know on our mitutoyo comparators people crank on the lil delicate handwheels like theyre pullstarting a lawnmower because they dont know the table's locked and assume that's how hard it is to move it; anything and everything threaded will get stripped by some muscledummy who puts his back into turning that extra-long hex wrench because it makes his delts pop, etc. the sky's the limit Ambrose Burnside fucked around with this message at 19:28 on Jun 22, 2018 |
|
#
?
Jun 22, 2018 19:20
|
|
|
frodnonnag posted:I guess this is as good as anywhere to post it. But does anyone else here have experience with optical comparators in manufacturing? My job involves repairing them but i'm not well versed in usage. Use one at work all the time. What do you want to know?
|
|
#
?
Jun 22, 2018 19:27
|
|
|
In volume production they're mostly for operators to quantify how much offset they need when the dimension already failed a go/no go and as a backup to dedicated comparator gages. Profile stuff like chamfers, thread depths, inside radii, groove widths. It's almost never a primary inspection tool because regular midshift checks are all done with go/no go and universals and first/last piece always goes to the marform lab
shame on an IGA fucked around with this message at 19:55 on Jun 22, 2018 |
|
#
?
Jun 22, 2018 19:49
|
|
|
We've got a pair and use them exclusively to check angles. All they do. All day. Angle angle.
|
|
#
?
Jun 22, 2018 19:52
|
|
|
Mostly i wanna see what people use so i can comment from the repair side of things for them.quote:i know on our mitutoyo comparators people crank on the lil delicate handwheels like theyre pullstarting a lawnmower because they dont know the table's locked and assume that's how hard it is to move it; anything and everything threaded will get stripped by some muscledummy who puts his back into turning that extra-long hex wrench because it makes his delts pop, etc. the sky's the limit Yep, just had a customer's story like this. Their comparator has motorized vertical movement, new guy starts measuring parts and on downward movement the table jams and starts making ungodly loud grinding noises and he keeps on measuring parts like this. All that was wrong with the machine was that it wasn't properly oiled so the shaft was binding and causing a gearbox to grind. frodnonnag fucked around with this message at 19:56 on Jun 22, 2018 |
|
#
?
Jun 22, 2018 19:53
|
|
|
lmao yeah the one and only line on our control plan that specifically called for the optical was a 30° chamfer E: An (relatively to everything within a 100' radius) ancient freestanding Spectra with 7-segment DRO shame on an IGA fucked around with this message at 20:00 on Jun 22, 2018 |
|
#
?
Jun 22, 2018 19:56
|
|
|
shame on an IGA posted:lmao yeah the one and only line on our control plan that specifically called for the optical was a 30° chamfer I know of comparators built back in like the 30s and 40s that are still used to this day for overlay checking. They are great for quick checks, we're just now getting into adding cnc capability.
|
|
#
?
Jun 22, 2018 19:58
|
|
|
Mostly checking form tools and chamfer Mills to figure out if they're correct and what their tip diameters are.
|
|
#
?
Jun 22, 2018 20:26
|
|
|
I was an engineering intern a company that uses them in production to check the profile of a ground part. Its a special cam lobe on one side and crimper/indenter on the other side. They have a custom overlay and it makes it really quick to check and the owner hates cap ex. Watching people use that and trying it out for myself a couple times is my only experience though.
|
|
#
?
Jun 23, 2018 01:54
|
|
|
Bought some stuff today!   Also grabbed some taper shank drill bits that I didn't get a picture of. Bonus: lovely pic of my lathe pre-restoration (about 8 years ago) 
|
|
#
?
Jun 24, 2018 00:38
|
|
|
I'm having a minor aneurysm or sth and accordingly can't successfully google this- is the ER16 spindle's M22x1.5 thread right-handed or left? Ambrose Burnside fucked around with this message at 04:54 on Jun 27, 2018 |
|
#
?
Jun 27, 2018 01:42
|
|
|
Ambrose Burnside posted:good lord the dickweed talking poo poo about your welds Lol I know right. Had to laugh. I'm in no way shape or form a professional welder, just an engineer with a penchant for making stuff, but I'm quite happy with what I made He shared an article/link to his welds and I mean they're nice but I'm certain he's a career professional welder, not some bloke with about 13 hours tig time ever building a cube entirely with fusion welds (bar maybe 2 of the corners which needed some filler)
|
|
#
?
Jun 27, 2018 02:43
|
|
|
https://imgur.com/gallery/ipWfcDF
|
|
#
?
Jun 27, 2018 04:02
|
|
|
|
| # ? May 9, 2024 21:42 |
|
|
e: quote is not edit, rookie move
|
|
#
?
Jun 27, 2018 04:52
|
|