Cut *any* thread pitch on a metal lathe
ฝัง
- เผยแพร่เมื่อ 9 ก.ค. 2024
- The threading table for my Vevor metal lathe was very inaccurate, so I set about to come up with a much more accurate thread cutting table. With five gears that can be swapped out, It's possible to approximate any arbitrary thread pitch with surprising accuracy. I also tried making some wooden gears to explore the possibility of just making custom gears for it.
I have uploaded my Python script and spreadsheets to github:
github.com/Matthias-Wandel/la...
My gear program: woodgears.ca/gear
Vevor mini lathe: s.vevor.com/bfQKTC - แนวปฏิบัติและการใช้ชีวิต
![กินอาหาร ถูก vs แพง!! จานละ 500,000 vs จานละ 50!! [Ver. 2024]](http://i.ytimg.com/vi/FF_HppKg4vs/mqdefault.jpg)
![[안방1열 풀캠4K] 베이비몬스터 'FOREVER' (BABYMONSTER FullCam)│@SBS Inkigayo 240707](http://i.ytimg.com/vi/_2HL-ogzzBU/mqdefault.jpg)
I have uploaded my Python script to github:
github.com/Matthias-Wandel/lathe-thread-gears
Thank you so much!
LOL - at this point you are sending machinists to hospital with a major headache.
Get Matthias to do a spreadsheet for your Maximat!
Not just machinists.
Why ? Because he wrote a few for loops that tested all the possible combinations of gears and found better ones ? Which part is supposed to give a headache ?
@@xl000 I think because machinists hate dead trees.
@@xl000 Dude! Lighten up! Stefan was joking around.
Common Internet Wisdom: Replace plastic gears with metal. Matthias: Make gears out of wood.
Channel isn't named "woodgears" for nothing.
I was hoping he would make wood gears for this
A benefit of plastic gears is if something jams the gear fails but the rest of the machine is undamaged. This is the idea behind shear pins, etc. You have known failure points on purpose.
@@JDeWittDIY- It must be noted that I did say "Common Internet Wisdom". ;-)
He didn't put a wooden gear into a rc car... Yet.
I'm hyped for the era of Metal Matthias, considering all he can do with wood alone, with metal he'll be unstoppable
Naww. I look at stuff machinists make, and I often think "I could have made that way faster out of wood!"
@@matthiaswandel Don't we overestimate what products need to be made from metal? Your videos have shown that often times something made from wood is plenty strong for its intended purpose.
Not! You wood workers are 1 Inch 😊spec. Good luck. Don't buy vevor, unless it's free👎😢
The new Pen Shaking Device will double as a melee weapon of heretofore unanticipated lethality.
3D printed metal!
This video had maths, spreadsheets, Python programming, gear design, wooden gear making and a lathe cutting threads. What more could you want. Brilliant video! I found the part with the wooden gears fitted in the lathe particularly amusing.
Face it, Mathias is a national treasure.
Those wooden gears are giving me 'it's stupid but it works' vibes...
And if it's stupid, but it works (safely!), then it's not stupid ❤
I’m glad Vervor came to their senses and hired you to calculate the new values.
"Just takes a little bit of programing" is the worst thing a woodworker can hear
... but makers love.
Experienced programmers also wince at that a bit😬
To be fair the programming of this is the easy part.
"The gear chart was bad so I made a new one" would have made for a good video title too.
Impressive work!
If by "good" you mean annoying and sensational, then sure. The title and thumbnail game is the absolute worst part of modern TH-cam, bring back descriptive titles and simple thumbnails. Down with the algorithm.
I've made gears (when i was a teen and didn't have a proper lathe) with baltic ply (i think it was 15 layers) and here's the kicker. If you lightly score the faces of the teeth with a thin saw, then dip the gear in thin (model making) cyanoacrylate, they'll last even longer than plastic gears, which are widely used. You can achieve good precision across the teeth if you make a jig sander using a dulled file and ball bearing rollers set in a V. It will cut only as deep as the face of the tooth. Had perfect threads cut with that lathe. Proper thread against my thread, perfect mesh. I never "hardened" the key hole, because i was afraid it was too strong, stronger than a plastic gear (for sure) so i just left that natural ply on purpose. I had to re-key the gear maybe... 5 times total in 4 years. Wood gears? 100% plausible and highly recommended. Cheaper and your determination to get them proper is the factor at play. I used Tamiya superglue, but any thin superglue (read cheap, because it's thinned out) will work.
Vevor hit the jackpot with you. Whatever they paid you to correct their mistakes is a bargain for them. Apparently they don’t have access to people of your caliber within reach
Has all metal lathe, slowly starts replacing parts with wood. Matthias you absolute madman, I love it.
You should adjust your program so it excludes combinations where certain size gears are in positions that prevent the gearbox from closing.
That's only for the Nervous Nellys. Run with it open.
@@sdspivey But if he's making the gear ratio table for Vevor, as a manufacturer, they have to consider whether the protective cover is on or off.
@@jarodmorris611 they should be nervous, next up a laser lathe with infinite speed control and no torque needed. made by Wandel anti-gravity Works for 59.95. why not
@@jarodmorris611they could also consider an oversized gear cover to allow for larger gear sets.
He allready made wooden planer and bandsaw. Just wait few years and hell make video about wooden metal lathe and he's selling the plans on his website.
Matthias, I love that you are literally helping companies improve their products.
Matthias, I did my time as an Apprentice Fitter Machinist, Tool Maker and Welder. The highest of compliments to you for your continued outstanding work. You are without peer as a Polymath in all you tackle. Bloody well done Mate 😁. Cheers from Michael. Australia.
The amount of work you put into this is incredible. As a machinist, the only way I could threads that match in this situation is to make everything custom. If you produce a nut and a bolt on the same machine, tools, and gear ratios, they would both not match the chart, but at least they would match each other.
Until the nut or bolt is lost with your 10mm socket. Then you’re stuffed because you can’t go buy another.
Apartment dweller here. No wood working tools, no metal working tools, no space but still enjoyed. You seem to be gifted with many talents. Thanks.
That lathe could fit in an apartment easily. Of course, you need some other tools too. But if you have a spare bedroom, that could be a small machine shop!
@@matthiaswandel Original NYCCNC style
@@matthiaswandel Works only if he has no wife or girlfriend or doesn't mind loosing either/both.
@@matthiaswandelNo spare rooms in my apartment unfortunately. Can your lathe be mounted vertically to decrease the required floor space? Maybe bolted to the wall in the shower for easy cleanup.
that's where I ran mine when I first got it, I just laid down drop cloths and hung some around me to catch flying swarf and keep it out of the carpet.
So this is like the old phrase “is the glass half empty or half full” The engineers state it was made to the wrong specifications… Matthias says “it’s wrong, but for a few I’ll make you a new glass that is correct”
Love it !!
So great to see you using the old gear program and cutting wooden gears again! Back to the roots :)
One thing I love about this channel is that Matthias looks at a problem that we often just throw money at and solves it faster and cheaper with data and wood.
Matthias, your intelligent, diligence and continued adaptability has made you one of my favourite TH-cam's, been warching you for at least a decade now, great work as always brother
I could actually see the gears turning in your head on this one. I went through this exercise about 15 years ago when I bought a mini-lathe from Harbor Freight. Thanks for the memories.. 😎
What ever Vevor is paying you, it’s not enough!
Talk about above and beyond.
I love his thought process whilst tearing down and testing products. One of a kind.
Neat! Thanks Matthias, you make it look easy and I'm sure it's not. The real value is how you show us what's possible, and from that we can derive ourown inspiration. Great stuff.
You forgot to put the key in the chuck :-D
😂😂😂
The wood gears in the thread section of a metal lathe is probably more than enough to fertilize the rage farm - LOL!
This is peak Matthias content
Love the analysis and improvements.
Now, this is the quality of content Matthias has got us used to 👍👍👍👍👍
I’d pay good money or perhaps a “small fee” to just spend a few hours in Matthias’ mind. Your thought process and execution of said process are mind blowing.
Matthias had to buy a metal lathe to finally bring us WOODGEARS-content back again ;)
Amazing!
I - and a ton of other machinists - would definitely buy a Matthias Wandel Gear-ratio calculator program! "Just" add a UI like you've done with the gear generator or bigprint?
The need for that is definitely there! I am in a German hobby-machinists Forum and this question after a program that can do that came up quite often in the past years. It would certainly be very helpful!
One related and interesting problem would be to either propose a set of gears that would have a better precision than what they provide, or propose a few additional gears that would dramatically increase the precision of the initial set.
It can easily be brute forced, given the low numbers of gears. And that way, Vevor can also provide and small hardware upgrade in addition for the revised table.
Yeah, some logic in his program to determine which gears could be used to cover the most combinations would be helpful.
You're a frickin' genius! Thanks for another awesome video.
feeling envious of your gears stash. that's some decades worth of collecting 😍
You never cease to amaze me
I love how you think. Your videos are so satisfying.
Looks like they hired the right man for the job.
ha ha ha, absolutely love how hard Matthias goes geeking out on gears especially when there is error in the product sold.
I wish you were my friend and we could hang out.
Your persistence and resilience, combined with your "I'm not done yet, it could be better" methodological approach is inspiring.
Thanks for sharing so much of your time and information.
Matthias, as usual…Amazing!!
You are doing such important work here Matthias. I have been wanting to get a lathe for cutting small screws for my repair shop, and most vehicles are metric pitch nowadays anyhow, but I could never get a solid answer to if a cheap lathe like Vevors would actually cut ALL the metirc threads I need. I sincerely hope Vevor follows through and updates their thread pitch table, and I will likely buy the lathe along with the extra gear set.
Thread tables on lathes being wrong seems to be fairly common. Adam Savage had the same problem with his lathe.
This another reason I always learn something from Matthias video's thanks.
glad to hear vevor is interested in improving their product, at least in cases where it doesn't increase long run per-unit cost.
I had no idea about the gears and ratios for thread pitches. Now I do. Thanks for explaining this Matthias!
Seems like Vevor should update their default set of gears - and have the extra set be complementary so there aren't so many duplicates.
Honestly this guy is a treasure!! Love it!
For those who like this, you should watch the Screw advance box joint jig videos, there is also some calculation on how to choose the gears from a set to move the jig from a certain distance.
Always great to watch your videos. Thanks
This is awesome! The perfect way to approach this problem, and a very interesting problem to approach!
Nice study, I had no idea there was so much variation with the supplied gears,
Thanks much
Dude. Your mind works in such a unique way! Holy cow.
Keep doing what you're doing. Fantastic video.
I get a kick out of the users who get these Vevor lathes fine tuned. Very inspiring. Well done once again, Matthias!
You are the man showing everybody, that you do not have to have some fancy tools to make stuff work. Matthias Wangel, you are my spirit animal
It's been a hot minute since we've seen the good ol' gear generator program. Always nice to see it
04:20 How do you know that your test coupler is threaded error-free? If your little lathe has errors in its table, how many other lathes might have "sloppy" fits? Asking for a friend... I too have used spreadsheets to do high accuracy computations from the known integer gear values - great fun!
Back in the early 1970s I was attempting to retrofit a 1954 "Simplex XL" brand 35 mm movie projector whose nameplate stated explicitly that the driveshaft needed to turn at 1764 RPM! After much hair-tearing, I finally realized that the projector manufacturer had been EXTREMELY clever, because most ordinary 60Hz 4-pole single-phase AC induction motors loaded at ~50% of their full power rating will run at or slightly above ~ 1750 RPM, and by using a 49:50 gear ratio somewhere INSIDE the projector it would run very close to exact speed. 1764 = 9*7*7*2*2, and 1800 = 9*5*5*2*2*2. I wanted to drive the projector in perfect interlock with other film-handling audio equipment that was driven from 60 Hz line-frequency 4-pole synchronous motors @ 1800 RPM and/or wound-rotor 3-phase interlock (selsyn) motors @ 1200 RPM, so I found some tooth-belt pulleys & belts at a local vendor to set up the simple 1800:1200 = 60:40 and the more difficult 49:50 = 98:100. This required an intermediate jack-shaft, but that was the easy part...
The couplers are cut with a tap, so likely to be correct. Threaded rods and long bolts are rolled from one end to the other, so thy are often a little bit off.
I like how you simplify machining work
if i was looking into a mini lathe, id totally get this just because they are willing to work with you and improve what they go and are outsourcing this support. very smart of them in my book.
I've discovered that is true in other areas too: raspberry pi's, and lenovo(formerly IBM) thinkpads because of all the smart people hacking on them
Matthias pulls out a gears drawer, and this one is full of metal gears.
Hi I bought your gear cutter programme when it first came out and still use it today
Well dang! So far over my head. I’m just glad there are really smart people like you I can turn too, to do the heavy lifting.
Thanks!!!
This reminds me of the time I wrote a program to mod my brown and sharp 0 dividing head to approximate a 127 tooth gear by adding more gearing that I could cut to its input. If I recall correctly the issue is that 127 is a large prime number where the dividing head is a 40:1 worm reducer. It was really a neat problem to minimize that error in tooth spacing while rotating the blank for the cutter. I struggled to make the program efficient enough but got there.
You could also use a wheel with angle markings and align the indexing pin to the precise value in degrees. Any error you make here with manual adjustment is reduced by a factor of 40 (because of the ratio of the worm reducer), which is pretty neat.
@@first_namelast_name4923 yea so I never ended up doing it that way even through I did the math because I ended up installing a dro on the mill which makes something similar to what you suggest the easiest solution. If every tooth is spaced 127/360 deg. Multiply that x 40 and all you have to do is drill 127 holes 14.11 deg apart in a circle/series of concentric circles. To make a custom dividing plate. That just becomes a lot of holes to drill and coordinates to follow but not a hard task.
Nothing stops you man!
This video is incredibly nerdy…but I love it. Matthias this one was great!
the way this guy just effortlessly writes a giant script doing a ton of math and computing accuracy and doing all the charts and stuff. you can tell he did some heavy lifting at RIM
"Now NASA comes". (A common expression in Brazil to tell someone made something crazy ingenuous)
You are the hero we need!
You've started the journey of building a metal lathe out of wood. I look forward to watching this.
Buena mákina maestro, un saludo
so much fun and very interesting to watch -- Note to tool manufactures be wary what you print bc there is a guy named Matthias that could end up doing a youtube video and prove your instructions are wrong. From measuring the curvature of lenses to mini lathe thread cutter ratios.. THIS IS THE ONLY CHANNEL I ACTUALLY LOOK FORWARD TO NEW VIDEOS.
Brilliant, Matthias! Fantastic work! 😃
I guess it's time to discover how to make metal gears!
Anyway, stay safe there with your family! 🖖😊
Back to where it all started! Nice!
Love it. We have been working out gear trains for years. My first lathe a south bend 9" had no gearbox and was completely manually picked. Tougt me how to do the calcs. On any lathe you just need to pick the pitch that puts the leadscrew at 1:1 with the spindle and add your gears according. Look up the old south bend or Hercus charts. Prime numbers are always the killers! Also a plastic gear is good practice as it strips in a crash and saves your lathe. All my machines have a sacrificial gear in the train somewhere. Awesome work love it! Toolmaker by the way guys. ❤️🤘
I need this program in my life
videos like this almost make me wish i knew how to use python
great work, matthias
Love the way you explained. Most videos are apologetic for dividing in to numbers but that’s great if you follow along
“Takes a little bit of programming…”
Man, you’re way, way beyond me.
Most excellent video. I’m interested to know if the lathe company follows through and supplies accurate gears.
It's worth remembering that he previously worked for Research in Motion, ie. the company that brought us the Blackberry.
@@joelhollingsworth2374
I’m very aware of the gentleman’s accomplishments. He’s quite the fellow.
Vevor has contracted with the best!
Now here's an interesting optimisation problem: If you can make your own gears of any size, which ones would be best? I.e. what's the set of n gears that gives you the best overall accuracy at your target pitches? Then graph out how accuracy improves as n increases, and pick the smallest with acceptable error. The computation starts to get *big* as n grows though, might need some real cleverness to make it efficient
Owners of minilathes should watch this video - I did 🙂
Good effort, Matthias!
I laughed out loud just looking at the thumbnail! Can't wait to watch it!
this blew my brain right out, amazing.
This is the content i signed up for when I subscribed to the woodgears channel.
You didn't mention this point, so I will. Within a certain tolerance, screw thread parts that mesh without binding will accept some intolerance and work fine if they are used for clamping loads. Where the error wouldn't be acceptable would be when the error compounds over length, such as your jigs that use 16TPI threaded rod to advance a particular interval per turn, so you'd be advancing more (or less) than 1 inch per 16 turns. But if the threaded interface will be static once assembled, perhaps even locktited, then a small percentage error is irrelevant if they fully mesh without either binding or excessive slop.
I have to say I'm quite impressed that Vevor both managed to put that bad of a table on their lathes, and are willing to pay a youtuber to make them a better one 😂.
That said, if you ever grow tired of making wood gears, 3d-printed gears work great as change-gears for these mini-lathes, I have made a whole stack of them for mine. They are a bit more noisy, but they hold very well.
Of course not every one have a 3d printer but 3d printed gears work very well for this application.
With your love of stepper motors, I see an “electronic lead screw” in your future.
Haven’t wrote a program in years except starting to tinker with arduino now and my lathe comes in handy with mechanical bits 😊
Of course you would make wooden gears for a metal lathe. I love it. I can't wait for version 2.0 with helical plywood gears :)
Absolutely brilliant....
There's lots of interesting extensions to this problem. Calculating the best gear or gears you need to make to minimize the error for a precision part. Calculate which of your gears make the least difference to your precision so you could give them to a friend who doesn't have enough gears, or calculate the best subdivision between your shared gears so that you both have the best accuracy possible. Very satisfying intersection of math and practicality
Another very interessting video even if I was not sure to watch it :D
This video is good proof that I, a mechanical engineering student who hates anything related to programming, should learn some decent Python...
This is the reason I love Matthias
A a fan of spreadsheets, I loved this.
Now THIS is the Matthias I enjoy. That's some serious nerding out! :D
This dude is an absolute wizard. A special mind for sure. I’m a mechanical engineer and I feel like a gorilla compared to this dude lol
To be able to frame the problem of how to crank out a thousand or more combinations from X set of gears using a computer is remarkable! Applied Maths for the win.
Fantastic video Matthias. Now you could pitch (pin intended) another idea to Vevor. Working out what gear tooth numbers wheels they have to provide in a set that is optimized for the minimalist number of wheels to get a set of metric and imperial pitches most often used in industry. And what gears to include in a de-luxe set.
The very first thing you have to work out is what is the pitch of the screw for advancing the support. If you have a metric screw you can get all standard thread pitches bang-on with a relatively small number of wheels. With a 127 tooth wheel you can convert it to an imperial pitch.
I'm so dumb. How do I get 1% of this man's brain
You've got to be the smartest guy on the web.