Hi Andrew, I just looked on Andy's channel. Helical hobbing is the ultimate way to go, as you're doing. However for small gears like we make (& we have time on our side) there's a much easier way to produce a really good result. Make a hob like Andy shows, but not helical, just 5 teeth (or grooves, however you look at it) one next to the other, the distance of the pitch required. Use this in the mill as you would a normal milling cutter or gear cutter, using your rotory table to turn the part one tooth at a time. The central (of the 5) cutters in the "hobb" will cut the whole tooth while the adjacent cutters cut the pressure angle in the adjacent teeth. Carry on around until all the teeth are cut. I've made gears including for the Hoglet this way & they're spot on. Make the cutter teeth deep enough to cut any gears in the range & you only need the one cutter. Think of it as a gear rack cutting the round gear. Hope this helps somebody.
Hi Dave - thanks for the suggestion - very interesting. I will certainly take that into consideration. As you say, others should find it useful too. Cheers Andrew
Blinding us with science again! 😂 Looks an intersting concept. I have seen mechanical hoggers, but never an electronic version. Looking forward to your updates on this project.
Hello Andrew. Gear hobs are the way to go. I made one several years ago and have used it many times. Andys Machines is a good resource, but the hobs really are not that complicated. I initially just used .500 aluminum rod for a trial run to see if I could make one. It turned out OK so while I had all the setups established on the mill I made another from .500 drill rod. All the calculations are in Law's book on gear cutting. Good luck.
Hi Earl - thanks for the info - I hope you are fit and well and playing with your Hoglet and other fine engines. Yeah the Law's book is a godsend - though I probably only understand half a dozen of the pages. I guess I will have a go at meking a hobb at some stage. All the best. Andrew
Hello Andrew, You might be interested in the You Tube channel DYI Gear Hobs by Robertt 4522. He shows making a straight hob which is the way I made mine. It involves making a lot fewer angle calculations and set-ups. I made my 42dp hob using a homemade manual indexer and it turned out fine. Be well. Earl
I was also going to mention Robert's videos on the straight rack profile hobbing method. Andrew really should see that technique as well as the excellent spiral hobbing method.
Take a look at 4th axis spindle they offer for the 6040 CNC routers. Those are usually a bearing pack plus a Nema 17 all in one and a small lathe chuck.
Hi - that looks absolutely perfect for what I need. I got really excited and then I saw the cost! I will try to make something similar. Thanks for the info. Cheers Andrew
One downside to hobbing like this is the need to cant the head over to match the helix angle of the hob - this is fine if you want to leave the mill set up as a dedicated hobbing machine but if you want to use it to mill something else you have to tram it in again . For this reason I will probably stick to making hobs with no helix and just use the stepper to move the indexing head to index to the next tooth . Also with rotary hobbing you have to be careful of missed steps as the synchronisation can go out of whack , they usually have a feedback line ( using another encoder) from the dividing device so the arduino can compensate for any missed steps in real time . I look forward to seeing you build this Andrew !
Hi Ian - many thanks for the advice. I fully understand where you are coming from regarding tilting the mill head and I too had resevations. However my mill head has an adjustable stop at the side, so moving the head back to vertical is just a matter of rotating it up to the stop. So I think I should be okay 🤞. Your method certainly sounds interesting, so I now have a plan B if this method fails or proves to be too awkward setting up. Cheers Andrew
Great video! And big 👍 to Andy's Machines he has a very clever channel. Re the indexing/rotary head, check ebay for their chinese 4 inch rotary tables, there are cheap and quite compact and I think are 40 turn worm. You can just take off thenhandle and mount a stepper motor to where the handle used to go.
Thanks Wiz. Unfortunately a 40:1 gear ratio on a rotary table will not work. For the code to work, the spindle to hold the gear blank must have less or equal steps per revolution as the rotary encoder steps per revolution. My rotary encode has 600 steps/rev. If I set the stepper driver to 200 steps/rev, I can't have more than a 3:1 ratio on the table. I think I need a spindle running in bearings with a 3:1 gear ratio. Or maybe a stepper driver set to 400 steps/rev with a 1.5:1 gear ratio on the table. engineerd3d mentions a 4th axis device which would probably work well. Only issue is the price!🤑 Cheers Andrew
@@learningturningmetal ... thank you for explaining, I understand better now. It had been some time since I watched Andy's gear hobber videos. It is a shame his software does not allow for a larger physical gear ratio at the gear blank (since that is usually turning much slower than the cutter). That would allow for some rotary table options. Looks like you might need to get a 3:1 timing pulley set like he used... I would think a 3:1 set and 200 step motor is better than a 1.5:1 and 400 step motor because maximising the mechanical gearing will give more torque rigidity to counter the cutting forces. I'm a electonics/software guy so to my way of thinking the very best solution would be to fix the software to work with a 36:1 or 40:1 minature rotary table. 🤔
Hi Jeremy. I do hope it turns out okay. I'm just working on a very simple design for a 3:1 ratio spindle to hold the gear blank. Nothing as sophisticated as AndysMachines! Cheers Andrew
Took me a wile to understand what you were on about. We built a gear cutting Arduino using a single point cutter. Set tooth count, Micro step for adjustment, Next tooth advance. No need for spindle speed. Yours will work a lot faster using a hob. Still don't understand the problem with the 90 to 1 reduction. Keep up the good work.
Hi John - yeah a built a World of Ward uint a while back and it works a treat. AndysMachines goes to some length discussing gear ratios and the relationship between the milling spindle and the gear blank. I don't fully understand it, but this is my interpretation: For the code to work, the spindle holding the gear blank must have less or equal steps per revolution as the rotary encoder steps per revolution. My rotary encode has 600 steps/rev. If I set the stepper driver to 200 steps/rev, I can't have more than a 3:1 ratio on the table. I think I need a spindle running in bearings with a 3:1 gear ratio. Or maybe a stepper driver set to 400 steps/rev with a 1.5:1 gear ratio on the table. I hope it turns out okay! The good news is that Douglas in Pennsylvania has got the two gear cutters I'm missing, so I can continue with the JH v-twin. All the best. Andrew
Hi John - whether I can get it to work is a different matter. Good news on the two missing gear cutters - Douglas in Pennsylvania has now got them and is in the process of shipping them to me.😃 All the best. Andrew
Since you are going down this route if you remind me what DP you need, i will look at my stock of pro (spiral toothed) gear hobs and see what i have. I bought a stash at a watch and clock fair years ago and have many doubles.
Hi Andrew, I am building Andy's system but have encountered a problem which I believed to be linked to the encoder, your comments seem to confirm this. would there be any chance of a copy of the instructions to make it work on 5 volts? Thanks RonC
Those encoders seem sol outdated compared to the ones found in inkjet printers. They're just a graduated plastic film disc which is mounted to the shaft with a tiny decoder reading it.
I raise my hat for Your work, but... If i get it right You have managed to measure the RPM of the mill spindle. Using an encoder is a huge overkill. A magnet and a Hall sensor would do that. I have such a board, an Arduino UNO with the keypad and LCD shield attached. I use it for hole circle computations for the moment. Making the code myself. Other applications, just to download them. Time 10:33. There's no analog output on that Arduino. Likely it's a PWM, Pulse Width Modulated, output. It will be interesting to follow Your oncoming videos. For works like this there's the Arduino Forum having really expert helpers, and me....
Hi Stefan - electronics is a subject I don't really understand, but it's an interesting little venture. Nothing ventured, nothing gained! I've got a computer programming background so coding isn't really an issue. Cheers for now. Andrew
I doubt that using only one "datapoint" per revolution is reliable enough. Rather than dealing with time, the goal is to synchronise the angles of workpiece and hob, the time base can be totally kept out of the equation. e.g. during ramping up or down the speed of the spindle and workpiece should idealy stay in sync.
@@timogross8191 Looking at Andys video and gear hobbing, I fully agree. For just RPM measuring it would be enough but not for this way of cutting gears.
So I'm not the only one that counts larger numbers as 1, 2, 3, ..., 9, 10 1, 2, 3, ..., 9, 20 1, 2, 3, ..., 9, 30 etc. Had had some real wired looks from people here in Midlands when I count like that. Wonder if it's a Yorkshire/Northern thing (I'm a Yorkshire lad m'sen)
![ประสบการณ์ จ้าง Developer ที่ผมจะไม่มีวันลืม [ภาค2]](http://i.ytimg.com/vi/pZbsSDRrmXA/mqdefault.jpg)
Hi Andrew, I just looked on Andy's channel. Helical hobbing is the ultimate way to go, as you're doing. However for small gears like we make (& we have time on our side) there's a much easier way to produce a really good result. Make a hob like Andy shows, but not helical, just 5 teeth (or grooves, however you look at it) one next to the other, the distance of the pitch required. Use this in the mill as you would a normal milling cutter or gear cutter, using your rotory table to turn the part one tooth at a time. The central (of the 5) cutters in the "hobb" will cut the whole tooth while the adjacent cutters cut the pressure angle in the adjacent teeth. Carry on around until all the teeth are cut. I've made gears including for the Hoglet this way & they're spot on. Make the cutter teeth deep enough to cut any gears in the range & you only need the one cutter. Think of it as a gear rack cutting the round gear. Hope this helps somebody.
That's a great idea!
Hi Dave - thanks for the suggestion - very interesting. I will certainly take that into consideration. As you say, others should find it useful too.
Cheers
Andrew
Mind blown , I think I need to watch this again .
Ha ha - I don't understand electronics. Just an excuse to play with my oscilloscope.
All good fun!
Cheers
Andrew
Hello Andrew,
This series is going to be a good one... I find gear hobbing very interesting. See you in part two.
Take care.
Paul,,
Cheers Paul!
Blinding us with science again! 😂 Looks an intersting concept. I have seen mechanical hoggers, but never an electronic version. Looking forward to your updates on this project.
Hi Tim - I'm not so sure about that. I know virtually nothing about electronics. Yeah - I hope it works out okay.
Cheers
Andrew
Nicely explained. More please.
Thanks mate!
Hello Andrew. Gear hobs are the way to go. I made one several years ago and have used it many times. Andys Machines is a good resource, but the hobs really are not that complicated. I initially just used .500 aluminum rod for a trial run to see if I could make one. It turned out OK so while I had all the setups established on the mill I made another from .500 drill rod. All the calculations are in Law's book on gear cutting. Good luck.
Hi Earl - thanks for the info - I hope you are fit and well and playing with your Hoglet and other fine engines. Yeah the Law's book is a godsend - though I probably only understand half a dozen of the pages. I guess I will have a go at meking a hobb at some stage.
All the best.
Andrew
Hello Andrew, You might be interested in the You Tube channel DYI Gear Hobs by Robertt 4522. He shows making a straight hob which is the way I made mine. It involves making a lot fewer angle calculations and set-ups. I made my 42dp hob using a homemade manual indexer and it turned out fine. Be well. Earl
I was also going to mention Robert's videos on the straight rack profile hobbing method. Andrew really should see that technique as well as the excellent spiral hobbing method.
@@abbottart2000 Hi Earl. That's very interesting. Maybe my plan B!
All the best.
Andrew
Good luck with your new project. Its somewhat similar to the electronic lead screw for lathe that I have seen on TH-cam.
Gday Andrew, this is going to be an interesting project, im really looking forward to the next instalment, thanks for sharing mate, cheers
Thanks Matty - I'm making this up as I go along so it could be a total disaster!
Cheers
Andrew
Very interested in seeing this to its completion Andrew.👍
Thanks mate!
Take a look at 4th axis spindle they offer for the 6040 CNC routers. Those are usually a bearing pack plus a Nema 17 all in one and a small lathe chuck.
Hi - that looks absolutely perfect for what I need. I got really excited and then I saw the cost! I will try to make something similar. Thanks for the info.
Cheers
Andrew
Good stuff Andrew
Thanks Adam. My new oscilloscope is real fun!
Cheers
Andrew
Thanks for sharing 👍
Cheers Craig!
One downside to hobbing like this is the need to cant the head over to match the helix angle of the hob - this is fine if you want to leave the mill set up as a dedicated hobbing machine but if you want to use it to mill something else you have to tram it in again .
For this reason I will probably stick to making hobs with no helix and just use the stepper to move the indexing head to index to the next tooth . Also with rotary hobbing you have to be careful of missed steps as the synchronisation can go out of whack , they usually have a feedback line ( using another encoder) from the dividing device so the arduino can compensate for any missed steps in real time . I look forward to seeing you build this Andrew !
Hi Ian - many thanks for the advice. I fully understand where you are coming from regarding tilting the mill head and I too had resevations. However my mill head has an adjustable stop at the side, so moving the head back to vertical is just a matter of rotating it up to the stop. So I think I should be okay 🤞. Your method certainly sounds interesting, so I now have a plan B if this method fails or proves to be too awkward setting up.
Cheers
Andrew
"Open"-Loop works O.K.ish, for hobby type work. Feedback is not a must, but a reliable signal chain is important.
Great video! And big 👍 to Andy's Machines he has a very clever channel.
Re the indexing/rotary head, check ebay for their chinese 4 inch rotary tables, there are cheap and quite compact and I think are 40 turn worm. You can just take off thenhandle and mount a stepper motor to where the handle used to go.
Thanks Wiz. Unfortunately a 40:1 gear ratio on a rotary table will not work. For the code to work, the spindle to hold the gear blank must have less or equal steps per revolution as the rotary encoder steps per revolution. My rotary encode has 600 steps/rev. If I set the stepper driver to 200 steps/rev, I can't have more than a 3:1 ratio on the table. I think I need a spindle running in bearings with a 3:1 gear ratio. Or maybe a stepper driver set to 400 steps/rev with a 1.5:1 gear ratio on the table. engineerd3d mentions a 4th axis device which would probably work well. Only issue is the price!🤑
Cheers
Andrew
@@learningturningmetal ... thank you for explaining, I understand better now. It had been some time since I watched Andy's gear hobber videos.
It is a shame his software does not allow for a larger physical gear ratio at the gear blank (since that is usually turning much slower than the cutter). That would allow for some rotary table options.
Looks like you might need to get a 3:1 timing pulley set like he used... I would think a 3:1 set and 200 step motor is better than a 1.5:1 and 400 step motor because maximising the mechanical gearing will give more torque rigidity to counter the cutting forces.
I'm a electonics/software guy so to my way of thinking the very best solution would be to fix the software to work with a 36:1 or 40:1 minature rotary table. 🤔
Been thinking of building the same thing
Me also. Spiral hobbing is the way to go I reckon.
Hi Jeremy. I do hope it turns out okay. I'm just working on a very simple design for a 3:1 ratio spindle to hold the gear blank. Nothing as sophisticated as AndysMachines!
Cheers
Andrew
Took me a wile to understand what you were on about. We built a gear cutting Arduino using a single point cutter. Set tooth count, Micro step for adjustment, Next tooth advance. No need for spindle speed. Yours will work a lot faster using a hob. Still don't understand the problem with the 90 to 1 reduction. Keep up the good work.
Hi John - yeah a built a World of Ward uint a while back and it works a treat. AndysMachines goes to some length discussing gear ratios and the relationship between the milling spindle and the gear blank. I don't fully understand it, but this is my interpretation: For the code to work, the spindle holding the gear blank must have less or equal steps per revolution as the rotary encoder steps per revolution. My rotary encode has 600 steps/rev. If I set the stepper driver to 200 steps/rev, I can't have more than a 3:1 ratio on the table. I think I need a spindle running in bearings with a 3:1 gear ratio. Or maybe a stepper driver set to 400 steps/rev with a 1.5:1 gear ratio on the table.
I hope it turns out okay!
The good news is that Douglas in Pennsylvania has got the two gear cutters I'm missing, so I can continue with the JH v-twin.
All the best.
Andrew
Very intriqued with this idea. Thanks for the look, understand the wait for parts.
Hi John - whether I can get it to work is a different matter. Good news on the two missing gear cutters - Douglas in Pennsylvania has now got them and is in the process of shipping them to me.😃
All the best.
Andrew
Since you are going down this route if you remind me what DP you need, i will look at my stock of pro (spiral toothed) gear hobs and see what i have. I bought a stash at a watch and clock fair years ago and have many doubles.
Hi Chris - I guess it must have some similarities.
All the best.
Andrew
Hi Andrew,
I am building Andy's system but have encountered a problem which I believed to be linked to the encoder, your comments seem to confirm this.
would there be any chance of a copy of the instructions to make it work on 5 volts?
Thanks
RonC
Hi Ron
Email me at learningturningmetal@gmail.com and I'll send you the details.
Cheers
Andrew
Hi Ron
Did Andrew reply to you ?
please provide the Arduino UNO code
Hi. The assembler code can be downloaded via AndysMachines TH-cam channel.
Cheers
Andrew
Those encoders seem sol outdated compared to the ones found in inkjet printers. They're just a graduated plastic film disc which is mounted to the shaft with a tiny decoder reading it.
Hi Bob - but they do work!😃
Cheers
Andrew
I raise my hat for Your work, but...
If i get it right You have managed to measure the RPM of the mill spindle. Using an encoder is a huge overkill. A magnet and a Hall sensor would do that.
I have such a board, an Arduino UNO with the keypad and LCD shield attached. I use it for hole circle computations for the moment. Making the code myself. Other applications, just to download them.
Time 10:33. There's no analog output on that Arduino. Likely it's a PWM, Pulse Width Modulated, output.
It will be interesting to follow Your oncoming videos.
For works like this there's the Arduino Forum having really expert helpers, and me....
Hi Stefan - electronics is a subject I don't really understand, but it's an interesting little venture. Nothing ventured, nothing gained! I've got a computer programming background so coding isn't really an issue.
Cheers for now.
Andrew
Knowing programming is always nice and useful in projects like that. I'll to remember You know that.
I doubt that using only one "datapoint" per revolution is reliable enough.
Rather than dealing with time, the goal is to synchronise the angles of workpiece and hob, the time base can be totally kept out of the equation.
e.g. during ramping up or down the speed of the spindle and workpiece should idealy stay in sync.
@@timogross8191 check out AndysMachines to see how the software works 😉
@@timogross8191 Looking at Andys video and gear hobbing, I fully agree. For just RPM measuring it would be enough but not for this way of cutting gears.
So I'm not the only one that counts larger numbers as
1, 2, 3, ..., 9, 10
1, 2, 3, ..., 9, 20
1, 2, 3, ..., 9, 30
etc.
Had had some real wired looks from people here in Midlands when I count like that. Wonder if it's a Yorkshire/Northern thing (I'm a Yorkshire lad m'sen)
Ha ha. My brain to mouth and back to brain has never been good. Maybe it's the Yorkshire Water - or lack of 😵💫
👍👍👍👍👍
🤞🤞🤞🤞🤞😃