James -- I, for one, greatly appreciate you comprehensively showing your methodical step-by-step requirements, thinking, calculations and hands-on tasks. It really makes projects like this seem a bit more approachable, providing a reference way of accomplishing your specific result. It introduces the topics that one might immediately learn from the video if familiar with related ideas, or for less-familiar topics, at least introduces them as areas to research a bit further. Well done!
Glad you liked it. This video was an experiment with a different, much more compressed editorial style. It's very difficult to make project videos that people actually watch. :)
Instead of cable feed is it possible to still do a hydraulic servo using a hydraulic cylinder and a scale I don't know if such things are still available they should be always liked hydraulic servos solid dead accurate leaks and all.
Yet another green field machine automation project by James. 💯 3D scanning, fusion360, specification, design, machining & software - I’m sitting comfortably in anticipation with the popcorn at the ready!🍿
Did not expect a SuperFastMatt mention. 😄I would also recommend the Making for Motorsport channel, he's done several videos on hobbyist 3D scanning and the processes around it.
This will be fun! And a lot of different fields of knowledge covered in one video is great as I can relate to several of my own projects/ideas/interests.
Great video as always. I’m glad you’re doing another automation project. I really enjoyed your electronic lead screw project. To anyone who hasn’t seen that project or who isn’t familiar with James stick around throughout the project, you’re in for a real treat.
The gear reduction will definitely help with the inertia mismatch since the reflected inertia is equal to the load inertia divided by the reduction ratio squared, though something tells me you already took this into consideration. Sesame has a line a planetary reducers that adhere(ish) to the the NEMA standard for stepper motors. Will probably cost as much as the motor, but will allow you more options than a belt drive since you can do higher ratios. Everyone knows speed and torque, but inertia is a killer....
This is giving me some great ideas for my old CNC mill conversion. If i don't get the Heidenhain up and running i will switch to the clear core. Great work as always.
Good stuff James. This is you in your element. If you can come up with good grinding cycles for different styles of grinding opperations . Im sure you could sell the controller as a kit for the retrofit market Awesome
Hi James, SDP is a good choice. Quality parts that don't cost an arm and a leg and their calculators are awesome. Have used them for pulleys and belts when adding computer control to several of my telescopes and even built a large German Equatorial Mount for an 8" Refractor that was used at our astronomy clubs observatory. Looking forward to the videos to come on the project.
Love this video. Think it's on a level rarely seen on UA-cam. Thanks for making my day a bit brighter. I usually gets my hands dirty in the shop. Why is touch screens more and more popular these days?
The only use i have for a cnc surface grinder would be for grinding forms on carbide and ceramic. 99 percent of that would be on round parts between centers or a drive head. You are going to make that grinder a clumsy POS and never develop grinding skills. Thanks very much for the video. There is much useful information for projects i would like to do. I look forward to seeing the finished project. If that surface grinder had a DedTrue centerless fixture or a powered rotary fixture mounted then the cnc retro fit would be kinda awsome.
Wish I had your knowledge of servo motors and the associated controls. I have a sawmill that I would love to automate. Essentially the same except just 2-axis setup, and bigger. Spent a lot of years cranking manual surface grinders. Your project is an interesting exercise, even if not particularly practical for most applications on a small machine.
for anyone interesting in this inertia thing: inertia matching optimises energy transfer i.e. efficiency. the electronics equivalent is impedance matching; you don't always want input and output impedances to be matched (think of output of a pre-amp into the input of a power amp). in practice, the total system inertia is usually as low as possible (i.e. smallest motor) without overwhelming the control loop of the servo (i.e. which is when it starts to oscillate) because that is most effective in terms of performance per dollar. between system inertia and rotor inertia you would want the rotor to far exceed the system inertia because from the motor's perspective that would approach driving no load, which isn't necessarily efficient, but it does provide very good performance; just think of how fast a servo motor starts and stops with no load attached. i had a 2.2 kW AC servo some time ago and it would literally jump off the table a little bit when changing speeds. went through a big rabbit hole when selecting motors for my machine lol
Hi James, looking forward to the rest of build, 3 axis of control… I can see a profile dresser being added at some stage. Your X axis position may creep because the transmission for the X is rope and pulley.
Having the knowledge to do mechanical, electronics and software is such an awesome advantage and opens the door to so many projects. Im good with mechanical stuff but missing the electronics and software knowledge
We have had many issues with HMIs from 4D systems. They often glitch, similar to what you were seeing on the ELS for a while, and they would periodically freeze without warning so that button presses would not be recognized for several seconds at a time. We finally settled on a raspberry pi screen which has been working super well. I'd suggest taking a look at those if you don't want to go down the hmi rabbit hole.
Also, I'm not sure what you're planning to use for a power supply but, as I'm sure you know, Clearpath motors operate with higher torque ratings of they have a 75V supply. Teknic offers two power supplies, the IPC 3 and 5. You should be able to operate all three motors with a single IPC 5, especially since they won't all be running at full gronk at the same time. They're not the cheapest but neither are Clearpaths. They just work and have the backing of Teknic which is worth the cost imo. I can vouch for the ClearCores as well. They work incredibly well but you need to pay more attention to the input and output wiring. The documentation helps a lot with this but pull up and pull down resistors are pretty commonly needed for things like limit switches/sensors - not quite as simple as an Arduino.
Amazing work but i have one concern. While you might think that your pulleys can handle the required torque, the technique used to mount them to the motor shaft does not. You should replace them with clamping-type pulleys, which are designed to handle high-torque applications like yours. When you keep using regular pulleys the set screws will damage the motor shaft, and the set screws will eventually unscrew themselves. You could possibly get away with keyway pulleys but clamping-type pulleys are what you should be looking for.
Hey James. Thanks for your very informative video! We went ahead and did the exact same conversion on a Sharp SG-618 (Taiwanese clone). As this was for a pressing job in my company, I had an engineer working full time on the project. We decided to get the enhanced and sealed version of the servos. Our machine is setup with a "Grind-Rite" centerless grinder attachement. The build is now complete we are CNC grinding shafts within 2-3um with an auto dressing/truing sub-routine (diamond wheel). Let me know me know if I can help.
Looking forward to this series as I've been thinking about doing the same to a KO Lee S714 manual grinder, but lack the expertise. I'm hoping that much of your design will translate to other grinders.
I love that we all buy tools to make better tools and use the better tools to upgrade existing tools that we own, thus the cycle of life. But lets not forget that we tell the wives it only cost £40 on clearance on some sale wrongly listed. looking forward to the updates.
i have the same machine and as projects allow i was going to do the same thing. will follow this one and hopefully you make it available as i will certainly stand in the queue.
My Brown and Sharpe no2 surface grinder has auto feeds. It also grinds to within a micron. Not bad for something built in the 1930s. Progress eh? Oh and it cost me £350 or under $500 😂
Automating the X and Z axis makes sense for the surface grinder but in my mind, automating the Y (Vertical) is not so much unless you set traversal limits to keep from crashing the wheel into the work in excess of what you intended.
@Clough42 James I can't wait to see how this pans out. Teknic's stated repeatability vs resolution leave gaps that could be orders of magnitude apart if both numbers are accurate. Happy Thanksgiving from the youtube community! Looking forward to your next installment.
Interesting project, exciting to see the result. The Clearpath controller price is far too reasonable considering the prices of their motors, I was expecting it to be a lot more!
This is extremely timely as I just started a 3 axis project with a ClearCore / clearpath motors. Definitely looking forward to seeing how you get along with the IDE. Also interested to see how you implement limit switches/ homing in the code. And your E-stop / enable circuit. Apparently you can attach a digital input to a motor control object to act as a limit switch but they don't really explain anything about it in the documentation that I've been able to find.
Yeah, there's a lot these motors can do that I don't fully understand. Like auto-homing against a hard stop. Not sure how that interacts with the step-direction interface.
@@Clough42 I saw they had a code example for that, using the HLFB signal. Although for my purposes I think I need an actual limit switch, and they don't have any examples for that using the SD motor types.
Please accept my professional opinion as a kind, respectful comment. As a tool and die maker, I've spent countless hours using a surface grinder, and it's not that hard. And as I'm sure you know, electronics can cause crashes too! Unless you're doing some type of production work, you will spend more time fiddling around with the electronics and programming than if you just used the grinder by hand. Automation is good for larger grinders which I have used and automation has it's place, but I don't think it's practical or advantageous for your application in a home shop. You can get yourself out of trouble much quicker on a manual machine. That being said, from the standpoint of providing entertaining and informative content that encourages others to explore new ideas, you will and always have accomplished that goal. Thanks for your videos.
When I would 3D scan big dies with a Creaform I would use MAGNAFLUX Dye Penetrant Developer. It worked good but the downside was also having to clean it up. One this I did think of was trying spray foot powder but I never had the time to try it.
Yes, dye pen developer, brand doesn't matter, is cheaper and is what everyone used until they developed the fancy sublimating sprays. It also has the benefit/disadvantage of sticking around until you have to clean it up. It does just wash/wipe off as it's basically just spray on chalk so it's not to bad. I've tried foot sprays and a bunch of other hobby solutions, it's not even close to as good or easy to work with. If you screwing around in your garage sure but if you're on the clock not worth it at all.
@@TheWidgetWorks Oh I am sure there are other brands. I was bringing this program online at my shop back in 2018 So at the time I did have some time to figure out a good process. Man I miss using a Creaform.
Hi from Canada. I have questions: When doing you torque calculations why did you not include the effect of the work piece(s), additional chucks, fixtures, etc.? At TS6.45 you said you only calculated for the table and magnetic chuck as load. Wouldn't the addition of your secondary chuck plus several pounds workpiece almost double the required torque? Lastly: could the manufacturer of your grinder have provided a 3D model? P.S. I really, really enjoy your channel.
James, when I began watching ytube videos. I have watched ones that helped me to pick up with a hobby. Shut as when you made a Chuck Key for Quinn. I'll continue, and what you're doing. But I spent nearly 44 years with controllers, PLC. So I'm a bit burned out. So if you wish, pls send me a reply
What I found the hard way is, the motor drivers don't like motors hand cranked when electronic control is OFF. I used cheap TMC2209 drivers with Nema17 motors not the expensive kind mentioned here. Just be sure to check if you plan to hand crank machine when not using electronics for moving.
The Clearpath servos can theoretically handle this if they're powered, but not enabled. They can even do this with the power hub and logic backup 24V. I haven't tried it.
Crazy idea here, if you're rigging up your own software for this, why not put a pair of inductive proximity sensors on either side of the grinding wheels contact point. If you have one on each side the logic to automate grinding a pass should be straight forward, you can use the sensors to control the back and forth of the y axis without having to ever setup the machine. 1 Go to the top of the Y Axis 2 Go to the centre of the X and Z Axis 3 Lower the Y till you get a signal on both sensors 4 Advance Z "towards you" until you loose Signal on both sensors 5 go in the distance of the sensor offset from the actual wheel 6 Go all the way to the right on the X Axis until you loose both Sensors 7 Drop the machine down a bit on the Y ( I'd guess the amount of thou you want to take off) 8 Go Left across X till you loose the sensor signals 9 Advance Z by some Variable Amount (I'd guess set by rotary encoder knob or touch screen) 10 Go the other way on X till you loose the signal repeat 9 and 10 till you loose signal on both sensors, voila full pass no setup. just make sure the piece is over the center and not off to one side or you would be grinding the magchuck. doing it this way it doesn't matter the size or the piece the machine can just hit go and it'll run until it's ground the top. the only issue I can see with this is the bracket to hold the sensors would need to be adjustable to take into account the wheel getting smaller as you dress it. I guess technically you could set up some sort of tool height sensor to set the height of the sensors but you'd need motor to control the height of the sensor bracket.
This. Working out your requirements and translating them to specifications before implementation is always sexier than going broke doing the wrong thing and living with the result. On something this complicated, it's way too easy to get it wrong. One of my favorite aspects of your ELS project was the thoroughness and care you took in your performance requirements and how they translated to your component selections. It really minimized the redesign/rework parts of the project. (Plus, you got a lot of great channel content and products to sell!) I'm sure you did some back-of-the-envelope guesstimates of the time & cost a similar approach would need for this grinder project, before deciding on a more OTS solution. When you're done, I hope we can get a comparison of those estimates vs. the actuals.
Hi James, I'm curious how you do the screen recordings in Fusion. You always move the views around and zoom in/out so smoothly - what's your technique?
Hi James are you aware of the Bridgeport /Harig EZ surf grinder ? Bridgeport installed their CNC control on a Harig surface grinder, very similar in size and form to your machine .Might be a useful reference for your project.
One question I'm interested in is the component of system behavior that corresponds to the fact that the servomotor's position/velocity/acceleration feedback is on the motor, rather than on the load. Is that somehow solved by the "Inertia Mismatch" calculations? It seems related, but I don't quite see how it tackles the issue directly.
That turns out to be an interesting question. The servo controller controls the motor, since that's where the encoder is located. The controller sees the motor's inertia directly, but the load's inertia is only visible through a coupling that has both elasticity and viscous damping. The motor mounting and the load also have some degree of damping. These external factors are the reason the servo needs to be tuned to the load. "Inertia Mismatch" is a way to calculate how much you might be asking the controller to compensate for, but it's only part of the picture. The real excitement occurs in the frequency domain due to phase and resonance. There's a fairly detailed treatment here, if you're interested: www.automate.org/industry-insights/understanding-the-mysteries-of-inertia-mismatch
Hello James, since you are the one I go to for CAD tutorials, I thought I'd ask you this question. I'm trying very hard to become a better designer and the terminology escapes me in this case. In Constructive Solid Geometry, you create primitives that either add to the final geometry, or are used to cut away parts of other primitives. Is there a specific term for the "tool" that I create to be used to cut away from the original geometry? I seem to think that the term "slug" is correct, but I really have no idea. Overkill really IS underrated, Sir!
Are there no scale bars or coded reference markers available for your 3D scanner? Worked professionally with Steinbichler and GOM systems (in the $100´000 price range) for a couple of years and even tho the marketing ppl likes to tell you how good their free matching functions in the software are, if you want good tolerances scale bars and coded markers are superb. If you want even better, photogrammetry is the way to go. GOM software used to be free in a somewhat limited, but still capable, version. But i guess its not free anymore since Zeiss came in and took over the show.
Hi,...There are not many detailed videos for Industrial Ac Servos especially for Mach3.! Those can't be used for Mach3 or what? The connections and setting parameters may vary from model to model I guess..! I think it would be awesome if We can use such industrial servos with Mach3. But the inertia tuning and all again ...oh ...!!
Since you invoked the name of Robin Renzetti, I will bring up Adam the Machinist, who takes CNC grinding to the next level. Pretty sure you have seen his videos, but mentioned him just in case.
James -- I, for one, greatly appreciate you comprehensively showing your methodical step-by-step requirements, thinking, calculations and hands-on tasks. It really makes projects like this seem a bit more approachable, providing a reference way of accomplishing your specific result. It introduces the topics that one might immediately learn from the video if familiar with related ideas, or for less-familiar topics, at least introduces them as areas to research a bit further. Well done!
Glad you liked it. This video was an experiment with a different, much more compressed editorial style. It's very difficult to make project videos that people actually watch. :)
@@Clough42 There are three or four "chapters" of this video that I'd watch another 30 - 40 minutes of. But I'm probably in the minority.
Instead of cable feed is it possible to still do a hydraulic servo using a hydraulic cylinder and a scale I don't know if such things are still available they should be always liked hydraulic servos solid dead accurate leaks and all.
Yet another green field machine automation project by James. 💯
3D scanning, fusion360, specification, design, machining & software - I’m sitting comfortably in anticipation with the popcorn at the ready!🍿
This is what I've been waiting for
the technology has gone so far ,thats just amazing , the 3D scanner for a machine this scale look just insane to me
Great video as usual!
Can't wait to see the rest of them.
I, too, almost understood what you are explaining. The lack of understanding is completely on my end. But, VERY interesting and I’ll keep watching.
Did not expect a SuperFastMatt mention. 😄I would also recommend the Making for Motorsport channel, he's done several videos on hobbyist 3D scanning and the processes around it.
James you are a God, I have no idea what you are doing but I can't stop watching. 😅🤣
This will be fun!
And a lot of different fields of knowledge covered in one video is great as I can relate to several of my own projects/ideas/interests.
Great video as always. I’m glad you’re doing another automation project. I really enjoyed your electronic lead screw project. To anyone who hasn’t seen that project or who isn’t familiar with James stick around throughout the project, you’re in for a real treat.
Thanks for that one - that was incredibly informative! I have used Clearpath SDSKs on my Milling Machine and they are absolutely fantastic…
The gear reduction will definitely help with the inertia mismatch since the reflected inertia is equal to the load inertia divided by the reduction ratio squared, though something tells me you already took this into consideration. Sesame has a line a planetary reducers that adhere(ish) to the the NEMA standard for stepper motors. Will probably cost as much as the motor, but will allow you more options than a belt drive since you can do higher ratios. Everyone knows speed and torque, but inertia is a killer....
You are correct. I did take the squared effect of the drive ratio into account. :)
I had no doubt that you would automate the surface grinder. And I'm glad I was right :)
This is giving me some great ideas for my old CNC mill conversion. If i don't get the Heidenhain up and running i will switch to the clear core. Great work as always.
Now that was surprisingly easy to follow and even almost understand!
Hot damn! I can’t wait for this series!!!!!! 🤩🤩🤩
CNC is fun when you are making the item. Thanks for the video keep on keeping on.
Your axis naming is correct...Z axis moves in and out through the center of rotation of the spindle!
Ive built/retrofitted 3 cnc setups. Ive always wanted to do this to my surface grinder. Subbed!
Very excited! Been waiting for this one
Good stuff James. This is you in your element. If you can come up with good grinding cycles for different styles of grinding opperations . Im sure you could sell the controller as a kit for the retrofit market
Awesome
Very exciting new project. I am looking forward to every detail.
Excellent stuff; looking forward to the rest of this series!👍😁🍻
Hi James,
SDP is a good choice. Quality parts that don't cost an arm and a leg and their calculators are awesome. Have used them for pulleys and belts when adding computer control to several of my telescopes and even built a large German Equatorial Mount for an 8" Refractor that was used at our astronomy clubs observatory. Looking forward to the videos to come on the project.
Roll on part 2, I'm hooked.
Love this video. Think it's on a level rarely seen on UA-cam. Thanks for making my day a bit brighter.
I usually gets my hands dirty in the shop. Why is touch screens more and more popular these days?
I avoided them with the ELS, but this is going to be more complicated and hard to do without spinning a big custom button panel.
2:45 - I have no idea what you’re saying, but I like it.
Looks like another great project!
Looking forward to seeing this conversion! Thanks for the great videos.
My "traverse shoulder" has me thinking about at least motorizing the left wheel on my Harig- I appreciate the video!
Definitely looking forward to this project! Keep it up James 😊
The only use i have for a cnc surface grinder would be for grinding forms on carbide and ceramic. 99 percent of that would be on round parts between centers or a drive head. You are going to make that grinder a clumsy POS and never develop grinding skills. Thanks very much for the video. There is much useful information for projects i would like to do. I look forward to seeing the finished project. If that surface grinder had a DedTrue centerless fixture or a powered rotary fixture mounted then the cnc retro fit would be kinda awsome.
Wish I had your knowledge of servo motors and the associated controls. I have a sawmill that I would love to automate. Essentially the same except just 2-axis setup, and bigger.
Spent a lot of years cranking manual surface grinders. Your project is an interesting exercise, even if not particularly practical for most applications on a small machine.
for anyone interesting in this inertia thing: inertia matching optimises energy transfer i.e. efficiency. the electronics equivalent is impedance matching; you don't always want input and output impedances to be matched (think of output of a pre-amp into the input of a power amp). in practice, the total system inertia is usually as low as possible (i.e. smallest motor) without overwhelming the control loop of the servo (i.e. which is when it starts to oscillate) because that is most effective in terms of performance per dollar. between system inertia and rotor inertia you would want the rotor to far exceed the system inertia because from the motor's perspective that would approach driving no load, which isn't necessarily efficient, but it does provide very good performance; just think of how fast a servo motor starts and stops with no load attached. i had a 2.2 kW AC servo some time ago and it would literally jump off the table a little bit when changing speeds. went through a big rabbit hole when selecting motors for my machine lol
You should make an incinerator for burning your trash. Great video! Thanks for sharing.
I look forward to future videos, thanks for sharing, as always!
I want to start a video with "this is my surface grinder" one day :)
I’m gonna have to look into 3d scanners. Would be so convenient
Hi James, looking forward to the rest of build, 3 axis of control… I can see a profile dresser being added at some stage. Your X axis position may creep because the transmission for the X is rope and pulley.
Awesome project!
Having the knowledge to do mechanical, electronics and software is such an awesome advantage and opens the door to so many projects. Im good with mechanical stuff but missing the electronics and software knowledge
We have had many issues with HMIs from 4D systems. They often glitch, similar to what you were seeing on the ELS for a while, and they would periodically freeze without warning so that button presses would not be recognized for several seconds at a time. We finally settled on a raspberry pi screen which has been working super well. I'd suggest taking a look at those if you don't want to go down the hmi rabbit hole.
Also, I'm not sure what you're planning to use for a power supply but, as I'm sure you know, Clearpath motors operate with higher torque ratings of they have a 75V supply. Teknic offers two power supplies, the IPC 3 and 5. You should be able to operate all three motors with a single IPC 5, especially since they won't all be running at full gronk at the same time. They're not the cheapest but neither are Clearpaths. They just work and have the backing of Teknic which is worth the cost imo.
I can vouch for the ClearCores as well. They work incredibly well but you need to pay more attention to the input and output wiring. The documentation helps a lot with this but pull up and pull down resistors are pretty commonly needed for things like limit switches/sensors - not quite as simple as an Arduino.
I have an IPC-3, which technically should be enough, but I ordered an IPC-5, just because the enclosed chassis is easier to package.
Ye all see what you can do when you go to school! Awesomeness.
Look forward to your next video in this series.
Excited for this project
This will be a great project, Thanks for sharing.
As usual James, freaking AWESOME! ❤😅 .....can't wait to see more!
You sure are a clever Dude... Like your videos!
Amazing work but i have one concern. While you might think that your pulleys can handle the required torque, the technique used to mount them to the motor shaft does not. You should replace them with clamping-type pulleys, which are designed to handle high-torque applications like yours.
When you keep using regular pulleys the set screws will damage the motor shaft, and the set screws will eventually unscrew themselves. You could possibly get away with keyway pulleys but clamping-type pulleys are what you should be looking for.
Hey James. Thanks for your very informative video! We went ahead and did the exact same conversion on a Sharp SG-618 (Taiwanese clone). As this was for a pressing job in my company, I had an engineer working full time on the project. We decided to get the enhanced and sealed version of the servos. Our machine is setup with a "Grind-Rite" centerless grinder attachement. The build is now complete we are CNC grinding shafts within 2-3um with an auto dressing/truing sub-routine (diamond wheel). Let me know me know if I can help.
that scanner does a great job!
Looking forward to this series as I've been thinking about doing the same to a KO Lee S714 manual grinder, but lack the expertise. I'm hoping that much of your design will translate to other grinders.
first time i see someone using a TI microcontroller, most of their example schematics in datasheets use them, it's tempting to try for fun.
They're great, but mostly targeted for spinning your own boards.
@Clough42 James, thanks for answering my question I asked you at the Bash this summer with this video!
I love that we all buy tools to make better tools and use the better tools to upgrade existing tools that we own, thus the cycle of life.
But lets not forget that we tell the wives it only cost £40 on clearance on some sale wrongly listed.
looking forward to the updates.
Thank you. I really enjoy this. I do the same sort of things on a smaller budget. I wish I had a scanner.
i have the same machine and as projects allow i was going to do the same thing.
will follow this one and hopefully you make it available as i will certainly stand in the queue.
I love this stuff!!!
My Brown and Sharpe no2 surface grinder has auto feeds. It also grinds to within a micron. Not bad for something built in the 1930s. Progress eh? Oh and it cost me £350 or under $500 😂
Yup old school has definite advantages
Automating the X and Z axis makes sense for the surface grinder but in my mind, automating the Y (Vertical) is not so much unless you set traversal limits to keep from crashing the wheel into the work in excess of what you intended.
@Clough42 James I can't wait to see how this pans out. Teknic's stated repeatability vs resolution leave gaps that could be orders of magnitude apart if both numbers are accurate. Happy Thanksgiving from the youtube community! Looking forward to your next installment.
Interesting project, exciting to see the result. The Clearpath controller price is far too reasonable considering the prices of their motors, I was expecting it to be a lot more!
Great project!
Cool project, it’s going to be fun to follow👍
3:19 dam that smile caught me off guard.
Great video thank you.
This is extremely timely as I just started a 3 axis project with a ClearCore / clearpath motors. Definitely looking forward to seeing how you get along with the IDE.
Also interested to see how you implement limit switches/ homing in the code. And your E-stop / enable circuit.
Apparently you can attach a digital input to a motor control object to act as a limit switch but they don't really explain anything about it in the documentation that I've been able to find.
Yeah, there's a lot these motors can do that I don't fully understand. Like auto-homing against a hard stop. Not sure how that interacts with the step-direction interface.
@@Clough42 I saw they had a code example for that, using the HLFB signal. Although for my purposes I think I need an actual limit switch, and they don't have any examples for that using the SD motor types.
@@MarkFunderburk I did a bunch of reading last night, and I think that's correct.
1051 thanks for showing the full process I really appreciate it :-)
WOW!!!…just….WOW!….enjoyed…KOKO
Axes designation is correct. It's the same as a horizontal mill.
Please accept my professional opinion as a kind, respectful comment. As a tool and die maker, I've spent countless hours using a surface grinder, and it's not that hard. And as I'm sure you know, electronics can cause crashes too! Unless you're doing some type of production work, you will spend more time fiddling around with the electronics and programming than if you just used the grinder by hand. Automation is good for larger grinders which I have used and automation has it's place, but I don't think it's practical or advantageous for your application in a home shop. You can get yourself out of trouble much quicker on a manual machine. That being said, from the standpoint of providing entertaining and informative content that encourages others to explore new ideas, you will and always have accomplished that goal. Thanks for your videos.
When I would 3D scan big dies with a Creaform I would use MAGNAFLUX Dye Penetrant Developer. It worked good but the downside was also having to clean it up. One this I did think of was trying spray foot powder but I never had the time to try it.
Yes, dye pen developer, brand doesn't matter, is cheaper and is what everyone used until they developed the fancy sublimating sprays. It also has the benefit/disadvantage of sticking around until you have to clean it up. It does just wash/wipe off as it's basically just spray on chalk so it's not to bad. I've tried foot sprays and a bunch of other hobby solutions, it's not even close to as good or easy to work with. If you screwing around in your garage sure but if you're on the clock not worth it at all.
@@TheWidgetWorks Oh I am sure there are other brands. I was bringing this program online at my shop back in 2018 So at the time I did have some time to figure out a good process. Man I miss using a Creaform.
Hi from Canada. I have questions: When doing you torque calculations why did you not include the effect of the work piece(s), additional chucks, fixtures, etc.? At TS6.45 you said you only calculated for the table and magnetic chuck as load. Wouldn't the addition of your secondary chuck plus several pounds workpiece almost double the required torque? Lastly: could the manufacturer of your grinder have provided a 3D model? P.S. I really, really enjoy your channel.
There's lots of margin on X. And I can always slow it down if it's an issue.
Had same thought as a work piece could easily weigh as much as that chuck.
You so geeked out on this video😂😂😂
James, when I began watching ytube videos. I have watched ones that helped me to pick up with a hobby. Shut as when you made a Chuck Key for Quinn. I'll continue, and what you're doing. But I spent nearly 44 years with controllers, PLC. So I'm a bit burned out. So if you wish, pls send me a reply
What I found the hard way is, the motor drivers don't like motors hand cranked when electronic control is OFF. I used cheap TMC2209 drivers with Nema17 motors not the expensive kind mentioned here. Just be sure to check if you plan to hand crank machine when not using electronics for moving.
The Clearpath servos can theoretically handle this if they're powered, but not enabled. They can even do this with the power hub and logic backup 24V. I haven't tried it.
Crazy idea here, if you're rigging up your own software for this, why not put a pair of inductive proximity sensors on either side of the grinding wheels contact point. If you have one on each side the logic to automate grinding a pass should be straight forward, you can use the sensors to control the back and forth of the y axis without having to ever setup the machine.
1 Go to the top of the Y Axis
2 Go to the centre of the X and Z Axis
3 Lower the Y till you get a signal on both sensors
4 Advance Z "towards you" until you loose Signal on both sensors
5 go in the distance of the sensor offset from the actual wheel
6 Go all the way to the right on the X Axis until you loose both Sensors
7 Drop the machine down a bit on the Y ( I'd guess the amount of thou you want to take off)
8 Go Left across X till you loose the sensor signals
9 Advance Z by some Variable Amount (I'd guess set by rotary encoder knob or touch screen)
10 Go the other way on X till you loose the signal
repeat 9 and 10 till you loose signal on both sensors,
voila full pass no setup. just make sure the piece is over the center and not off to one side or you would be grinding the magchuck.
doing it this way it doesn't matter the size or the piece the machine can just hit go and it'll run until it's ground the top. the only issue I can see with this is the bracket to hold the sensors would need to be adjustable to take into account the wheel getting smaller as you dress it. I guess technically you could set up some sort of tool height sensor to set the height of the sensors but you'd need motor to control the height of the sensor bracket.
Awesome video! When you’re done you should sell it as a kit!
And invite the lawsuits and the usual bs on the phone all day listening to the whining I think not.
You say this wasn't the "sexy" part of the project..... I beg to differ!
This. Working out your requirements and translating them to specifications before implementation is always sexier than going broke doing the wrong thing and living with the result. On something this complicated, it's way too easy to get it wrong. One of my favorite aspects of your ELS project was the thoroughness and care you took in your performance requirements and how they translated to your component selections. It really minimized the redesign/rework parts of the project. (Plus, you got a lot of great channel content and products to sell!)
I'm sure you did some back-of-the-envelope guesstimates of the time & cost a similar approach would need for this grinder project, before deciding on a more OTS solution. When you're done, I hope we can get a comparison of those estimates vs. the actuals.
Intriguing! Any plans to put a scale on Y? I would on mine, but I'm curious to see how it goes for you.
Planning to use an algorithm to get reliable positioning, but I guess we'll see how that works
Hi James, I'm curious how you do the screen recordings in Fusion. You always move the views around and zoom in/out so smoothly - what's your technique?
I'm pretty sure he's said the space mouse (a 3d modeling tool) in some of his other videos.
@@ChristopherGoggans Yep, 3D mouse is the way to go when doinf 3D modeling of any kind.
Thanks for your video… ive been wanting to cnc my harig super 6x18 for a while now. How much is this gonna cost?
Hi James are you aware of the Bridgeport /Harig EZ surf grinder ? Bridgeport installed their CNC control on a Harig surface grinder, very similar
in size and form to your machine .Might be a useful reference for your project.
Dont reinvent existing technology
One question I'm interested in is the component of system behavior that corresponds to the fact that the servomotor's position/velocity/acceleration feedback is on the motor, rather than on the load. Is that somehow solved by the "Inertia Mismatch" calculations? It seems related, but I don't quite see how it tackles the issue directly.
That turns out to be an interesting question. The servo controller controls the motor, since that's where the encoder is located. The controller sees the motor's inertia directly, but the load's inertia is only visible through a coupling that has both elasticity and viscous damping.
The motor mounting and the load also have some degree of damping. These external factors are the reason the servo needs to be tuned to the load. "Inertia Mismatch" is a way to calculate how much you might be asking the controller to compensate for, but it's only part of the picture. The real excitement occurs in the frequency domain due to phase and resonance. There's a fairly detailed treatment here, if you're interested: www.automate.org/industry-insights/understanding-the-mysteries-of-inertia-mismatch
got any additional features for your ELS in the pipe line? position stop or perhaps another axis? spindle position?
Pretty much everything in this video went over my head. *_Surprising though the surface grinder has no auto feed._*
Oh they come with an auto feed for more$$$ when hes done he will have spent more but where's the fun in that.
Hello James, since you are the one I go to for CAD tutorials, I thought I'd ask you this question. I'm trying very hard to become a better designer and the terminology escapes me in this case.
In Constructive Solid Geometry, you create primitives that either add to the final geometry, or are used to cut away parts of other primitives. Is there a specific term for the "tool" that I create to be used to cut away from the original geometry? I seem to think that the term "slug" is correct, but I really have no idea.
Overkill really IS underrated, Sir!
Fusion just calls it a tool body.
Sir, Can you do a video with other industrial servos like Yaskawa,Panasonic etc? And can we use mach3 usb for such servos. ?
Are there no scale bars or coded reference markers available for your 3D scanner? Worked professionally with Steinbichler and GOM systems (in the $100´000 price range) for a couple of years and even tho the marketing ppl likes to tell you how good their free matching functions in the software are, if you want good tolerances scale bars and coded markers are superb. If you want even better, photogrammetry is the way to go.
GOM software used to be free in a somewhat limited, but still capable, version. But i guess its not free anymore since Zeiss came in and took over the show.
I don't think so. The blue laser units are the go-to solution in this class, but they're $5-10k.
Renzetti will be proud. Now....go hit sub-micron dimensions already!
Doing the fly-by-wire like Tony's Maho?
So refreshing to hear metric units instead of bananas and matchsticks 😂
Did you consider contacting the grinder's manufacturer to see if they have a CAD model?
I didn't.
Z is always depth of cut. as in the wheel up and down.
Depth of cut on a lathe can be two directions
@@douglassmith2055 So can a surface grinder (side wheel). So can a mill. So can a router. Plunge is Z
Hi,...There are not many detailed videos for Industrial Ac Servos especially for Mach3.! Those can't be used for Mach3 or what? The connections and setting parameters may vary from model to model I guess..! I think it would be awesome if We can use such industrial servos with Mach3. But the inertia tuning and all again ...oh ...!!
Yeah, Clearpath can be used with MACH3.
Since you invoked the name of Robin Renzetti, I will bring up Adam the Machinist, who takes CNC grinding to the next level. Pretty sure you have seen his videos, but mentioned him just in case.
What about space for the motor connector?
Hell ya!
Pragmatic Lee recently did something similar on his channel.
Holy crap did I "get a lot of presents" this week. :D
Hmmm, I didn't give myself a Christmas present for many years now...🤔
Nah, wouldn't fit in my shed anyway. 🤭