What parameter do you program to motor (if allow change settings) ? Tested by me NEMA 23 3,1Nm by factory defaultnis ser maybe to 50% of avaialble "power" after change settings to max limit is just stronger. Separate parameter for stall torque and for spinning. (iCS-2331-L)
Excellent video and testing. As I always tell my engineers in the words of the great W.Edwards Deming "without data you are just another person with an opinion"
Wow ! At last, an objective test. I've learned alot. Yes, and pinch of salt is always to be taken when reading specs, but your tests are truly revealing. The important point here is the test method, and not subjective reviews. Data doesn't lie, manufacturers do.... Well done, keep it at it.
I always learning something listening to your videos. And that is saying something since my wife constantly reminds me that I know everything. Look forward to the follow on video.
James, thank you for an intelligent comparison of these motors. Your channel is one of the _few_ channels where I'm impressed with content _and_ intelligence.
Thank you for this detailed test, this is very valuable for making an educated decision. I was considering going with closed loop, just because it's supposedly better, but seeing this, at the rpm I'll be using it, the difference to a digital driver is very small! You just saved me some money.
Well done James for putting in the effort to set all this up and do the editing! My CNC router runs 3Nm steppers on old analogue 80V drivers but I’m running at 70V as this really boosts the output over 48V.
This channel is the way, the truth, and the light. Also, you've lost a LOT of weight, James. Good on you man, keep taking care of yourself you look so much healthier than 2 years ago.
Thanks! I feel better, too. Last weekend, I needed to move a 350lb piece of machinery, and I just lifted it onto the pallet. That was really satisfying. :)
This was indeed helpful to my climbing the learning curve on these CNC motors and drivers. I especially liked the comparison at 31:20 and have very much "taken it for what it's worth". Good one! 😉
Thanks James. As always a very informative lesson & clearly articulated. You have a real gift in being able to show & explain complicated technical details in a way that most laymen can understand. Keep up the interesting videos, please... :) (& I like your quiet dry sense of humour too).
As an engineer, I agree with you. You always have to test key components to see if the data sheet is accurate. I've seen lots of power MOSFETs that have a safe-operating area much smaller shown in the data sheet.
This is gold!!! I wasn’t knowledgeable in any of these things before watching your videos. Now, I feel like I’ve been exposed to enough information (all from your videos) that I’m comfortable on understanding everything I’d need for buying the ELS parts. Thanks all to you James 💪🏻 That 180watt stepperonline motor your going to test sounds like the one for me if it does decent on your testing 🤞🏻 Great video sir 😎
I requested a quote for 4 Servos at DMM Technology and the guys offered me a very nice discount so, Is my way to go, plus DMM servos have a 16 bits absolute encoders which is awesome. I would compare a DST 0.4kW AC Servo Motor (164 usd) + DYN4 AC Servo Drive (235 usd) with that Clearpath. Thanks so much for your channel, it is always professionally done in content and editing, I just like it even before see it.
Thank you for doing this, lots of hearsay on the web, my first cnc was in 2004, and no one did this 👍 , this is a demostrative example to base decisions on. Btw, I know John at automation technologies personally and he’s a great person. I’ve done most of these comparisons and I’m going to the get the clear path for my z-axis on the g704👍😎. Awesome video, James.
Hi James, this investigation is valuable well beyond ELS as I'm sure you know and that is reflected in the title which will help with other search terms. Very interesting results and thank you. BobUK.
Just a thought to help ease up on the pressure -- if you tie the string around anything you can use as a lever, it would allow you to pull it against the vice and take a lot of variance out of pulling the string directly.
This is a great test, but also an excellent demonstration of why you shouldn't use the "smooth curves" plotting style in Excel with sparsely sampled data!
Really nice analysis. The last integrated servo are actually made by leadshine. I got the 200w made by JMC. They also have the same 180w ones. I hope the test comes out nice Becuase I got 3 of them! Great content as usually
@@marcus_w0 same for me. Got them directly from JMC. I have now discover that the 200w version exist also with 10mm shaft. The 11m shaft size was a real nightmare. I am using steel pulley and che 4mm keyway in the was a real challenge. Good to know that they are working well for you! I bought them Rodger with a 1.8kw AC servo to replace my Nema24 on the BF20L (simil G0704) and the spibdle motor. All servos is the way to go :-)
I really wanted to build a proper motor dyno with a load cell at each end of a friction strap and a leadscrew to gradually apply the force, but realized I could spend the next six months developing it. that would be fun, but wouldn't really get me closer to my goals. :)
@@Clough42 Your method of measuring torque is excellent for those of us still trying to grasp the fundamentals of torque and horsepower. Nothing is hidden inside some arduino code communicating with a load cell; everything is out in the open and easy to understand. I think your method is both educational AND serves as a low-cost torque measurement system. I like it.
Fantastic video! Loved it :) Your test rig is genius! I own a set of ClearPaths on my 3D printer and I love them. Quiet and powerful they are truly impressive equipment.
Super great video ... thanks!! Not sure it's been suggested but if you add compliance to your cord at the finger-pull-end (spring?), you'll get a much smoother tension control in finger pull vs output tension resulting in more consistent measurements in pull-out torque. It will adhere to the winch equation -> LoadTension = InputTension * e^(Drum-CordFriction*CordWrapAngle) for the mechanical gain of your cord+drum winch system. You'll be able to control the InputTension (and hence output tension) better.
Love this test. I was for a while ruminating over if I should invest in a servo system for my cnc. Have to say they don't appear to perform as well as I thought. Also, the clearpath system running at 72v should obviously take the lead given the voltage hike. I would imagine most of the other closed loop systems would have performed fairly comparable at that voltage, so this test is skewed in favour imo. That being said, I'll stick with my current dsp open loop system 👍👍
GREAT INFO! I would not have known how to test these. I have Nema 23s and a couple large Nema 34s and never knew really what the optimal speed/torque would be.
Thanks my dear, I will try to follow your input and transform my lathe, too. It will be a long term project as last time I had used trubopascla (25 years ago). I bought an arduino (knowing that probably it is not powerfull enough) but if it will work just a bit I will feel I won. Thanks for your videos, as a (small) youtuber I know how challanging it is and you make more then professional videos. I thank you a lot for your time and send my best regards from Florence. PS: When I should finish the project I will send you the link to my video...but don't expect it too soon :-)
Great video James!!! I purchased your ELS kit but have not installed it yet. I too did lots of research for the most affordable closed loop stepper for this ELS project. Many forums as well as the guys from Arizona CNC Kits (Who have tested loads of Nema 23/24 closed loops steppers) & Franco from Franco CNC directed me to a very affordable 4.0 Nm(566.56oz.in) closed loop stepper/32 bit digital driver (Stepperonline item# 1-CL57Y-24HE40). This closed loop stepper/driver kits sell for $90.56. It "appears" to be a nice fit for your ELS project. As I will not be using these components for 5+ months, I would be happy to loan the closed loop stepper/driver to you to put it on your well thought out testing rig. Thanks... Richard
Hey. Look at my little 570oz stepper GO! I need upgrade to the KL-5056D. I wonder what the KL-5056E would do. Does the full step, half step, quarter step change anything with the torque at speeds? I think I had it set for 1/8 steps. The Gecko drivers does some magic with the steps by converting the pulse train to full steps when it can. Good stuff!!!
Hi James and thanks for an incredibly useful video. Would be very valuable to measure the power off ‘residual torque’ (for lack of a better expression) of these steppers / servos. Sometimes you want to operate a machine manually while the stepper / servo is coupled but off. Then if considerable, ‘residual torque’ can play havoc with manual accuracy. Many thanks for an excellent channel.
I was excited to see the results for this, and you did not disappoint! As soon as you hooked up the Clearpath my first thought was "he better have run autotune in that setup". My experience with them is that they are garbage with factory settings, you absolutely cannot get away with running them untuned. Thanks for doing this test!
@@Clough42 sorry, forgot to add that I was glad when you said you'd done the autotune. But yeah, they use a lot of "smart" functions for handling torque based off expected performance.
@@Clough42 one more thing, if you want to test some more motors might I recommend Leadshine? They don't seem to be really well known, but they've got some supposedly high torque motors that are more compact than a lot of the standard Chinese clone steppers.
Same here, although my professor would have turned me inside out if i had used the smoothed graph lines in a diagram. "If you did´nt measure it, don´t graph it."
Awesome video. Earned a new sub. My garage cnc uses the Clearpath SDSK-2346P-RLN and I really like them. Between the Clearpath and a Jianken 3.2kw spindle you can barely hear my machine running beyond the noise from the cutter. I remember the days when we all ran routers and it would run you out of the room with the noise. Again, great video. Thanks!
Great video! I built a similar machine (CP nema 32's with Jianken spindle) It is great :) FWIW I don't think the Teknic PS and cables are magic. I used a big, open-frame supply from AD, made up my own cables and they're fine. Just follow the specs and you can save a few bucks
@@rbyt2010 If you got the ATC option on your Jianken spindle Piotr Fox posted a link on some lower cost ISO25 tool holders. I bought them and have to say the runout measured as good as the $100 a piece tool holders from Parfaite. I think I spent just over $100 for 5 tool holders.
Really interesting James! Thank you for that content. I’d love to have an ELS for my 15x60 lathe. I only know a little about this stuff so designing a full system is outside of my comfort zone for sure.
It would be nice back2back test also following combo: motor 23HS2430B, CL driver 2HSS86H. Together with an 1000ppr encoder it forms quite low cost combo with reasonable performance. I've tested it set to 3A RMS, 6.8A max. current draw in my application by a proper rotational torque meter, it tops at about 90W, but the resulted speed/torque characteristics can't be directly comparable with the methodology of this video for sure. sidenote: drivers configuration (esp. the current limit in closed loop and both PI controllers terms) are key to maximise the perfo. Tuning the RMS current up to the thermal limits of given motor for application duty cycle is the way to get the most of the system.
3D printed hanger also elasts little when force applied, so you have to substract that too, but doesn't matter if you are using same setup for motor to motor comparison.
ShOcK & AWE comprehensive video. A UA-cam video that all others will be compared. You "are" the Servo-Stepper Whisperer. Must See TV is back. Yeah, I'm pretty wound up. Subbed! :)
The resonance should change with different microstep settings I guess (the resonance point will shift). You're hitting midband resonance. The other Servo is a JMC 180W Nema23 Motor, they're really nice. That motor should be more than enough for your setup. Mabye I skipped it but nowhere I heard something about your current settings.
Very informative, thanks! I'd be very interested in a module on measuring/calculating required torque for a specific application, i.e. what size motor is needed to drive a lathe leadscrew or whatever. (I had to just go with the old SWAG method for my ELS.)
True, that would be helpful. Although there would be a lot of assumptions about the friction in the system, which would require a big safety margin. For the rest, it´s simple physics. You would need the mass / geometry of every accelerating part though.
@@Clough42 holding torque really does not mean anything when you are trying to move things. But it is how manufacturers suck customers in. They should just weld the shafts on stepper motors. 1,000 foot pounds of holding torque! This puppy ain't going no place.
100 IPM is not bad. It's what I limit my CNC machine to. Any faster and I have trouble stopping it when I jog. But if you want to go faster you might have more power. It depends how close your drives are to your motors rated current. Being over or under will cause power loss. If you don't measure it there's no real way of knowing for sure. Don't believe what the drive says it is set at. It may be right and it may not be. My money is on if you've never checked it then it is probably not right. Break into a phase circuit with an ammeter and then jog real slow looking for the highest current reading you can find. Just start and stop a lot and take readings. They'll go up and down depending on where you are in the sequence. But after a bit you should be able to figure out where the high is. Compare that reading to what your motor current should be. It is like the Price is Right rules. As close as you can get it without going over is how you win. Over current stepper motors run like crap. So under is better than over. Over and you lose. Way under and you lose then too. I'd rather be a little under than a little over though. Make a video about it. I want to see what kind of drives you're running.
@@AmateurRedneckWorkshop yeah you might be able to get more. What voltage do you run your drives at? Low voltage steppers run out of power at slower speed. Every volt nets you more. A surprising amount more. But every drive has an input limit. Better drives run higher voltage. It's what makes them better.
I like your shop, mine is a tiny shop 8' x 16'. No off the shelf mill or lathe. I have a Skil drill press that I have built a conversion package for I call it SkillMan, that allows me to due some machining.
Maybe the higher reading at 1500rpm is due to how you are measuring the torque. If it's turning faster, then you have to account for the inertia of the core itself adding up to that final torque reading at the moment the cord brake grabs onto the cylinder.
Given the work that @vez3d and @miragec are doing for fast 3d printers, it would be super interesting to see the odrive with their 270kv & 150kv bldc motors tested with this methodology for comparison. I’ve been consider using some nema23 stepper or changing to the clearpath or odrive.
I hope you'll make what ever changes in software and such available again based on where you end up on the ELS...if so, I'm in for the changes (Unless it's Clear Path!)
You deserve far more viewers with content like this.
I couldn't agree more. :)
@@Clough42 with content like this, they’ll come pouring in at some point no doubt
Sorry but I can only confirm.
What parameter do you program to motor (if allow change settings) ? Tested by me NEMA 23 3,1Nm by factory defaultnis ser maybe to 50% of avaialble "power" after change settings to max limit is just stronger. Separate parameter for stall torque and for spinning. (iCS-2331-L)
not to @@Clough42 why stepper motor a not a dc motor out of the mobility scooter ?, can you test one of them.
Excellent video and testing. As I always tell my engineers in the words of the great W.Edwards Deming "without data you are just another person with an opinion"
This is some thorough testing but when you know your stuff, you know your stuff. Thanks for your effort 👍
Wow ! At last, an objective test. I've learned alot. Yes, and pinch of salt is always to be taken when reading specs, but your tests are truly revealing. The important point here is the test method, and not subjective reviews. Data doesn't lie, manufacturers do.... Well done, keep it at it.
I am amazed at your patience. Very impressive.
I always learning something listening to your videos. And that is saying something since my wife constantly reminds me that I know everything. Look forward to the follow on video.
James, thank you for an intelligent comparison of these motors. Your channel is one of the _few_ channels where I'm impressed with content _and_ intelligence.
Very understandably explained, even for non-English speakers. One of the best videos I have ever seen. Please keep it up! Thumbs up!
Most excellent analysis...hmmm, now I understand why I’ve been creating “multi-start” threads so successfully 🤔
Excellent testing. Just the right amount of introduction upfront and then into the testing. Your production skills are fantastic. Thank you!
Thank you for this detailed test, this is very valuable for making an educated decision. I was considering going with closed loop, just because it's supposedly better, but seeing this, at the rpm I'll be using it, the difference to a digital driver is very small! You just saved me some money.
And now you're on servos 😉
@@mmmikem Indeed, the quest for more and more speed has finally lead me there.
Well done James for putting in the effort to set all this up and do the editing! My CNC router runs 3Nm steppers on old analogue 80V drivers but I’m running at 70V as this really boosts the output over 48V.
"This is why we test". Top-notch video!
Fantastic testing and thoroughness. Thanks for the video
Best video on this topic after watching several videos, thank you.
This channel is the way, the truth, and the light. Also, you've lost a LOT of weight, James. Good on you man, keep taking care of yourself you look so much healthier than 2 years ago.
Thanks! I feel better, too. Last weekend, I needed to move a 350lb piece of machinery, and I just lifted it onto the pallet. That was really satisfying. :)
This was indeed helpful to my climbing the learning curve on these CNC motors and drivers. I especially liked the comparison at 31:20 and have very much "taken it for what it's worth". Good one! 😉
"I'm all out of motors" and presumably money, given the price of some of these motors, yikes! Nice procedure and results!
I have some new clearpath motors if any one is interested.
@@arinchang6579 Definitely interested!
Thanks James. As always a very informative lesson & clearly articulated. You have a real gift in being able to show & explain complicated technical details in a way that most laymen can understand. Keep up the interesting videos, please... :) (& I like your quiet dry sense of humour too).
The test method is brilliant, I mean rolling the string over the motor hub and pulling on one end
I just found you today, and I like your videos. Pure simple engineer mindset you have, very smart one.
Great comparison! Thank you for testing these so thoroughly and reporting your results.
As an engineer, I agree with you. You always have to test key components to see if the data sheet is accurate. I've seen lots of power MOSFETs that have a safe-operating area much smaller shown in the data sheet.
One of my favorite channels. Quality content.
This is gold!!! I wasn’t knowledgeable in any of these things before watching your videos. Now, I feel like I’ve been exposed to enough information (all from your videos) that I’m comfortable on understanding everything I’d need for buying the ELS parts. Thanks all to you James 💪🏻
That 180watt stepperonline motor your going to test sounds like the one for me if it does decent on your testing 🤞🏻 Great video sir 😎
Perhaps the best thing I've watched on UA-cam this year. Well done. Great content. 👍
Very interesting. Thanks a lot for putting in the work and sharing the results!
I requested a quote for 4 Servos at DMM Technology and the guys offered me a very nice discount so, Is my way to go, plus DMM servos have a 16 bits absolute encoders which is awesome.
I would compare a DST 0.4kW AC Servo Motor (164 usd) + DYN4 AC Servo Drive (235 usd) with that Clearpath.
Thanks so much for your channel, it is always professionally done in content and editing, I just like it even before see it.
So many motors... So little time... So much money. :)
Thank you for doing this, lots of hearsay on the web, my first cnc was in 2004, and no one did this 👍 , this is a demostrative example to base decisions on. Btw, I know John at automation technologies personally and he’s a great person. I’ve done most of these comparisons and I’m going to the get the clear path for my z-axis on the g704👍😎. Awesome video, James.
This is very helpful. Thanks for taking the time to pull this together.
I have closed loop Nema 24s & 34 on my pm-25mv and they seem to be adequate. I really enjoyed your entire presentation. Another great video.
OOo really looking forward to that clearpath/stepperonline battle!
Hi James, this investigation is valuable well beyond ELS as I'm sure you know and that is reflected in the title which will help with other search terms. Very interesting results and thank you. BobUK.
fantastic and helpfull testing , thank you
finally someone using a prony brake instead of the scale test :) Nice work
The DOE was fantastic, I know it involved a lot of work. The data looks very interesting, especially when upgrading to a digital driver. Thanks! 👍🏼
Excellent video! Can't wait to see the comparison of the ClearPath vs. the StepperOnline Easy Servo!
Expecting the Project Farm music to start any moment now.
We're gonna test that!
Just a thought to help ease up on the pressure -- if you tie the string around anything you can use as a lever, it would allow you to pull it against the vice and take a lot of variance out of pulling the string directly.
This is a great test, but also an excellent demonstration of why you shouldn't use the "smooth curves" plotting style in Excel with sparsely sampled data!
Please take a look at the popular JMC servo with integrated driver 180 Watt. There is also a 400Watt version, and a new 200watt.
Hadn't heard of the idea of "electronic leadscrew" before this video - very cool idea!
Really nice analysis. The last integrated servo are actually made by leadshine. I got the 200w made by JMC. They also have the same 180w ones. I hope the test comes out nice Becuase I got 3 of them! Great content as usually
@@marcus_w0 same for me. Got them directly from JMC. I have now discover that the 200w version exist also with 10mm shaft. The 11m shaft size was a real nightmare. I am using steel pulley and che 4mm keyway in the was a real challenge.
Good to know that they are working well for you! I bought them Rodger with a 1.8kw AC servo to replace my Nema24 on the BF20L (simil G0704) and the spibdle motor. All servos is the way to go :-)
Very helpful! Always love your videos keep up the great work
Fantastic as always! Thanks for your work!!! Huge help.
Great video! Hope to see last test shootout, using this to pull the trigger on a buy.
I very much like your method of measuring torque; I had not seen that before. I love electro - mechanical projects. Thanks.
I really wanted to build a proper motor dyno with a load cell at each end of a friction strap and a leadscrew to gradually apply the force, but realized I could spend the next six months developing it. that would be fun, but wouldn't really get me closer to my goals. :)
@@Clough42 Your method of measuring torque is excellent for those of us still trying to grasp the fundamentals of torque and horsepower. Nothing is hidden inside some arduino code communicating with a load cell; everything is out in the open and easy to understand. I think your method is both educational AND serves as a low-cost torque measurement system. I like it.
Fantastic video! Loved it :) Your test rig is genius! I own a set of ClearPaths on my 3D printer and I love them. Quiet and powerful they are truly impressive equipment.
Super great video ... thanks!! Not sure it's been suggested but if you add compliance to your cord at the finger-pull-end (spring?), you'll get a much smoother tension control in finger pull vs output tension resulting in more consistent measurements in pull-out torque. It will adhere to the winch equation -> LoadTension = InputTension * e^(Drum-CordFriction*CordWrapAngle) for the mechanical gain of your cord+drum winch system. You'll be able to control the InputTension (and hence output tension) better.
Oh, good idea. I didn't think of that.
Love data! LOVE a well-presented analysis of said data. And live your test rig.
Great content! I didn't fall asleep until 39:20 when it got quiet!
This is very valuable information. Thanks for taking all the time to do this study.
Killer video James. Gold
Very cool project. I'm building this for my old Grizzly G4000 lathe, going with the Rattm Motor 3N.m Closed Loop Set
Nice work standardizing your test method, good head to head data.
I just started watching this. My wife looks over and says "oh, it's your friend" 😁
Bravo for knowing the proper name of the Prony brake! ;-)
Love this test. I was for a while ruminating over if I should invest in a servo system for my cnc. Have to say they don't appear to perform as well as I thought. Also, the clearpath system running at 72v should obviously take the lead given the voltage hike. I would imagine most of the other closed loop systems would have performed fairly comparable at that voltage, so this test is skewed in favour imo. That being said, I'll stick with my current dsp open loop system 👍👍
GREAT INFO! I would not have known how to test these. I have Nema 23s and a couple large Nema 34s and never knew really what the optimal speed/torque would be.
Thanks my dear, I will try to follow your input and transform my lathe, too. It will be a long term project as last time I had used trubopascla (25 years ago). I bought an arduino (knowing that probably it is not powerfull enough) but if it will work just a bit I will feel I won. Thanks for your videos, as a (small) youtuber I know how challanging it is and you make more then professional videos. I thank you a lot for your time and send my best regards from Florence.
PS: When I should finish the project I will send you the link to my video...but don't expect it too soon :-)
Nice job one more time!
As I understand probably the second line on the clear path motor is the 100%dude factor load.
Really enjoyed this content! DMM Servo is what I use. Would love to see what they do against the Clearpath
Great video James!!! I purchased your ELS kit but have not installed it yet. I too did lots of research for the most affordable closed loop stepper for this ELS project. Many forums as well as the guys from Arizona CNC Kits (Who have tested loads of Nema 23/24 closed loops steppers) & Franco from Franco CNC directed me to a very affordable 4.0 Nm(566.56oz.in) closed loop stepper/32 bit digital driver (Stepperonline item# 1-CL57Y-24HE40). This closed loop stepper/driver kits sell for $90.56. It "appears" to be a nice fit for your ELS project. As I will not be using these components for 5+ months, I would be happy to loan the closed loop stepper/driver to you to put it on your well thought out testing rig. Thanks... Richard
I saw that "4.0Nm" kit, but if you look at the torque curve, it only claims just over 2.5Nm pull-out torque.
Hey. Look at my little 570oz stepper GO! I need upgrade to the KL-5056D. I wonder what the KL-5056E would do. Does the full step, half step, quarter step change anything with the torque at speeds? I think I had it set for 1/8 steps. The Gecko drivers does some magic with the steps by converting the pulse train to full steps when it can. Good stuff!!!
Incredibly well done analysis !!!
Hi James and thanks for an incredibly useful video. Would be very valuable to measure the power off ‘residual torque’ (for lack of a better expression) of these steppers / servos. Sometimes you want to operate a machine manually while the stepper / servo is coupled but off. Then if considerable, ‘residual torque’ can play havoc with manual accuracy. Many thanks for an excellent channel.
Excellent information I Need a lot of that next year as I am working on my Project
Absokuly amazing information!! Thanks James.
I was excited to see the results for this, and you did not disappoint!
As soon as you hooked up the Clearpath my first thought was "he better have run autotune in that setup". My experience with them is that they are garbage with factory settings, you absolutely cannot get away with running them untuned.
Thanks for doing this test!
I did. Before autotune, the spool had enough inertia to cause all kinds of crazy behavior.
@@Clough42 sorry, forgot to add that I was glad when you said you'd done the autotune. But yeah, they use a lot of "smart" functions for handling torque based off expected performance.
@@Clough42 one more thing, if you want to test some more motors might I recommend Leadshine? They don't seem to be really well known, but they've got some supposedly high torque motors that are more compact than a lot of the standard Chinese clone steppers.
Just as a notice, for the hybrid motor, the torque curve was for 75v and you run 48v if I'm not mistaken.
Brought back memories of physics lab! Enjoyed this one very much. Thanks.
Same here, although my professor would have turned me inside out if i had used the smoothed graph lines in a diagram. "If you did´nt measure it, don´t graph it."
Fair enough. When you get a lot of lines, the smooth lines make it easier to track a single line across the graph.
This just energized me. Great work!
Also, I would love to seem the math you used in that spread sheet to calculate Torque
Awesome video. Earned a new sub. My garage cnc uses the Clearpath SDSK-2346P-RLN and I really like them. Between the Clearpath and a Jianken 3.2kw spindle you can barely hear my machine running beyond the noise from the cutter. I remember the days when we all ran routers and it would run you out of the room with the noise. Again, great video. Thanks!
Yeah, I need to tear in to my mill to hunt down some backlash that has crept in, and I'm seriously thinking about the Clearpaths.
@@Clough42 My issues were in the shaft couplings. Changed to Zero-Max brand and everything improved 100%.
Great video! I built a similar machine (CP nema 32's with Jianken spindle) It is great :) FWIW I don't think the Teknic PS and cables are magic. I used a big, open-frame supply from AD, made up my own cables and they're fine. Just follow the specs and you can save a few bucks
@@rbyt2010 If you got the ATC option on your Jianken spindle Piotr Fox posted a link on some lower cost ISO25 tool holders. I bought them and have to say the runout measured as good as the $100 a piece tool holders from Parfaite. I think I spent just over $100 for 5 tool holders.
@@machine2747 Thanks, will check it out. Little behind on Piotr...
Very nice, always want to see such a thorough comparison between those motors. Also would like to see the comparison between spindle motors.
Really interesting James! Thank you for that content. I’d love to have an ELS for my 15x60 lathe. I only know a little about this stuff so designing a full system is outside of my comfort zone for sure.
The Voltage is important for the torque. So 72 V against 48V differs wide.
Seems like cash thrown is the critical factor.
Current limits torque, not voltage. Voltage limits speed. Voltage limits current at high speeds.
Very good presentation just watched it
Awesome video! Detailed and insightful. Thank you!
It would be nice back2back test also following combo: motor 23HS2430B, CL driver 2HSS86H. Together with an 1000ppr encoder it forms quite low cost combo with reasonable performance. I've tested it set to 3A RMS, 6.8A max. current draw in my application by a proper rotational torque meter, it tops at about 90W, but the resulted speed/torque characteristics can't be directly comparable with the methodology of this video for sure.
sidenote: drivers configuration (esp. the current limit in closed loop and both PI controllers terms) are key to maximise the perfo. Tuning the RMS current up to the thermal limits of given motor for application duty cycle is the way to get the most of the system.
An awesome, truly useful video. Many thanks!
3D printed hanger also elasts little when force applied, so you have to substract that too, but doesn't matter if you are using same setup for motor to motor comparison.
ShOcK & AWE comprehensive video. A UA-cam video that all others will be compared. You "are" the Servo-Stepper Whisperer. Must See TV is back. Yeah, I'm pretty wound up. Subbed! :)
The resonance should change with different microstep settings I guess (the resonance point will shift). You're hitting midband resonance. The other Servo is a JMC 180W Nema23 Motor, they're really nice.
That motor should be more than enough for your setup. Mabye I skipped it but nowhere I heard something about your current settings.
Superb presentation and details of the comparison cheers
Very informative, thanks!
I'd be very interested in a module on measuring/calculating required torque for a specific application, i.e. what size motor is needed to drive a lathe leadscrew or whatever. (I had to just go with the old SWAG method for my ELS.)
True, that would be helpful. Although there would be a lot of assumptions about the friction in the system, which would require a big safety margin. For the rest, it´s simple physics. You would need the mass / geometry of every accelerating part though.
Interesting comparison. My steppers on my cnc plasma seem to run out of go at about 100 inch per minute. Thanks for the video.
It totally depends on the dynamics. The KL-5056D with the "566oz.in" motors can run my G0704 at 400IPM, but I don't recommend it.
@@Clough42 holding torque really does not mean anything when you are trying to move things. But it is how manufacturers suck customers in. They should just weld the shafts on stepper motors. 1,000 foot pounds of holding torque! This puppy ain't going no place.
100 IPM is not bad. It's what I limit my CNC machine to. Any faster and I have trouble stopping it when I jog. But if you want to go faster you might have more power. It depends how close your drives are to your motors rated current. Being over or under will cause power loss. If you don't measure it there's no real way of knowing for sure. Don't believe what the drive says it is set at. It may be right and it may not be. My money is on if you've never checked it then it is probably not right. Break into a phase circuit with an ammeter and then jog real slow looking for the highest current reading you can find. Just start and stop a lot and take readings. They'll go up and down depending on where you are in the sequence. But after a bit you should be able to figure out where the high is. Compare that reading to what your motor current should be. It is like the Price is Right rules. As close as you can get it without going over is how you win. Over current stepper motors run like crap. So under is better than over. Over and you lose. Way under and you lose then too. I'd rather be a little under than a little over though. Make a video about it. I want to see what kind of drives you're running.
@@1pcfred Thanks for the tips. I will keep the video in mind but no promise.
@@AmateurRedneckWorkshop yeah you might be able to get more. What voltage do you run your drives at? Low voltage steppers run out of power at slower speed. Every volt nets you more. A surprising amount more. But every drive has an input limit. Better drives run higher voltage. It's what makes them better.
I like your shop, mine is a tiny shop 8' x 16'. No off the shelf mill or lathe. I have a Skil drill press that I have built a conversion package for I call it SkillMan, that allows me to due some machining.
Fantastic video, both for the methodology (I was wondering how I would set up a rig for testing) and the comparison of the results.
please, feel free to talk more about motors.
Yeah, that was excellent
lol
@@ikbendusan Genuine comment. I /actually/ want him to? :)
This is a great video, thank you.
thanks very much for doing this video, very interesting.
Maybe the higher reading at 1500rpm is due to how you are measuring the torque. If it's turning faster, then you have to account for the inertia of the core itself adding up to that final torque reading at the moment the cord brake grabs onto the cylinder.
Given the work that @vez3d and @miragec are doing for fast 3d printers, it would be super interesting to see the odrive with their 270kv & 150kv bldc motors tested with this methodology for comparison. I’ve been consider using some nema23 stepper or changing to the clearpath or odrive.
Very interesting. Thank you James for all the work and time you’ve put into this. 👍👍😎👍👍
I hope you'll make what ever changes in software and such available again based on where you end up on the ELS...if so, I'm in for the changes (Unless it's Clear Path!)
All of these motors, including the Clearpath are compatible with the software as-is.
Thanks for this valuable information
Will try
Nice setup!
thanks for testing. great infomation!
It is soooo reliefing that you're using the metric system. Cheers mates!
Very nice motor test