PDA

View Full Version : Some questions to JAZZCNC or any other closed loop stepper experts



A_Camera
11-06-2021, 01:30 PM
I am currently considering this kit: https://www.ebay.co.uk/itm/393192724008

It's a 4 Nm closed loop closed loop stepper, and my intentions are to replace the 1.8 Nm stepper which is used in my "poor man's lathe" project: http://www.mycncuk.com/threads/14589-4th-axis-as-a-poor-man-s-lathe which I started just for fun and to see if a rotary axis can be used as a simple "lathe" type of machine. It turned out that at least drilling works well, so I continued with this activity and will test even turning probably during the weekend. Currently I manage to run the stepper at 2280rpm reliably but because of the 1:6 gearing, that results in 380rpm on the chuck, which I am well aware of is a bit low, especially for soft material. So I have two changes which I planned.

1. To replace the gearing with a 1:3 ratio. This will result in double speed, but also half the current torque. Maybe that's still better on soft material than low speed and higher torque, but to compensate the loss of torque, I also plan...

2. to replace the stepper with a closed loop stepper. In theory, it should mean I get much better torque all the way because the driver compensates for the step loss and applies more current when higher torque is necessary. In theory, I should also be able to run the motor at faster speeds than a stepper motor could run. So the combination of those two is beneficial (higher motor speed and torque), the motor I am aiming at is a 4 Nm motor which is more than twice of the stepper I now use. Of course, I still plan to change the gear ratio also, so with all that, I aim at having around 1000 chuck rpm which would be phenomenal considering this fairly cheap solution.

I have no experience with closed loop steppers, so I have a few questions.

Are all closed loop steppers and drivers equal? I don't mean specification wise, but more like quality wise. Does it make a difference who is selling them? What about the drivers? Does it matter which one is used for the same motor? Any pitfalls? I know you prefer AC servos, but currently I am concentrating on DC. I also know there are larger motors which are better suitable for this sort of task, but those are too large for my current ambitions, so I'd like to keep the NEMA 23/24 form factor and also don't want to build or buy a new PSU, so I plan to use my latest 48V 340W DIY PSU, even though the above kit comes with a PSU. Can I trust the torque table? I know it says 1600 microstep, which is wrong since they must mean 8 microstep (1600 steps/rev) but if that is true then at least at 1500rpm I will get much more torque since I intend to run it at 2 microstep (400 steps / rev).

A last question, which is maybe the most important to find an answer to. The listing says maximum rpm is 2000. I asked the seller if that is true and he replied within an hour (very good response time) that the maximum rpm is indeed 2000 rpm. He did not explain why, even though I asked. The encoder is good for 5000 rpm, so why is there a limit? Or is it possible that the seller does not know everything? I mean, normally, a stepper can be run at almost any speed in theory, as long as the pulses are supplied within the specifications. In theory. In reality, that's different, which is why my stepper stalls at 2400 rpm, but at 2280 rpm it is running happily forever. So why is there a 2000 rpm limit on the closed loop stepper? Is that a built in sort of hard limit? Or is due to something else, or maybe it is like the stepper, after a certain rpm it becomes unreliable, so the seller says 2000 rpm is the maximum, because he knows that it works up to that value?

OK, many questions and long text, but I hope I can get at least some answers, hopefully at least the last part will get answered, because I would like to increase the speed, and my hope is to be able to run the stepper at 3000 rpm. Maybe that's too ambitious...

JAZZCNC
11-06-2021, 08:23 PM
Ok first off Not all closed-loop drives are equal.!! (or motors)

I haven't used those drives or motors so this doesn't apply to them at all, however, In the past, I have used drives that look very similar (probably come out of the same factory) and wasn't very impressed.
The reason is two-fold, firstly, compared to the Lichuan drives I use now they gave much rougher stepping action, mostly down to the way they handle resonance I believe.
Secondly, they gave lower torque at higher RPM, which I believe is down to how they handle the current/volts combined with resonance plus only allowing 50V max. They struggled much past 1500rpm which is where the extra volts come into play, the 4.5Nm Lichuan's I use now will happily shift a 40-50Kg gantry with 10mm pitch screw between 20-25Mtr/min using 55Vac which translates to 2000 to 2500Rpm.

Regards the motors then most closed-loop steppers are pretty much the same motor as a standard stepper, just with an encoder fitted. It's the drives that make the difference.
I say most because some are better than others and again I find the Lichuan motors to be made just that bit nicer than some others. Also, you have 3 phase steppers and these are different animals altogether. They are smoother and have more torque and spin faster, unfortunately, you don't find many in nema23/4 with much above 3Nm ratings and 50Vdc drives, that said they will easily match a 4Nm 2 phase using the same voltage.

On large machines like 8x4, or 10x5, I fit 10nm closed-loop Nema 34's with mains voltage drives and they stop for nothing and spin crazy fast. Much faster than same size 2 phase motors.

These days I nearly always fit drives that allow AC voltages for simplicity and cost savings, I have run them on both AC and DC using toroidal transformers with smoothing Caps and can see no difference in performance. Now I just hook them up to a 55VAC transformer(toroidal) and away they go.


Have you considered looking at DC servos.? These spin plenty fast, usually 3000rpm + and give constant torque with lower voltages like 48Vdc.
Something Like these https://www.ebay.co.uk/itm/283050605349?hash=item41e7217b25:g:hlAAAOSwKJhbRIe E

They are low torque but it's constant across the RPM and they will spin up to 4500rpm with no or low loads. This will work better with your ratios as you'll have far more torque at the higher rpm's than a stepper will have.

Other than that then I'd say spend a little more and get a 200w AC servo with a brake and do the job right.!

Muzzer
12-06-2021, 12:24 AM
Those "DC servos" are closed loop steppers with the driver and encoder integrated into the motor, which is pretty neat. You just supply the DC power (48V or so) and the step/dir signals and it does the rest. These seem to be clones of the Leadshine Integrated Easy Servo family, although I've no idea how well they've been cloned. http://www.leadshine.com/producttypes.aspx?type=products&category=stepper-products&producttype=integrated-stepper-motors

I used a 2Nm Leadshine on the Z axis of my Bridgeport conversion and it's been seamless. Having said that, I've only used proper AC servos since, the last lot being the Lichuan A4 servos that JAZZCNC has been using. They are a bit basic on the software front compared to a Yaskawa but seem to work adequately and are a fraction of the price. You can also swap the cooling fans for something more civilised easily enough if you don't want to have to wear ear defenders(!).

JAZZCNC
12-06-2021, 01:34 PM
Those "DC servos" are closed loop steppers with the driver and encoder integrated into the motor, which is pretty neat. You just supply the DC power (48V or so) and the step/dir signals and it does the rest. These seem to be clones of the Leadshine Integrated Easy Servo family, although I've no idea how well they've been cloned.

Don't think they are steppers Muzzer, I have never used any but I've seen them fit on machines and they certainly don't sound like steppers. They sound just like servo's whisper quiet and smooth, plus they spin much much faster. I have used plenty of Leadshine Easy servos which are definitely steppers, 3 phase steppers.

Muzzer
12-06-2021, 02:43 PM
Perhaps you are right but there is mention of "1.2 degree step angle" and "closed loop stepper vs open loop stepper" in the description. The photo of the internals also looks pretty much identical to my Leadshine. It's often hard to tell from the websites, as they are often cobbled together from all sorts of random graphics and machine translated Chinglish.

I see that JMC do both integrated stepper and integrated servo motors in the same mechanical concept, so there ARE some out there. They do a 400W, 60mm, 3.5Nm (pk), 3000rpm true AC servo for about 100 on AliExpress etc, which is starting to sound quite interesting. It's only 48V though, so would need a meaty PSU. It's possible the link earlier is to a JMC or one of its clones of course.

https://www.jmc-motor.com/product/978.html
https://www.jmc-motor.com/ntegratedservomotors.html

JAZZCNC
12-06-2021, 03:26 PM
Perhaps you are right but there is mention of "1.2 degree step angle" and "closed loop stepper vs open loop stepper" in the description. The photo of the internals also looks pretty much identical to my Leadshine. It's often hard to tell from the websites, as they are often cobbled together from all sorts of random graphics and machine translated Chinglish.

I see that JMC do both integrated stepper and integrated servo motors in the same mechanical concept, so there ARE some out there. They do a 400W, 60mm, 3.5Nm (pk), 3000rpm true AC servo for about 100 on AliExpress etc, which is starting to sound quite interesting. It's only 48V though, so would need a meaty PSU. It's possible the link earlier is to a JMC or one of its clones of course.

https://www.jmc-motor.com/product/978.html
https://www.jmc-motor.com/ntegratedservomotors.html

Ah maybe we are at cross purposes here, I was talking about the link I gave to DC servos which are just like those JMC motors, the link the OP gave was for a closed-loop stepper.
To be honest it was the JMC I was thinking about but couldn't remember where I'd seen them..:hysterical:

A_Camera
13-06-2021, 07:26 PM
Thank you for the advices and answers. I really appreciate your comments.

There is a lot to be considered and since I have no experinece with closed loop steppers or servo, I really don't know why a weaker servo is more expensive than a stronger closed loop stepper.

I had a closer look at this one: https://www.omc-stepperonline.com/integrated-servo-motor/nema-23-integrated-easy-servo-motor-180w-3000rpm-0-6nm-84-98oz-in-20-50vdc-brushless-dc-servo-motor.html

It is very similar to the one recommended by Jazz, bur it is a Stepperonline version. There is though one difference which immediately is obvious, that is the the Stepperonline version has no enable signal. It is not clear to me when the motor is enabled and when not. I mean, I am actually using the enable signal, and I want to be able to disable it so that I can manually rotate the spindle if I want to, without having to power off everything. It is not clear to me how that can be done unless there is an enable signal. It is also not very clear to me if this is a 3-phase servo or not. The one in Dean's link looks like a 3-phase motor.

But there is another question... that is about the torque. Both Dean's, and the one here have considerably less torque than 4Nm... so would a closed loop stepper with 4Nm be equal to these two...? I mean, high rpm and smooth, quiet running is one thing and is very nice, but if it is at the expense of less torque then I don't see the benefits. I need this motor to be able to hold the chuck when I use it as a rotary axis, not just rotating fast. So what's the secret? Can these servos be used as ordinary rotation axis steppers also, and can they hold the position when used on the CNC mill?

JAZZCNC
13-06-2021, 09:22 PM
There is a lot to be considered and since I have no experinece with closed loop steppers or servo, I really don't know why a weaker servo is more expensive than a stronger closed loop stepper.

The servo at first might appear to be weaker, however, the stepper torque rating is when the motor is stood still, the moment it starts turning the torque drops away, and as the rpm's rise it drops dramatically to the point where at say 1500rpm you will have very little torque left. A servo motor rating is continuous across the RPM so at the rated speed you will have the motor's full torque rating, also they have a peak rating which can be between 3 x and 5x the rated torque for a number of seconds.
Often this is more than enough to accelerate an axis up to speed before dropping back to the rated torque which again is often enough to drive the axis and then peak again to stop it.
Then if you apply a ratio like what you will be doing, let's say 6:1, then your 0.6Nm becomes 3.6Nm at the output, and let us say the peak of 3x the rated then you have nearly 11Nm of torque for short bursts of time.

Now I'm not saying this motor is the answer to your question but if you need speed and torque then you really do need a servo, DC or AC.
The servos tend to be more money because they are so much more sophisticated in terms of electronics and what can be done with them.


I had a closer look at this one:

It is very similar to the one recommended by Jazz, bur it is a Stepperonline version. There is though one difference which immediately is obvious, that is the the Stepperonline version has no enable signal. It is not clear to me when the motor is enabled and when not.

It does, however, this must be done using the RS232 port. Not ideal I know but this one of the advanced features a Servo provides among others. Though I do think it's silly not providing a dedicated enable pin.


It is also not very clear to me if this is a 3-phase servo or not. The one in Dean's link looks like a 3-phase motor.

The link I gave wasn't a recommendation for that motor but more to show the type of motor, that is a DC Servo motor with integrated drive. After Muzzer's comment, I looked again and agree with him that it does read like it's a Stepper but says it's a DC Servo, these two things conflict with each other because while both uses DC current they are internally structured very differently, so I wouldn't trust the advert description on this.


But there is another question... that is about the torque. Both Dean's, and the one here have considerably less torque than 4Nm... so would a closed loop stepper with 4Nm be equal to these two...? I mean, high rpm and smooth, quiet running is one thing and is very nice, but if it is at the expense of less torque then I don't see the benefits. I need this motor to be able to hold the chuck when I use it as a rotary axis, not just rotating fast. So what's the secret? Can these servos be used as ordinary rotation axis steppers also, and can they hold the position when used on the CNC mill?

Well without knowing the torque you require then it's nearly impossible to answer, but my gut feeling is that you will need a little more than 0,6Nm and the answer to me is to use either an AC or DC servo with a higher Nm rating and with a brake.
This will give you both higher RPM you would like and the brake will hold the motor when at standstill. If this was an industrial 4th axis it would have a separate brake to lock the rotation.

Obviously, this would be more expensive than your stepper setup but it's the only way you will get higher RPM's and torque with reliable results without getting silly about it.
End of the day a typical 4th axis isn't a lathe, it's not designed for high rpm's but for positional indexing. If you want Lathe performance with accurate indexing then you will need to dig deep as it gets expensive very quickly to do it right.
(Look for Simpson36 on youtube if you want an idea of how complex it can get.)

A_Camera
14-06-2021, 07:03 PM
Thank you once again. I did watch those videos and seen his work (I am one of his followers) but... There is no way I want to compete with Simpson36 or anyone else. I am too old for competitions, just want to have fun and learn something new. My goal is just that. In the end, if I like it, or see a huge need of a real lathe then I'll buy one. To be honest, I don't think my machine will ever be able to be as good as a cheap 800-1000 USD cheap Chinese micro lathe, even though, mine might well end up being more expensive in the end.

Regarding the needed torque, honestly, I don't know what I need. All I know is that the current stepper is labelled 1.8Nm, but that's always holding torque. I don't know what the real torque is at 2280 rpm, which is what the rpm is now when the chuck runs at 380 rpm, but it is enough to drill in steel. This weekend I did some turning tests, which was a disaster but not mainly because of the stepper, but because of the extremely weak cross slide I was trying to use. Anyway, I could only turn soft plastic... :lemo:

I understand that your link was just an example, not a direct recommendation, but it seems like all 180W servo has about the same parameters. Had a look at JMC servos as well, and also the Stepperonline, but I don't think I want Stepperonline, due to that crazy RS232 enable solution. I don't know how they were thinking leaving out the EN signal and saving one opto coupler, I want a controllable enable/disable, and I don't regard RS232 is equal to that, since I can't use the same signal I am using for the other steppers, so Stepperonline is out of the picture.

The other thing about the torque is that I want to change the gearing from 6:1 to 3:1 or 2:1 (not decided yet which one) to get higher rpm, so if I do that I will lose torque but gain rpm. Anyway, if those servos are really that good, then I could test with 2:1 reduction, which could give me 1500 rpm on the chuck and about the same torque as I have now with the plain stepper, which should be fine to play with, so right now I am very much tempted to try that out. It's not too late yet to take a step up and use NEMA34, but I would prefer not to, because it would mean a huge change. It would also mean that it starts to become too large for my CNC, which means I might as well give up, which I don't want to do at this stage.

Anyway, it is fun to test this this out, even if in the end, I will not use it for anything serious, but of course, it would be more fun if I could use it for something as well.

JAZZCNC
14-06-2021, 09:51 PM
It's not too late yet to take a step up and use NEMA34, but I would prefer not to, because it would mean a huge change. It would also mean that it starts to become too large for my CNC, which means I might as well give up, which I don't want to do at this stage.

That would be an expensive mistake, large motors like Nema 34's require lots of voltage to spin fast and by lots I mean 150V+ to get any reasonable speed with torque.

The price of AC servos has come down to such a level that they nearly match High voltage drives which are required to get speed and torque from large steppers. So I would look at those before large steppers.

I have these 600w AC servos on my plasma machine and while only 1.9Nm they massively outperform the 4.5Nm closed-loop stepper motors. Which at just over $200 is a very good price, I would fit these and pay the extra over the JMC or any of those types of DC servo every time. (They also do them with a brake)

https://www.aliexpress.com/item/33046845943.html?spm=a2g0o.store_pc_groupList.8148 356.3.5be81468ni78om

Like I said before if you are going to go this far then do it right the first time because while digging deep hurts I always find it works out cheaper than experimenting and far less hassle.

Muzzer
14-06-2021, 11:10 PM
.....I don't know what the real torque is at 2280 rpm, which is what the rpm is now when the chuck runs at 380 rpm, but it is enough to drill in steel.


Sorry, I hadn't twigged that you were planning on using this motor for the SPINDLE drive. That's probably not the best idea if I have understood what you are planning. I'd seriously recommend you use a proper VFD rather than a stepper or servo drive, which are intended for position control, rather than speed control.

Forgive me if I've got the wrong end of the stick here. I have no desire to criticise so much as help you avoid disappointment. Perhaps Jazz or someone else will dive in here, in the interest of constructive advice.

JAZZCNC
15-06-2021, 08:18 PM
Sorry, I hadn't twigged that you were planning on using this motor for the SPINDLE drive. That's probably not the best idea if I have understood what you are planning.

No Muzzer, it's for a 4th axis.

A_Camera
15-06-2021, 11:08 PM
Sorry, I hadn't twigged that you were planning on using this motor for the SPINDLE drive. That's probably not the best idea if I have understood what you are planning. I'd seriously recommend you use a proper VFD rather than a stepper or servo drive, which are intended for position control, rather than speed control.

Forgive me if I've got the wrong end of the stick here. I have no desire to criticise so much as help you avoid disappointment. Perhaps Jazz or someone else will dive in here, in the interest of constructive advice.

You have misunderstood the thread. :sneakiness: I have a very good VFD and have no intentions to replace that. This is a 4th axis which I experiment with as a lathe / 4th axis combo.

A_Camera
16-06-2021, 10:47 AM
That would be an expensive mistake, large motors like Nema 34's require lots of voltage to spin fast and by lots I mean 150V+ to get any reasonable speed with torque.

I don't mean stepper, but servo.


The price of AC servos has come down to such a level that they nearly match High voltage drives which are required to get speed and torque from large steppers. So I would look at those before large steppers.

I have these 600w AC servos on my plasma machine and while only 1.9Nm they massively outperform the 4.5Nm closed-loop stepper motors. Which at just over $200 is a very good price, I would fit these and pay the extra over the JMC or any of those types of DC servo every time. (They also do them with a brake)

https://www.aliexpress.com/item/33046845943.html?spm=a2g0o.store_pc_groupList.8148 356.3.5be81468ni78om

Like I said before if you are going to go this far then do it right the first time because while digging deep hurts I always find it works out cheaper than experimenting and far less hassle.

Yes, but my problem is space, so I would like to avoid larger motors. All AC servos I have seen take up a lot of space, and even though the PSU is just a simple transformer, it does need a large one and that takes space, unless it is driven directly from the mains. At this stage, I'll avoid both for now. If need arise, I may change my view, or like I said before, buy a real lathe, which functionally will most probably be better than what I can build. To be honest, I think it would be more point in upgrading a real lathe with a large AC servo then doing it in my DIY "poor man's lathe" which will probably end up being pretty expensive in the end, I know that, but I want to try, learn and improve incrementally. I know that such approach costs money, but for me it is more fun this way. Starting a huge project to build a perfect and powerful lathe takes a long time and I don't think I have the patience to do that. Also, I am just an amateur "machinist", don't have any real need for this except as a hobby, so I am not at this stage interested in starting a huge project which will never be finished.

I feel that this hobby is like a black hole, it can swallow everything, including ones family life, fortune, space and time, so I must set some sort of realistic limit to my ambitions. I am well aware that this, once ready, will be far away from perfect, but that's OK for me, even if some parts will in the end not be used, or replaced shortly after a test.

Right now, I am aiming at some DC servos, but I think before that decision is made, I will build the X and Z out of real parts, which for me will be based on some HGR20 and HGR15 linear rails plus 1605 ball screws.

Muzzer
17-06-2021, 10:50 AM
I think there may be some confusion over terminology here. The Lichuan servos that Dean linked to are generally known as "AC servos" and don't need a separate psu - they take 220V mains directly. The term "DC servo" is not clearly defined but generally involve brushed motors ie an H bridge output and a 2 terminal motor. These aren't used much these days.

You might also talk about "brushless DC servos" (BLDC) which are pretty much the same as an "AC servo" but they drive the motors with a trapezoidal 3 phase voltage rather than the sinusoidal one seen in an AC servo.

The various open and closed loop steppers have 200 or more teeth per rev, whereas an AC servo is likely to have only a few poles per rev.

An AC servo motor will be more power dense than an equivalent AC induction motor (VFD etc) and also more power dense than a stepper. AC servo and AC induction will both create a fairly flay torque speed characteristic, whereas the torque from a stepper falls off quickly with speed. Gearing down a stepper "to increase the torque" may not actually gain you much / any torque in some cases due to that torque fall off, unless you are focusing on stall torque.

As Dean says, there's a lot to be said for going AC servo and those Lichuans are very keenly priced. There's no right or wrong here but that's just some input from my side. I develop motors and inverters in my day job and (sadly) use them from time to time in my workshop.

A_Camera
17-06-2021, 03:17 PM
OK, thanks. Anyway, there are AC servos which use less voltage than the mains, but in all cases they require a beefy transformer, a large driver and even the motors are larger. I have no space or need for that at this moment, even if it would be much better for the purpose. Remember that this is mostly just a test, a proof of a concept with a goal to build something to play and learn with.

I think the only difference between AC and DC servo is that an AC servo driver has the rectifier built into the driver, while a DC servo must be fed using an external DC PSU with a rectifier circuit in the PSU. When I say DC servo, I mean just that the driver needs DC current, but it is still a servo, and NOT a closed loop stepper. Basically, I have written off the use of closed loop stepper, mainly based on what Dean said before, except that I will not at this stage buy a large one, but aiming at a 180W type from JMC, namly this one:

https://www.aliexpress.com/item/32947958273.html?spm=a2g0s.8937460.0.0.79092e0eprg amN

I checked the manual and the software, and must say I am impressed considering readability, configurability and contents, and since this will be my first servo, I think that is important. It fits my size needs and I don't have to build a new PSU / control box. If in the end it turns out to be too weak, I can always move the motor to my CNC and replace a NEMA23 stepper of one axis, and buy a new and stronger servo with break as Dean suggested, but that will not be now. First I want to finish the project and see it working.

I think you are wrong about gearing. Torque is multiplied by the gearing, so indeed I gain a lot. Of course, since the stepper is pretty weak, it will not be much at high RPM, but never the less, it will be six times (minus some losses, but we can ignore that here). Of course, gearing down is giving the most torque benefits in holding, but there is a benefit even when rotating. Of course, that would not be needed at all if I had a large enough motor, but that's for later times.

Muzzer
17-06-2021, 06:06 PM
I think you are wrong about gearing. Torque is multiplied by the gearing, so indeed I gain a lot. Of course, since the stepper is pretty weak, it will not be much at high RPM, but never the less, it will be six times (minus some losses, but we can ignore that here). Of course, gearing down is giving the most torque benefits in holding, but there is a benefit even when rotating. Of course, that would not be needed at all if I had a large enough motor, but that's for later times.

You seem happy to redefine terminology, which is fine but may cause confusion when you try to communicate with others.

Have a look at the torque-speed curves for the stepper motor you are considering. If you fitted a 2:1 reduction gear you'd double the stall torque perhaps. But if the torque at a new, doubled operating speed had dropped to half of what it was, you'd have gained nothing - you'd now have the same final torque at the same speed but you would have halved your max speed (the all important "rapids"!). My conclusion is that you might as well go direct drive with steppers.

For a servo it's different - you really need to step the speed down by a factor of 2 or 3 to boost the final torque to something more generally useful. Also, not many ballscrews will be happy to run at the max speed of a servo which is usually around 3000rpm, give or take. So to me, steppers are best used in direct drive and servos best used with a belt drive reduction ratio of between 2 or 3. But as I said, there's no right or wrong!

Muzzer
17-06-2021, 06:19 PM
Here's a randomly chosen image from Oriental Motor, showing a typical torque speed characteristic for a stepper motor:
30083
https://www.orientalmotor.com/stepper-motors/technology/speed-torque-curves-for-stepper-motors.html

If you gear down the output by 2:1, you might move from eg 500rpm to 1000rpm and in the process you'd drop from a pullout torque (loss of position) of ~45oz-in to 27oz-in (forgive the Mercan units). After the supposed doubling up of torque through your expensive reduction stage, you'd actually end up with ~54oz-in. That's an increase of ~20%, not 100%. See what I mean? Some of the Chinesium steppers have even worse torque curves. Sure, you double the stall torque but unless you plan to use your machine as a press, that's not much use to anyone.

m_c
17-06-2021, 09:23 PM
Muzzer, did you look at the latest link A_Camera posted?

It's for a low DC voltage supplied integrated AC servo motor...

A_Camera
17-06-2021, 10:15 PM
Muzzer, try to understand something. I bought this 4th axis rotary chuck and it comes with a belt giving 1:6 reduction. It is NOT something which is a choice, the belt is a necessary thing in it, not an "expensive reduction gear" which I design. In fact, direct driving would be impossible because there is a hole through the rear, so unless you want to give that up, you MUST have a belt or some sort of arrangement to drive the chuck. Besides, if we take your own torque chart as example and my actual 1:6 gearing then you can see that at 2280rpm that motor should have approximately 10 oz-in (units make no difference in the calculation). Now because of the gearing, the torque is 6 x 10 = 60oz-in but the rpm is down to 380. So, yes, while the torque is not that much bigger, but still bigger, and is definitely 6 times the value read from the chart. Not much is gained compared to direct drive at low rpm, but never the less, there is a gain, even with your motor chart. I have also said that the gearing is mostly a help between zero and very low rpm, I am aware of that, but still, there is no way you can directly drive this chuck, except through putting in a 1:1 belt, but that would mean I lose even holding torque, which is the very reason these belt ratios are used the first place.

Now, to make something clear, this is the motor I am considering: https://www.jmc-motor.com/product/980.html
It says: "Nema 23 Integrated AC servo motor 180W" ...and it is not a printing error, because I am sure it is the same as any other AC servo, except that the driver is fed using maximum 50VDC, so the rectifier is outside the driver, as I said before. The motor has a 0.6Nm (85 oz-in) torque all the way to 3000rpm with a max torque of 1.5Nm. Now, by using the same 1:6 belt I'd get 3.6Nm holding torque AND all the way to my current maximum chuck speed, which is 380rpm. In other words, I gain at the speedy end, but lose on holding torque, since my current steppers have 1.8Nm x 6 = 10.8Nm, but my current stepper has 0.5Nm at 1500rpm, which is quite a bit from 2280rpm (my current speed). I have no torque table above 1500rpm. My guess is that the torque is very low at that high speed. With the servo I could go to 3000 rpm (or even more), so that would definitely be beneficial for the rotation, but not for holding torque.

Yes, I know that combining a 4th axis and use it as a lathe is not a good idea according to a professional, but to me, this is just for fun, I don't aim at competing with anyone except myself. I aim at getting experience, knowledge, having fun, enjoying my time, exercise a hobby and so on.

Correct me if I am wrong, but don't say that I "happy to redefine terminology", because I am not. I did not make any calculations before, but definitely not redefined anything and torque is definitely increasing when you add a reduction gear. I think you understand that very well, but if you don't understand what I am saying that's a different thing, and is perfectly possible that I am not that clear, but that does not justify such statement. I am not stupid, I can read charts, make calculations (in ANY sort of units) and so on. You are not the only one with engineering background and long and prosperous career.

m_c
17-06-2021, 11:59 PM
Correct me if I am wrong, but don't say that I "happy to redefine terminology", because I am not. I did not make any calculations before, but definitely not redefined anything and torque is definitely increasing when you add a reduction gear. I think you understand that very well, but if you don't understand what I am saying that's a different thing, and is perfectly possible that I am not that clear, but that does not justify such statement. I am not stupid, I can read charts, make calculations (in ANY sort of units) and so on. You are not the only one with engineering background and long and prosperous career.

Erm..you did though -
I think the only difference between AC and DC servo is that an AC servo driver has the rectifier built into the driver, while a DC servo must be fed using an external DC PSU with a rectifier circuit in the PSU. When I say DC servo, I mean just that the driver needs DC current, but it is still a servo, and NOT a closed loop stepper.

How any given motor controller gets powered, does not define the type of motor attached to said controller.
An AC motor doesn't suddenly become a DC motor because the controller needs DC power.

A_Camera
18-06-2021, 09:38 AM
Erm..you did though -

Where and in what way?

If you have a 1:6 gear you WILL get a 6 times increase of torque at ANY given rpm between zero and infinity. Period.
If you have a 1:6 gear you WILL get a 1/6 reduction of rpm at ANY given rpm higher than zero rpm. Period.

What is wrong in the above two sentences? If something was not clear in anything I said before then of course, a clarification should be asked for, but with an explanation of what is not clear, or what is wrong.

Of course, it should be clear and obvious to anyone that I don't mean the holding torque. Holding torque is only multiplied by six at zero rpm. For a servo if the torque is constant, the increase is 6 times all the way. Isn't all that true? I mean, I am an engineer (with more than forty years of experience and a few of my own inventions), but not on the mechanical side and I may have missed something, or didn't use the right expression or words, but that's not the same as I am a stupid idiot who is trying to redefine the shape of the earth or something. However, without explaining my mistakes nobody is helped and nobody is learning anything.

But OK, this is the Internet... :beer:


How any given motor controller gets powered, does not define the type of motor attached to said controller.
An AC motor doesn't suddenly become a DC motor because the controller needs DC power.

I know that, and as far as I remember, I made that clear before also. Some people talk about AC (powered) servo as if it was something special. The only special thing about them is that the rectifier is built in the driver, because otherwise internally, I think they are the same. But of course, I never had one, never opened a driver for a servo, so I don't know for sure, but I strongly suspect that is the case. The only advantage is when you have really high powered servos which can be fed directly from the wall mains. It makes wiring much simpler. Of course, it is simpler also if all you need is a transformer, but if you already have a fat PSU with a good enough voltage then there is no benefit in going to AC powered servo, unless you need to replace that PSU due to the size of the servo. In that case, of course it would be dumb to buy (or build) a new PSU when one can use the AC directly from the wall.

m_c
18-06-2021, 09:55 AM
Where and in what way?

If you have a 1:6 gear you WILL get a 6 times increase of torque at ANY given rpm between zero and infinity. Period.
If you have a 1:6 gear you WILL get a 1/6 reduction of rpm at ANY given rpm higher than zero rpm. Period.

What is wrong in the above two sentences? If something was not clear in anything I said before then of course, a clarification should be asked for, but with an explanation of what is not clear, or what is wrong.

Of course, it should be clear and obvious to anyone that I don't mean the holding torque. Holding torque is only multiplied by six at zero rpm. For a servo if the torque is constant, the increase is 6 times all the way. Isn't all that true? I mean, I am an engineer (with more than forty years of experience and a few of my own inventions), but not on the mechanical side and I may have missed something, or didn't use the right expression or words, but that's not the same as I am a stupid idiot who is trying to redefine the shape of the earth or something. However, without explaining my mistakes nobody is helped and nobody is learning anything.

But OK, this is the Internet... :beer:



I know that, and as far as I remember, I made that clear before also. Some people talk about AC (powered) servo as if it was something special. The only special thing about them is that the rectifier is built in the driver, because otherwise internally, I think they are the same. But of course, I never had one, never opened a driver for a servo, so I don't know for sure, but I strongly suspect that is the case. The only advantage is when you have really high powered servos which can be fed directly from the wall mains. It makes wiring much simpler. Of course, it is simpler also if all you need is a transformer, but if you already have a fat PSU with a good enough voltage then there is no benefit in going to AC powered servo, unless you need to replace that PSU due to the size of the servo. In that case, of course it would be dumb to buy (or build) a new PSU when one can use the AC directly from the wall.

And as per usual, you completely missed, or simply chose to ignore the bits Muzzer and I were correcting you on, with a big long rambling post.
I'm out. Have fun, interpreting things in whatever way you wish to chose, even if that interpretation is completely wrong.

A_Camera
18-06-2021, 12:05 PM
And as per usual, you completely missed, or simply chose to ignore the bits Muzzer and I were correcting you on, with a big long rambling post.
I'm out. Have fun, interpreting things in whatever way you wish to chose, even if that interpretation is completely wrong.

...and that is called constructive comment...? :hysterical:

Bye...

routercnc
18-06-2021, 07:46 PM
I have just skip read this post but the following video released by Clough42 as part of his lathe electronic leadscrew motor selection shows comparison testing of steppers and servos. Might be of some interest:

https://m.youtube.com/watch?v=et3oXoA2Hqo

A_Camera
27-07-2021, 06:19 PM
I have just skip read this post but the following video released by Clough42 as part of his lathe electronic leadscrew motor selection shows comparison testing of steppers and servos. Might be of some interest:

https://m.youtube.com/watch?v=et3oXoA2Hqo

Very interesting. Thank you. In the meantime, I bought a 180W servo motor, but to be honest, I am so far not impressed by that one. I feel that the holding torque is really poor, but maybe there are some parameters I missed and it can be configured better. However, this is version 552, and those have been replaced with version 6 which supposed to be better. I actually complained to the seller because he sold it as the "latest version" but it was obvious that it's not. I contacted the factory and got a reply that this version should not be sold overseas any more. Then I contacted the seller again and he offered me 100% refund, which was accepted. So now I have a free servo I can play with, but in the end, I may need to buy a 400W version if I want good holding torque also.