[WhiteHare]

Well, according to one blog,

"Generally, 12 [V] is the smallest voltage used to drive actuator motors, with higher voltages at 24 [V], 48 [V] and even 80[V] being used for motion control systems. A good rule of thumb is to use between 10 and 24 times the motor’s nameplate voltage for the system bus voltage."

Stepper Motor Voltage Explained – Blog

The nema 17's that I'm upgrading to have a nameplate voltage of 3.36 volts. So, according to the quoted rule of thumb, they should be driven somewhere in the range of 33.6v to 80.64v. That's quite a bit above the 12v that the woodpecker control board would limit it to! In fact, it's only just barely within the 36v upper-end of what an A4988 is rated to handle.

[NIC 77]

To give you an example, the project I am currently working on for my industrial 3d printer, I plan to use a Duet 2 or Duet 3 board running at 24V....

But that would not stop me from using industrial servos at 220VAC if I wanted to. I'd just need to check that the Duet could push the step/dir inputs at the correct voltage for the servo drivers.

I may end up using JMC servos at 36-48V. I have not decided what to do there.

I had a look at the Ustepper....those are starting to get pricey.

I have to wonder if your money is better spent on upgrading your linear components than on trying to push to 60V.

[NIC 77]

Well, according to one blog,
Stepper Motor Voltage Explained – Blog

The nema 17's that I'm upgrading to have a nameplate voltage of 3.36 volts. So, according to the quoted rule of thumb, they should be driven somewhere in the range of 33.6v to 80.64v. That's quite a bit above the 12v that the woodpecker control board would limit it to! In fact, it's only just barely within the 36v upper-end of what an A4988 is rated to handle.

Yeah, good idea to get away from 12V.

I would look up the spec sheet for the motor to figure out what the max Voltage is. Do you have a part number? You might overheat it. Plus if you changed to an 8mm lead screw, with an anti-backlash nut that might fix your problem better than trying to push to 60V.

You are talking about what the board is limited to. I am saying that this is not relevant to an external driver. Sounded like you were trying to push your entire board up to 60V?

IMO, just run it at 24V or 36V.

Like I said, I'm not an expert on the subject.

[WhiteHare]

I would look up the spec sheet for the motor to figure out what the max Voltage is. Do you have a part number?

Yes, it's 17HS24-2104S

I don't think there is such a thing as "max voltage" per se though. Maximum current is all that matters. Name plate voltage is just max current times the winding resistance. The driver only uses the extra voltage to overcome inductive reactance and get to the target current faster. Once the target current is achieved, the driver drops the voltage to something more sustainable that won't "burn out" the motor. The reason the motor doesn't "burn out" is that the driver never allows the current to exceed whatever maximum limit was set.

Paraphrasing what I've just recently read on other cnczone threads, it would be:
1. Torque is directly related to current. If you double the current, you double the torque.

2. Using a higher voltage means you can get higher acceleration. In particular, you can get to a target current level faster. Therefore, indirectly, higher voltage leads to higher (average) torque, because the faster you get to a higher current the more time the motor dwells at the desired torque, which means a higher average torque.

That seems consistent with your chart (below): double the voltage did lead to higher torque, but not double the torque.

[WhiteHare]

You are talking about what the board is limited to. I am saying that this is not relevant to an external driver.

Agreed. Most likely an external driver wouldn't have these limitations, but, then again, you can't really be sure without looking under the hood. Just as the Woodpecker board's silkscreen says "12v-36" for its supply voltage, and then behind your back buck converts it all to 12v anyway, an external driver could be doing the same thing. "Why would it?" you might ask. Well, for whatever reason (or lack thereof) that the Woodpecker is doing it. The Protoneer GRBL board (the el cheapo board I received yesterday) doesn't change the input voltage before connecting it to the stepsticks, and so it is proof that the Woodpecker didn't need to either.

[Al_The_Man]

Something missing seems to be the reason for the higher voltage when using steppers.
Stepper motors have a fixed operating voltage and current stated on the plate, and if you were to run them with a fixed voltage, then that plate voltage should not be exceeded as you would then exceed the current. With possible destruction.
But as you increase speed with a stepper, the inductive reactance increases, therefore limiting the current (torque), In order to overcome this drop in torque, a higher voltage is used for the drive, which in turn maintains the very important plate CURRENT, and should not exceed it.
IOW, from zero rpm to maximum operating RPM, the drive maintains the current as a constant.
Al.
.

[WhiteHare]

IMO, just run it at 24V or 36V.

Yes, most likely, since with stepsticks going a lot higher maybe requires more effort than its worth. So, if it turns out that the 24-45v range doesn't satisfy, then adopting some higher voltage external drivers would probably be the next easiest thing to try.

[WhiteHare]

Something missing seems to be the reason for the higher voltage when using steppers.
Stepper motors have a fixed operating voltage and current stated on the plate, and if you were to run them with a fixed voltage, then that plate voltage should not be exceeded as you would then exceed the current. With possible destruction.
But as you increase speed with a stepper, the inductive reactance increases, therefore limiting the current (torque), In order to overcome this drop in torque, a higher voltage is used for the drive, which in turn maintains the very important plate CURRENT, and should not exceed it.
IOW, from zero rpm to maximum operating RPM, the drive maintains the current as a constant.
Al.
.

Agreed. That's how I'm assuming it works.

So, based on that, do you agree that being able to drive from higher voltages is a good thing? I mean it may not be strictly necessary (lower voltages just mean arriving at your destination a bit more slowly), but in general it's a good thing and therefore preferable. Right? The only downside I've read is that if you go too high, then you may start to encounter resonance because of the "stiffer" response.

[Al_The_Man]

You start to encounter the effects of increased inductive reactance at a comparatively low rpm, so the necessity to maintain the rated current, comes in early in the required RPM range, so modern stepper drives all use a higher voltage, the drive manuf, should offer the required data.
Go to sites such as Gecko, you should find technical info there that explains the difference in drive settings.
Al.

[WhiteHare]

You start to encounter the effects of increased inductive reactance at a comparatively low rpm, so the necessity to maintain the rated current, comes in early in the required RPM range, so modern stepper drives all use a higher voltage, the drive manuf, should offer the required data.
Al.

Which data exactly? Here's the datasheet: https://www.omc-stepperonline.com/do...HS24-2104S.pdf

Is that a typical datasheet? Not much there. Certainly no guidance as to preferred application voltages that I can see. In fact, no mention of voltages at all.

The torque curve that's provided was generated at 24v, if that means anything: https://www.omc-stepperonline.com/do...rque_Curve.pdf That's about the only hint. Are we to assume that 24v is prescriptive? I'm guessing 24v half-step was chosen simply to have a common baseline to compare different steppers, and there's no significance beyond that.

[Al_The_Man]

That is just motor data, as per my post, I was referring to the manuf. specs of the drive that normally include able to set for a particular motor specs.
Al.

[WhiteHare]

That is just motor data, as per my post, I was referring to the manuf. specs of the drive that normally include able to set for a particular motor specs.
Al.

Thanks for clarifying. I misunderstood your earlier post because I normally understand drive and driver to be two different things. I'll take a closer look at the driver documentation.

Thanks for the pointer!

[Al_The_Man]

As to sizing a motor there is quite a few sites out there, e.g. https://www.orientalmotor.com/motor-sizing/index.html
Kollmorgen used to have a nice graphic sizing program, just plug in you mechanics and feedrate reqd and it spit out the motor size. had a feature which included motor-load inertia, which at one time was considered important.Had to be under 10:1.
I think Orientalmotor mention it.
Although I have worked on many steppers systems, I have never implemented them in a system, all my projects were servo's both DC & BLDC.
Al.

[WhiteHare]

I ordered one of these tramming tools:
https://www.cnczone.com/forums/attac...d=447310&stc=1
so that I can do accurate repeatability tests on the z-axis, and that is because I will need to be deadnuts accurate if I am to selectively CNC mill the solder mask off from a PCB without destroying the underlying solder pads. i'm not sure I will have enough headroom on the z-axis to fit this tool into place--or use it for that matter--so I'll just have to just see how that goes after it arrives. Also, with a tool such as this, I should be able to confirm whether or not the z-axis closed loop driver is performing as it should be.

The tool itself is meant for measuring perpendicularity of the spindle relative to the xy-table, but I just couldn't find any other kind of tool for taking a z-axis repeatability measurement.

[CitizenOfDreams]

I ordered one of these tramming tools:
https://www.cnczone.com/forums/attac...d=447310&stc=1
so that I can do accurate repeatability tests on the z-axis, and that is because I will need to be deadnuts accurate if I am to selectively CNC mill the solder mask off from a PCB without destroying the underlying solder pads.

I hate to rain on your bubble, but neither bed tramming nor closed loop steppers will get you closer to your goal. Your machine may be 100% flat, level and trammed, but the PCB itself never is. That's why even commercial prototyping machines use depth limiters and spring tools.

For a hobby environment, surface probing is probably the best method to mill PCBs. Once you develop the procedure, the results are quite good and quite repeatable - look at my photos in this thread if you haven't already.

[routalot]

I ordered one of these tramming tools:
https://www.cnczone.com/forums/attac...d=447310&stc=1
so that I can do accurate repeatability tests on the z-axis, and that is because I will need to be deadnuts accurate if I am to selectively CNC mill the solder mask off from a PCB without destroying the underlying solder pads. i'm not sure I will have enough headroom on the z-axis to fit this tool into place--or use it for that matter--so I'll just have to just see how that goes after it arrives. Also, with a tool such as this, I should be able to confirm whether or not the z-axis closed loop driver is performing as it should be.

The tool itself is meant for measuring perpendicularity of the spindle relative to the xy-table, but I just couldn't find any other kind of tool for taking a z-axis repeatability measurement.

Looks like a nice gadget but you don't have to have anything that complicated.I prefer a single dial gauge attached to an arm that goes in the collet.One gauge giving a constant reading through a full rotation is the target.As for repeatability,you just attach the gauge directly to a pin in the collet.Unless you have some backlash in your ballscrew and very sticky slides,why would you expect not to find repeatability?Gravity and a clear signal to the steppers pretty much guarantee repeatability.

[WhiteHare]

why would you expect not to find repeatability?

Because at the moment if I were to position an end-mill such that it just barely touches the top of a workpiece, zero the z-axis, then raise spindle 30mm, and then execute a "return to zero" at maximum speed back to zero (i.e. a so-called "rapid"), it will overshoot the zero point and crash down into the workpiece. The motion planning should have executed enough deceleration to prevent the overshoot but still return the endmill to the zeroed position., but it didn't. I need to find out why and fix it. Right now it is obvious and I don't need a gauge to tell me that it's overshooting, but to lock in a permanent solution, I'll also need to make sure it isn't overshooting or undershooting by just a hair, which would maybe (?) add up to a more noticeable cumulative error when summed over many similar movements.

[WhiteHare]

I hate to rain on your bubble, but neither bed tramming nor closed loop steppers will get you closer to your goal. Your machine may be 100% flat, level and trammed, but the PCB itself never is. That's why even commercial prototyping machines use depth limiters and spring tools.

For a hobby environment, surface probing is probably the best method to mill PCBs. Once you develop the procedure, the results are quite good and quite repeatable - look at my photos in this thread if you haven't already.

Actually, I think we are in "violent agreement" on the points you raised. However, I have at least one thing (see post directly below) that I need to fix first. Without that in place, I don't see how I can be confident of probing out an accurate contour map (i.e. auto bed-level a PCB) or be confident of the z-axis to accurately follow it during the trace isolation milling. Yes, within certain bounds the spring-bit may make the solder mask removal less accuracy dependent, but although I received the bit, I haven't yet had a chance to test it, so meanwhile I'm covering all the possibilities as best I can. In any case, I don't see having an accurate z-axis as being in conflict with using a spring-bit for that purpose. At worst it may be overkill for the purpose of solder-mask removal, but it still supports that objective, not conflicts with it.

[CitizenOfDreams]

Because at the moment if I were to position an end-mill such that it just barely touches the top of a workpiece, zero the z-axis, then raise spindle 30mm, and then execute a "return to zero" at maximum speed back to zero (i.e. a so-called "rapid"), it will overshoot the zero point and crash down into the workpiece. The motion planning should have executed enough deceleration to prevent the overshoot but still return the endmill to the zeroed position., but it didn't.

If we are talking about steppers, any overshoot you see must be purely mechanical (inertia of the axis). The motion planner does decelerate, but it has no way to know how much deceleration is "enough" - you have to tell it (in the controller settings) what rate to use.

Anyway, you don't normally rapid all the way to the part. I don't think that would end well even if you had a perfectly tuned machine.

[CitizenOfDreams]

Question is - do you really have a cumulative error (i. e. lost steps), or just backlash? If you move the Z axis up and down randomly for a few minutes, does it return to the same zero point or does it shift?