[ibvolman]

I am closing in on bringing my CNC router build to the finish line. That said, I am having a bit of an issue with motors stalling on my long axis, and on my Z. I abandoned the idea of using rack and pinion to drive my Y axis, and changed to twin 20mm ball screws, My stepper motors are Nema 34 878 oz inch. I thought they should be strong enough to handle the weight of the gantry and to lift the spindle. The X axis moves great. I can achieve a speed of 200 in/min without the motor bogging down, loosing steps, or just flat out stalling. I cannot say that about the Z axis and the Y axis motors. I can achieve only 75 in/min on both of these axis before the motors stall out, make a God awful noise, and fail to move the axis as I have commanded on the MDI or by continuous jog.

My gantry is very heavy, made of 0.750" aluminum, a huge extruded piece, and a hefty spindle. I am curious if my stepper motors are just not strong enough. I did find some Nema 34 1841 oz/in motors that are 5A per phase. I am curious for some insight from some more experienced CNC builders if it would be worth my investment to upgrade to the heavier torque motors, or should I just live with the slow movements? I have no idea what will happen when I finally get a router bit ripping through some hard wood. I can imagine that will only add to the load, and feed speed would need to be drastically decreased to handle it.

On a whim I thought that perhaps my linear rails mounted to the side of the extruded might not be the optimal thing, and I modified my gantry slightly to mount them to the top of the extruded rails along the Y axis. I really cannot tell that this made a difference and in hind sight my X axis is mounted along on the gantry beam much the same as the Y axis shown in the picture.

I am primarily building this CNC to manufacture some of my guitar necks and fingerboards, but as I am approaching retirement age I would also like to use it for other wood projects to generate some income. That said, I do not want to end up with 10 inches per minute feed rate.

[joeavaerage]

Hi,
what the bet that the steppers have plenty enough torque, but TOO high inductance and so the stepper starts missing steps or stalling even at low speeds.

Manufacturers make high torque steppers, and they catch the eye of unwary buyers......because they are powerful?......right??. The problem is they will have high inductance so that beyond a 100rpm or so
they lose so much torque that they crap out.


What is the pitch of the Z axis ballscrew? Do you have any belt or gear reduction between the Z axis stepper and the ballscrew?

What is the inductance of the stepper? No guesses, read it off the spec sheet.

All steppers lose torque the faster they go, thats plain physics and can't be avoided. Inductance is a good measure of how bad that degradation will be. The lower the better.

For 34 size steppers look for 2mH to 4mH, 2mH preferred and reject anything over 4mH.

The best steppers are very seldom the ones with the highest torque (and highest inductance to the point of being useless for CNC), but a moderate torque and LOW inductance.

The second thing you need to make steppers run at speed is high voltage. What voltage are your drivers and power supply? You should have at the very least 80VDC drivers and an 80VDC supply.
110VDC and 110VDC supply would be better.

Craig

[ibvolman]

Hi,
what the bet that the steppers have plenty enough torque, but TOO high inductance and so the stepper starts missing steps or stalling even at low speeds.

Manufacturers make high torque steppers, and they catch the eye of unwary buyers......because they are powerful?......right??. The problem is they will have high inductance so that beyond a 100rpm or so
they lose so much torque that they crap out.


What is the pitch of the Z axis ballscrew? Do you have any belt or gear reduction between the Z axis stepper and the ballscrew?

What is the inductance of the stepper? No guesses, read it off the spec sheet.

All steppers lose torque the faster they go, thats plain physics and can't be avoided. Inductance is a good measure of how bad that degradation will be. The lower the better.

For 34 size steppers look for 2mH to 4mH, 2mH preferred and reject anything over 4mH.

The best steppers are very seldom the ones with the highest torque (and highest inductance to the point of being useless for CNC), but a moderate torque and LOW inductance.

The second thing you need to make steppers run at speed is high voltage. What voltage are your drivers and power supply? You should have at the very least 80VDC drivers and an 80VDC supply.
110VDC and 110VDC supply would be better.

Craig

I have the Motionking 34HS9801 steppers. Phase inductance of these is 4.1 according to the spec sheet. I am using 5mm pitch ball screws on all axis all direct drive, no belts or gear reduction. My drivers are 80V rated. Unfortunately my power supply is 54V, and I am almost positive it isn't high enough watts after doing a bit of research. The closest to the power rating I have found on an 80VDC power supply is 1500W. Do you suspect that this is my issue in such slow rapids? These steppers are grunting at me when I set speeds over 75 in/min. I was excited to get a bit over 200 on the X axis. Cant figure why that one works so well, but the Y and Z have to be set slow to get smooth motion without stalling, grunting, squealing, and pretty much telling me NO...lol.

I do appreciate your time to answer my questions. This has been a challenge to build this machine. I have spent well over 5 years gathering parts and piecing things together. Unfortunately I purchased my steppers, drivers, and power supply several years ago without any direction. Live and learn I suppose.

[joeavaerage]

Hi,
with a 5mm pitch screw if you apply, or rather the stepper applies 1Nm of torque then the screw will exert 1250N of thrust. In short there should be any amount of torque for the mass of your machine.

My drivers are 80V rated. Unfortunately my power supply is 54V, and I am almost positive it isn't high enough watts after doing a bit of research.

54V is not really enough, but it should do better that you've got. Don't get concerned with the power. Sure it would be nice to have a high power supply as that ensures that the voltage remains high in those
brief moments of peak load, so I don't think a low power supply is the reason for the problem you describe.

Are the motors bipolar (4 wire) or unipolar (6 wire)? If you have unipolar motors but have mis-wired them the effective inductance is four times that of one coil alone, ie 16mH rather than 4mH. That could explain the issue
you describe.

Craig

[peteeng]

Hi Ibvolman - As Craig says potentially you have tonnes of force available to you for moving your machine. How are you driving the "twin" drives? do you have a separate driver for each motor? Is your X axis parallel to the gantry? I say this because you are having trouble with "twin" drives yet not the single drive. The single drive is doing its thing correctly. A 5mm pitch and 200"/min (5m/min and 1000rpm) is the top you should expect for this sort of set up. If you are driving two motors with one driver then that's your issue. Peter

[ibvolman]

Hi,
with a 5mm pitch screw if you apply, or rather the stepper applies 1Nm of torque then the screw will exert 1250N of thrust. In short there should be any amount of torque for the mass of your machine.



54V is not really enough, but it should do better that you've got. Don't get concerned with the power. Sure it would be nice to have a high power supply as that ensures that the voltage remains high in those
brief moments of peak load, so I don't think a low power supply is the reason for the problem you describe.

Are the motors bipolar (4 wire) or unipolar (6 wire)? If you have unipolar motors but have mis-wired them the effective inductance is four times that of one coil alone, ie 16mH rather than 4mH. That could explain the issue
you describe.

Craig

Craig, I am using 8 wire motors. I supplied the wire diagram for my particular motor. What I have is the yellow and blue wire tied together, and the orange and brown wires ties together. Then the red and black make up my A leads, and the white and green make up the B leads. Not sure if it is exactly correct, but that is the way I understood the diagram.

What insight could you provide to the way I should wire these? I appreciate your help.

[ibvolman]

Hi Ibvolman - As Craig says potentially you have tonnes of force available to you for moving your machine. How are you driving the "twin" drives? do you have a separate driver for each motor? Is your X axis parallel to the gantry? I say this because you are having trouble with "twin" drives yet not the single drive. The single drive is doing its thing correctly. A 5mm pitch and 200"/min (5m/min and 1000rpm) is the top you should expect for this sort of set up. If you are driving two motors with one driver then that's your issue. Peter

Hi Peter. I am driving each axis ball screw with its own motor. So so I have a 4 motor setup. Each motor has its own drivers, and the slaved drive has the exact same settings on the drivers. In fact, all 4 drivers are set the same on the dip switches. I have set my current on my drivers to the 3.71 amp selection as my motors are rated at 4 amps, and the next highest selection is 4,28 amps.

My gantry is square to the Y axis when I get ready to jog or input a MDI command to move the Y axis. It remains square at slower speeds, but as previously mentioned, it will not move properly with anything over 75 in/min. When one of the motors decides it is going to jam the axis visibly comes out of square until I let off of the jog and hit jog a few times. Then it comes back into square enough to move it back against my switches to square it up properly.

On my X axis I have it set at 200 in/min. For some reason I was able to get slightly over 245 in/min on that axis, but opted for the slower speed of 200 as my machine only gives me 24" in the X direction. Don't want to crash it too hard...lol. But that one axis works the way I envisioned all axis working. I have issues with the Y twin axis, and the Z axis as having to run on the slow side. I figured the Z axis is trying to lift the heavy spindle, so I chalked it up to that fact. However, when I experienced the same issue with the Y axis I figured something is not right with my stepper setup, or my motors do not have enough torque.

Thoughts?

[joeavaerage]

Hi,
well, there's your problem right there! You have wired those motors in Bipolar Series. If the inductance of any one coil is 4mH then the inductance of two coils in series is 16mH. (Inductance is proportional to the square
of the number of turns. Ergo with two coils in series and mutually coupled, the effect is one coil of double the turns, i.e. 4mH becomes 16mH)

Try this: Bipolar parallel

Red & Blue together making the A winding
Yellow & Black together making the /A winding

White & Brown together making the B winding
Orange & Green together making the /B winding

Not sure how the manufacturer intended for the current rating to be interpreted. I would guess 4A means 4A per winding....thus with two windings in parallel that would mean 8A.
That does seem very high though. I would in the first instance try 4A for the two windings combined. Try it an see if the motors get warm. Then try upping the current to 6A and observe the temperature
rise. The current limit is all about thermal load....and you can experiment to where you think the motors are hot enough. It's not uncommon for a stepper to run at 60C, although I'd
prefer they stay 40C or lower. I suspect once you have them wired correctly that you will need only a fraction of their full output to make your machine sing and could thus run them at well less than
rated current and still have plenty of boogie.

Even with Bipolar Parallel and an effective 2mH per winding I think about 1000rpm is all that could be expected with only 54V available. Were you to up that to 80V then you might reliably get
1250rpm maybe 1500rpm.

Note I'm a little unsure what happens to the effective inductance when two coils are joined in parallel. If they are NOT mutually coupled then the inductive impedance is the combination of two impedances in
parallel per normal, i.e. 4mH becomes 2mH. However, when the coils are mutually coupled the I suspect that the effective inductance goes as the square root of the two coils, i.e. 4mH becomes 2mH.
The net result is the same but the logic by which it is obtained differs. I really REALLY do not want to get my textbooks out.......mutual inductances always confused the sh*****tter out of me!

Craig

[ibvolman]

Hi,
well, there's your problem right there! You have wired those motors in Bipolar Series. If the inductance of any one coil is 4mH then the inductance of two coils in series is 16mH. (Inductance is proportional to the square
of the number of turns. Ergo with two coils in series and mutually coupled, the effect is one coil of double the turns, i.e. 4mH becomes 16mH)

Try this: Bipolar parallel

Red & Blue together making the A winding
Yellow & Black together making the /A winding

White & Brown together making the B winding
Orange & Green together making the /B winding


Even with Bipolar Parallel and an effective 2mH per winding I think about 1000rpm is all that could be expected with only 54V available. Were you to up that to 80V then you might reliably get
1250rpm maybe 1500rpm.

Craig

Let me make sure I understand correctly. The last thing I want is to do something stupid and let the magic smoke out of my motors. I provided a diagram of how I understand what you are saying concerning the wiring.

Oh, and I did go ahead and order an 80VDC 1500W power supply. I figure what is the worst that could happen.

[joeavaerage]

Hi,
yes, that is correct.

Just by wiring in this manner has improved the inductive time octant by 16-fold, and thus the top speed has increased by like amount.

Inductive time constant = resistance x inductance.

The two windings in parallel have half the resistance of one winding alone, and one quarter of two windings in series. The inductance of the parallel pair
is 2mH verses 16mH of the series pair. 1/4 x 1/4 =1/16.

It is the inductive time constant that determines the effective top speed of a variable reluctance motor (stepper).

Craig

[ibvolman]

Once again, thank you Craig. I will swap my wiring over to this manner this coming weekend and give it a try.