Matching stepper driver output (V) to stepper rating

[macona]

Bart,

In an earlier post in this thread you mentioned micro-stepping capability might allow Mach-3 to deliver a decent engraving capability. How much a a micro-step do you think would be needed? 1/8, 1/16, more?

- Leon

Microstepping is not a replacement for true mechanical resolution. If you want higher res belt it down. What microstepping does do is make the machine run smoother. The higher microstep rate the smoother motion as it more closely resembles a sine wave. The disadvantage of microstepping is a pretty heavy toll on the torque of the motor.

[bdring]

Macona is correct in general stepping terms, but if your machine still runs fine in fine microstep mode, it is a good fix for the Mach3 issue. You really never need better than 1000 dpi resolution, so leave it in that mode for general cutting so you have high rapid moves. You can then flip a few dip switches and let Mach3 have higher resolution. It sure beats swapping pulleys. If you have a heavy machine, fine microstepping might not be an option.

Mach3 puts out a fixed, short pulse per step. At lower lower resolution, like 1000 dpi, this just is not enough power density. More pulse per inch means more laser power per inch. Note: I think you need to leave the pulse engine at 25MHz for Mach3 engrave mode.

[LeonS]

I understand both macona's explanation and Bart's explanation of why, in this particular instance, it does make a difference due to the controller logic.

Thanks for the info!
- Leon

[LeonS]

@r691175002 (boy that rolls right off the tongue doesn't it?)

Thanks for clarifying your status on the RetinaEngrave card. Bummer about the flash failing. I take it that a utility would normally update the firmware through the USB connect like an Arduino, right? I hope things go well when you get your controller back. I'll be very interested in your experiences. Has FSE been supportive by providing special versions of their firmware?

@Mike-f

Thanks for the testimonial and specs on using Mach3 for engraving. Great photos! The macro shots are awesome to look at. It really makes one appreciate the impressive technology we are playing with.

Cheers! and thanks again for the responses,
Leon

[evokanivo]

I hope someone can let me know whether the following combination of PSU voltage, motor ratings, and driver ratings will work. I am mostly concerned about the small nema 8 and want to check whether using a current set/limit resistor with a G251X is all that's required.

I have a 48V psu and am connecting up to 6 steppers:
-Two nema 23 steppers (rated 3.5V, 3.50 A, 3 mH), via G251X's with 3.5k current set resistor
-Two nema 8 steppers (rated 3.9 V, 0.60 A, 1.7 mH), via G251X's with 0.5k current set resistor
-Two nema 11 steppers (rated 3.8V, 0.67 A, 4.2 mH), via G251X's with 0.6k current set resistor

nema 8: http://www.pololu.com/catalog/product/1204
nema 11: http://www.pololu.com/catalog/product/1205

From the geckodrive manual:

32 * (√mH inductance) = Power Supply Voltage
If your motor has 2mH of inductance, the equation would look as follows.
32 * (√2) = 45.12V
In the example above 45.12V is the maximum voltage for that specific motor; any voltage above this will create unnecessary heating and any voltage below that will get you proportionally less speed than you could otherwise get at 45.12V.

This concerns me because the nema 8 is 1.7 mH, so 41 V is the max allowable PSU voltage. Will using a current set resistor take care of that or do I have to get another PSU? If so, I was thinking of just using a step-down switching regulator like this:
http://www.dimensionengineering.com/SWADJHV.htm

[frob]

From the geckodrive manual:

32 * (√2) = 45.12V

Ive no experience with these but lots in electronics so for what its worth,
i believe the rule above is meant as a rule of thumb, its not meant to be a precise threshold.
You have less than 3V difference (6%) in the result from your actual PSU voltage so i'd think its close enough.
Also your cables to the motors add some inductance so the total inductance might could get it right on the money as well.
If you want to compensate to get it right on, the easiest and cheapest way would be to add a pair of inductors in series with the motor windings.

[evokanivo]

Thanks for the response.

The root(2) is from the geckodrive example - the nema 8's are root(1.7)*32 = 41 V, so it's 7 off. Seems a bit much, no?

When you say add some inductors - how about something like this: http://search.digikey.com/us/en/product ... -ND/774814

And just out of curiosity, would it be equivalent to use the following if I were to omit connecting pins 2 and 3?
http://www.murata-ps.com/data/magnetics/kmp_5100.pdf

[frob]

The root(2) is from the geckodrive example - the nema 8's are root(1.7)*32 = 41 V, so it's 7 off. Seems a bit much, no?
When you say add some inductors - how about something like this: http://search.digikey.com/us/en/product ... -ND/774814
And just out of curiosity, would it be equivalent to use the following if I were to omit connecting pins 2 and 3?
http://www.murata-ps.com/data/magnetics/kmp_5100.pdf

First thing i'd do is try it as is and see if it works as is.
You're not driving much power at all so a little extra inefficiency shouldn't even come close to causing issues.
A good rule i try (though often fail) to live by is, if it ain't broke, don't fix it.

If you add inductors, its not a demanding use for them at all so just about any one will work, as long as you don't exceed their current ratings ( get at least double the motor current to account for ripple currents).
A shielded inductor is always better - especially when there's 2 in proximity and it avoids any stray EMI from them or influence from nearby metals or magnetic fields.

Just for fun i just did a simple empirical test to see if its worth trying to account for motor lead inductance - I grabbed a ~ 9' length of 20 awg zip cord speaker wire from my junk bin, shorted the ends and measured the inductance with an LCR bridge - hard to get an accurate reading at such low inductance but seems to be about 0.1 uH or less. (meter has .1uH resolution) So probably not. That's because the individual conductors are tightly paired so their mutual inductance largely cancels out. To show that, i took the same wire but this time shorted both ends and laid it in a rough loop on the floor, and measured 1.8uH.
If i now take a short piece of 22 awg wire and loop it 10 times through a 1" powdered iron toroid that happened to be on my desk, i get about 4.5uH. 2 turns produces .4uH, 3 gets .9uH.
Your common mode choke could work as one inductor, and is self shielding. in general those ferrite chokes are designed to be lossy and absorb high -frequency energy, though the .5uH one doesn't really do much that way until you get to several hundred kHz so it shouldn't be a real factor. it probably only has 2-5 turns of wire so you could easily make your own with any handy toroid.

[evokanivo]

Sounds good. I will just try it out and see if the motors get hot. Aside, your tests yielded a few uH, but I need several hundred uH, so I don't think I could reasonably make my own. But it's an interesting data point about the inductance of a few turns of wire.