Machsupport Forum
Mach Discussion => General Mach Discussion => Topic started by: Kristin D on August 31, 2008, 08:28:34 PM
-
I can't really find any clear information on the effects of microstepping and torque, right now I have my board set for 8 microsteps (jumper selectable) just happened that way as I was careless reading the settings. I have been just getting set up but last night I had to mill off some fins on the heatsinks I was making and the machine groaned and jammed a couple of times. So I am wondering if I should pop another jumper or two and go to 1/4 step or 1/2 step or even full step mode? Seems from what I have been able to find microstepping will not improve resolution and may actually make it worse if it takes a few bumps to get the motor to move or switch direction.
Kristin
-
Kristin,
Yes, microstepping will reduce torque somewhat. But, if you're having problems with the motors stalling, your real problem is that either:
a) You're pushing the machine too hard or,
b) Your motors are too small or,
c) Your controllers are not very good, or are limiting current (Geckos are highly recommended) or,
d) You're not running high enough power supply voltage or,
e) Any combination of the above
What kind of machine is this? What size motors? What, if any reduction (including leadscrew pitch)? What stepper controllers are you using? What is your power supply voltage, storage, curring capability?
Regards,
Ray L.
-
Questions, questions...
It's a Taig mill, motors are 205 Oz unipolar, HobbyCNC Pro driver board (hey for $25 on eBay I couldn't pass it up), 20TPI leadscrews & adjustable nuts (new a couple of weeks ago). I think you hit it with supply voltage, think I am only getting about 20 VDC @ 10A out of the gutted Astron PS I am using, got to dig out another transformer and build another for 30-32 Volts I can run the board up to 42VDC but don't have a transformer on hand with 36V output.
I have adjusted the gibs, adjusted the leadscrew nuts and it's not mechanicaly binding anywhere, really not trying to run fast(6 IPM) or take more than .01/cut with a sharp USA made four flute 3/8"end mill. Could just have been the heatsink was pretty soft aluminum and was giving me fits?
Kristin
-
Kristin,
There are books written on the subject of microstepping. And if you want to be mathematicaly challenged read some of them. If i remember correctly microsteps only uses a small portion of the current available to hold position and max current is available at the 1/2 or full steps. But the controller, motor, wiring etc can greatly influence the torque at the microstep
location. I tested on some small steppers ( < 100 in oz ) using a torque gage and found that you could break the holding torque at a microstep position ( don't know what microstep position it was at) using less torque then at say a full step postion ( if thats where it was). When it skipped it will seek one of those postions and maybe it took 50-60% less torque as compared to breaking the hold at a major position.
Theory is interesting but on the practicaly end..........
I watched my 200 oz in steppers on the mill cut right thru a 1/2" hardened steel clamp at 1/4" deep and break the 1/4" carbide end mill at the end and they never missed a step! I was able to return to 0 location and could repeat if stupid enough to do so! ;D The engraving machine will split .001" easily into ten or 20 parts using 10 or 20 steps jogs, but then the resolution is coming from the axis mechanics. Frankly it's very hard and time consuming to attempt to measure inaccuracy of a microstep.
That said I'll stick with the microstepping / smoot running and tune and use the motors appropriately and leave the books on the shelf.
Also some of the testing which I tried.
;) RICH
-
Kristin,
Find where the steppers start to skip. Back acceleration down to 60 or 70%of where they started skipping. Make some test cuts in material and see if you left enough of a margin in the settings for cutting. Of course, like Ray noted,
the tuning and capabilites change with electrical supply available.
What you really want is power and that is achieved by the combination of speed and torque. If you had a motor curve
you could get in the ball park rather quickly. Even with or without a motor curve some testing will pay dividends.
RICH
-
Kristin,
Find where the steppers start to skip. Back acceleration down to 60 or 70%of where they started skipping. Make some test cuts in material and see if you left enough of a margin in the settings for cutting. Of course, like Ray noted,
the tuning and capabilites change with electrical supply available.
What you really want is power and that is achieved by the combination of speed and torque. If you had a motor curve
you could get in the ball park rather quickly. Even with or without a motor curve some testing will pay dividends.
RICH
Rich,
I think the problem is really my power supply plenty of amps but low by 1/3 to 1/2 the rated voltage. Or it could have just been the dead soft aluminum didn't like being cut, the other day I was cutting some acrylic with a 1/8 mill and it took the end off of the 1/8"x1" steel hold down strap without a cough or hicup proabably a .05 slice full depth!
Kristin
-
Questions, questions...
It's a Taig mill, motors are 205 Oz unipolar, HobbyCNC Pro driver board (hey for $25 on eBay I couldn't pass it up), 20TPI leadscrews & adjustable nuts (new a couple of weeks ago). I think you hit it with supply voltage, think I am only getting about 20 VDC @ 10A out of the gutted Astron PS I am using, got to dig out another transformer and build another for 30-32 Volts I can run the board up to 42VDC but don't have a transformer on hand with 36V output.
I have adjusted the gibs, adjusted the leadscrew nuts and it's not mechanicaly binding anywhere, really not trying to run fast(6 IPM) or take more than .01/cut with a sharp USA made four flute 3/8"end mill. Could just have been the heatsink was pretty soft aluminum and was giving me fits?
Kristin
Kristin,
OK, well you've got several things working against you. Unipolar motor controllers, and low supply voltage. The motors and drivers are probably OK for such a small machine, however. I'd beef up the power supply - get as close as you can to the 42V limit, and you'll see MUCH better performance. At 20V, your performance will be very poor.
You do have to be careful with soft aluminum, as it gets "gummy" and will really clog the cutter. Using a good lubricant will help a lot. WD-40 works pretty well. Also, don't go too high in RPM. You're far better off keeping RPM low, and increasing feedrate. Excessive RPM creates a lot of heat, which will tend to clog the cutter, as well as dulling it much faster.
Good luck, and let us know how you make out!
Regards,
Ray L.
-
MY NOTES ON DESIGN OF A POWER SUPPLY:
4 to 20x motor voltage ( GEICO DRIVE RECOMENDATIONS) [ of course stay within limits of your drive]
< 4X drive won't run smoothly
> 20X motor heating
OVERDRIVE RATIO ? - 25:1 power supply voltage to motor rating ( similar to 4 to 20 above )
POWER SUPPLY AMPERAGE: should be greater the all drive requirements
Others may want to commnet on the above, but believe they are good quidelines.
BTW, when cutting heat sink or other AL which was anodized, that surface is harder than you think.
RICH
-
Thanks Ray,
I did use WD-40 (actually Walmart spray oil or "SPROIL" as I am fond of calling it) but I think your right about the voltage, I have to tear the cover off of one of these wheelchair chargers I have here and check the transformer voltage, I recall something like near 30V output when I was charging a couple of batteries in series, I just don't remember if it had any internal filtering except the regulator board. That would give me a 25% boost in power right there and I can hunt down another transformer later.
Kristin
-
Kristin,
Put lotsa capacitance on whatever you use. I would think you'd want 10,000 uF at an absolute minimum, so you don't get voltage sag under heavy load.
Regards,
Ray L.
-
I have some big honking caps in the Astron and a couple of other saved junk electronics just have to make sure the voltage on them is enough, don't want them to go boom and fly around! ;D
Getting to late tonight to fool around, I'll report my findings tomorrow.
Kristin
-
Ray,
Been on a capacitor search, closest I have found is a 10,000Uf but only rated 30V BUT with 40V surge listed right next to that. So what does that mean? First time I have seen a surge voltage listed on one of these things. Looks like it will work with my 25VAC transformer from the wheel chair charger, that has two 25 Volt secondaries that seem to be wired in parallel through the charge control board. Any Ideas as to figuring phase on the secondaries, I was thinking just connect one lead from one to one from the other, one way (the wrong way) should cancel out to zero the other should show 50VAC series voltage.
From what I figure 25*1.414= 28.535VDC after the diode and cap and not accounting for the diode drop so the 30V cap should work with out providing me with the "big bang theory".
Kristin
Kristin,
Put lotsa capacitance on whatever you use. I would think you'd want 10,000 uF at an absolute minimum, so you don't get voltage sag under heavy load.
Regards,
Ray L.
-
Kristin,
The surge rating means it will tolerate that voltage for a *brief* period, not long-term. You'll need to find something with a higher voltage rating. While you're at it, better pick up a new battery for the ol' calculator. 25 * 1.414 = 35.35V , not 28.535V :-)
Your plan for sorting the phase on the secondaries is spot on.
Regards,
Ray L.
-
Ray, the calculator battery is fine :) it was me putting 1.1414 into it rather than 1.4141 :-[
Got it working short term, think I need a new battery for my cheap voltmeter too getting about 33VDC output with the cap connected, got to get the good meter out and find the new test leads I bought a few months ago.
I also got an answer on my LazyCam problem direct from Art Fenerty, seems zooming in before offsetting makes all the difference in the world! Two projects in one day is almost too much!
Kristin
-
Kristin, finding higher voltage lower capacitance might be easier and all you do is connect a few in parallel to get your required capacitance.
Hood
-
Kristin, finding higher voltage lower capacitance might be easier and all you do is connect a few in parallel to get your required capacitance.
Hood
Yup,
Know all about that but getting BIG caps here is tough, the 30V/40surge is working and has not proved the "big bang" theory yet! Now to find another 10,000 or so Uf for Free! ;D
Kristin