HELP! Losing Steps

[bassman2914]

for the past 10 hours today and some time yesterday i have been testing my stepper motors since i finally found a computer with a working parallel port. When testing my motor which are 2.4amp 1.8 degree 160ish oz-in steppers being run on an atx 12v power supply i lose a ton of steps!!(on average i'd say about 2-3 per second) It seems to be completely random and it's extremely frustrating. I've testing every setting including but not limited to kernel speed, acceleration, pulse length, micro-stepping, rpm's(somtimes it's worse at high speeds but it's usually worse at low speeds, around 60-100rpm), buffer speed, the computer optimization stuff on mach's website, compatibility modes, parallel card drivers, every mach version i could find and more. The things i haven't been able to test is a different computer (i have one but not a parallel card) and a higher voltage power supply. (i fried it before i could test it, black should never touch red ) the one thing i have noticed is i can get higher speeds and less jerking from the motor if i hold it really tight as to not allow any lost energy in the motor to be used to twist it. I have a friend with a mill that i could test my stuff with but that would tell me nothing since it would be a different computer, different settings, different driver and a different power supply. that's too many variables to get any conclusive answer and he wouldn't have the patience to test one at a time(neither would I). My guess is the extremely low voltage is not allowing the coils to lose/build their charge fast enough causing random jerks and lost steps(that was something i just thought of and has absolutely no reason behind it besides the fact that it sounds possible but it's what i'm going with/hoping. ) The only way i've been able to successfully keep all the steps it by going at a speed that is essentially one step at a time. If anyone has any clue why this may be happening please help.

[ger21]

the one thing i have noticed is i can get higher speeds and less jerking from the motor if i hold it really tight as to not allow any lost energy in the motor to be used to twist it

Sounds like resonance. What drives are you using? What microstepping?

[bassman2914]

I'm using a Chinese TB6560 3 Axis Driver Board (CNC TB6560 3 Axis Stepper Motor Driver Controller Board | eBay) I've been using single or half step but i don't think the problem went away when i use even 1/8 or 1/16 but i can do some tests. I don't think I can afford the power loss from micro stepping but i'll try.

[bassman2914]

I never thought of it being resonance. I think you're right because it operates much better on half step than full step but the controller sucks when it comes to 1/8 or 1/16 so i won't use those. On full step my steppers do exactly what's in this video ([nomedia="http://www.youtube.com/watch?v=ME5udDx_560&feature=related"]YouTube - ‪G203V Z axis Bad Movement‬‏[/nomedia]) at 300rpm. On half step I can get to 400rpm before it does something like that. Is there any way to dampen the resonance more so I can get to 600rpm(60ipm) or with the motor being hooked up to the lead screw help. Thank You

[ger21]

http://www.cnczone.com/forums/steppe...tml#post256639

But I doubt you'll get 600rpm on only 12V.

[bassman2914]

YEAH!! Tried making my own dampers(didn't look through that thread though). I looked around for some aluminum but saw something even better. I just took a piece of wood about 1"x4"x.75" and drilled a slightly smaller than .25" hole in the center(must be centered!!) and just pushed it onto the shaft. On half step mode before damper i could get 40ipm(400rpm) and with the damper i got up to 180ipm(1800rpm). On single step mode before damper i got 25imp(250rpm) and after I got 300ipm!!!(3000rpm), that's a 1100% increase, a piece of wood!!! That's on 12volts too so when i get it running at 35v i should have all the power i need at 60ipm. Still have really bad resonance at 6ipm on single step so i have to accelerate fast to avoid losing steps(but i'm not going to use single step so it doesn't matter much). One problem though, my motors are all single shaft so for the tests i put the damper on the main shaft. Is there any way to extend the shaft? I could make a funky looking motor mount that would allow the damper to be in the front but that'd be a last resort. I'm looking to get new motors around 260 oz-in (mine are 160 right now) but those also don't have dual shafts. does anyone know where i can get steppers with dual shafts besides the 425oz-in ones on ebay which are expensive/overkill/will not help max cutting speed. P.S.-my machine is going to have a cutting area of 15"x15"x4" and it's going to be all aluminum(besides the spoil board) and the gantry will weigh around 60lbs(it's gonna be a beast made for cutting aluminum). do you guys think with these motor sizes i can get 60ipm in aluminum with a linear cutting force of around 10lbs. Thank You

[doorknob]

A lot of the motors on this page are available with dual shaft:

Stepper Motor

BTW, even though the driver board says that it will handle up to 36 volts, it might be a good idea to leave some "breathing room" instead of getting too close to that value (to avoid letting the smoke out of the TB6560s)...

[bassman2914]

Yeah good idea. How much breathing room do you think is good. I was looking a one supply that was 32-40volts and one that was 34-38v. The 34-38 is about 2 weeks faster shipping but if 34volts isn't enough breaking room i'll go with the other one. Thanks for the link. Their a little pricier than ebay but they have dual shafts, i know they're high quality, they have all the specs and i don't have to wait a month for shipping.

[doorknob]

I don't have a good basis for making an exact recommendation of how much "breathing room" to allow. In my own case, I am not planning to apply more than 24 volts to my board (well, maybe I'd go as high as 30), but YMMV.

The datasheet that I have seen for the TB6560AHQ IC shows an absolute maximum voltage rating of 40 volts. Maybe the board's rating of 36 volts max was intended to provide an adequate buffer zone, but I have seen so many reports of burned out TB6560's that IMHO caution is in order.

It is possible that people are burning out their TB6560s due to other reasons - for example, disconnecting one or more motor wires while the driver board has power applied to it is a good way to burn out almost any motor driver. Maybe some people are exceeding the allowable current limit instead of the max voltage rating. Maybe some of the TB6560 chips are substandard, or black-market clones. Who knows? But whatever the cause, a lot of people are burning up a lot of TB6560s.

It might be interesting to monitor the voltage applied to the TB6560 with an oscilloscope while motors are being driven, to see what amplitude of voltage spikes might show up there, and use that as guidance for deciding how much "breathing room" to allow...

[James Newton]

? "power loss from microstepping"? where did that come from? Microstepping doesn't cause power loss... unless it isn't being done right. I'm curious where anyone got that idea?

[doorknob]

? "power loss from microstepping"? where did that come from? Microstepping doesn't cause power loss... unless it isn't being done right. I'm curious where anyone got that idea?

Like everything else it seems, the answer is "it depends".

Power is related to torque. At some speeds, microstepping will be worse than full stepping in the torque department, while at other speeds, microstepping will be better (also depending on the details of the drive, just to complicate things).

For example, see the discussion at What effect does microstepping have on torque?.

[James Newton]

Yeah, no... if you read through that thread, the only time microstepping is worse is when it isn't turned off at high speeds, which top end microstepping drivers will do automatically. And it's a pretty minor loss even if they don't: the area difference between a sign and a square wave isn't much and the part where the rotor is most distant from the coil is further reduced by the square of the distance (I believe). Also, my experience, which Mariss doesn't confirm, is that there is less med band resonance with microstepping.... at least with linear microstepping.

But it certainly does depend on the drive as you say.

I'm just bothered by blanket statements against microstepping when it's clearly /better/ at lower and medium speeds.