PDA

View Full Version : Stepper motor question.



DICKEYBIRD
07-29-2012, 11:12 AM
I'm trying to figure out what's wrong with the X-axis on the ORAC CNC lathe retrofit. The Z axis has the normal smooth whine I'm used to hearing but the X has a lower pitch, somewhat "groan-ey" sound. I timed the 2 axes over the same distance/feed rate with a stop watch and they are precisely the same.

The velocity is only set to 20 IPM in Mach right now. If I speed it up to 30 IPM, the Z axis still moves and sounds very smooth; it just moves much faster. The X axis stutters, groans & loses steps even with no cutting load applied. The 2 stepper motors are identical; the Z axis has a 5mm pitch ballscrew, the X is 2.5 mm but the timing belt ratio is doubled on the X. The steps per in Mach is set the same for both. You'd think both would work pretty close to the same.

I've been having trouble with the speed control function in the CNC4PC C11 B.O.B. and thought maybe I'd screwed up one of the I.C.'s on it or maybe there's something wrong with the SmartStep/3 driver board. To test, I exchanged the X/Y B.O.B. outputs and the problem stayed the same so electronics shouldn't be the problem.

The wiring is 18g shielded cable, almost 8 ft on the X, 5 on the Z. I've tried it both ways, shield grounded on 1 end only and grounded on both ends. No difference either way. These are 4 wire Zylotex 23H276-28-4B motors.

What do ya'll think? Wire too long; needs a damper unit; stepper motor bad??

dalee100
07-29-2012, 11:55 AM
Hi,

Doubtful that the stepper is bad. They are pretty bullet-proof as a rule. Very generally speaking, when I see dropped steps it's either because the accel/decel are set too high, (and just because they are identical motors, doesn't mean they will behave identically), or there is binding some where on the offending axis. The difficulty with locating the binding is that it can be intermittent.

dalee

philbur
07-29-2012, 12:04 PM
Switching steppers should verify that alternative.

Phil:)

DICKEYBIRD
07-29-2012, 12:13 PM
Thanks Dale. I've tried various lower accel/decel settings and all it does is slow down. I noticed that if I grab the pulley, I can definitey feel a distinct vibration in the X axis at all speeds whereas the Z axis is smooth.

I whipped up a 1 3/4 x 5/8" steel fllywheel and mounted it on the other end of the double ended motor and it got worse so maybe it's a resonance problem?? I thought resonance occurred at particular speeds only...not across the range??

Evan
07-29-2012, 12:19 PM
Check that the step signal is occurring on the same edge as the other steppers in the system. In Mach "Ports and Pins" setup you are able to set whether the signal is recognized on the rising or falling edge of a change in state. It sounds like the X step signal is set to the wrong edge. Make it the same as the others.

If set to the wrong edge it will lose steps any time the direction changes as the direction signal will overlap the step signal.

Also, in the "Motor Tuning" section check the acceleration value and turn it down a bit. That is axis dependent since the amount of angular momentum of the entire drive for a particular axis depends on the amount of mass and its distance from the normal of that drive. Larger pulleys for instance have larger angular momentum and require slower acceleration. The acceleration model in Mach is a very simple linear increase in step rate from zero to maximum and isn't nearly as capable of smooth acceleration as systems that provide S curve acceleration. That is unfortunate. I used S curve acceleration when I wrote stepper drivers for a plotter back in the 80s.

Evan
07-29-2012, 12:23 PM
We overlapped posts. From what you say it sounds like a driver is weak or you have a poor connection on one of the coils.

Resonance is also definite possibility but a simple flywheel isn't an effective fix although it can sometimes work. It will require turning down the acceleration a lot though.

I will look up an example of a "real" resonance damper.

Evan
07-29-2012, 12:29 PM
This is an example of a resonance damper. It has a flywheel inside mounted on a bearing and immersed in a viscous fluid. The outer shell is light weight and is fixed to the stepper shaft. The flywheel is free to move and connected radially only by the viscous liquid. It has only a slight effect on acceleration capability.

http://ixian.ca/pics9/stepperdamper.jpg

DICKEYBIRD
07-29-2012, 12:31 PM
Switching steppers should verify that alternative.

Phil:)Unfortunately these motors don't have built-in terminal boxes and that'll be a last resort. I hate splicing wires.

Evan
07-29-2012, 12:35 PM
What is the actual driver you are using?

DICKEYBIRD
07-29-2012, 12:48 PM
Thanks for your input Evan. I checked the step/dir low/active settings and they're the same. I switched the wires to drive the X stepper from the Z outputs on the driver and the problem is exactly the same. I'm going to compare the motor coil resistances from the end of the cables and see if there's much difference there. I'll report back in a bit.

ps: Man I really hated to hear about your health issues and wish you all the best my friend. You are definitely one of the good guys!:D

DICKEYBIRD
07-29-2012, 12:54 PM
What is the actual driver you are using?Oops, missed your post. It's the same type I have in the MicroMill; an old Smartstep/3 from Optimised Control in England. It's half-step, 2A per phase @ 40V

Evan
07-29-2012, 01:21 PM
It's clearly not the driver then. If it worked before and doesn't work now the problem is in the X axis somewhere. Take the stepper motor off the axis and put a pulley or flywheel on it and try it then. If it still has a problem then it is either the motor or the wiring and the wiring is the first bet. First, if there is any excess wiring length and it is coiled to keep it neat then recoil it in a figure eight. That will eliminate any inductance in the coiled part which may be causing ringing. Next, put a .1 ohm resistor in series with each of the motor coils. This will reduce any ringing caused by the inductance of the wiring. If the problem goes away either leave the resistors in or then separate the wiring and twist each coil pair together to make it a twisted pair. The resistors need to able handle the wattage. I really doubt that is the problem though.

If the stepper works fine off the axis then you have either some resistance or friction or you have a loose mechanical connection in the system.

A loose mechanical connection will cause missed steps.

DICKEYBIRD
07-29-2012, 01:22 PM
Well crap, looks like I'm gonna have to strip out all my carefully routed and sheathed X stepper wiring. The Z motor/wiring ohms out totally different from the X when measured at the driver end. I'm pretty confident in the splices I've made so it may be frayed wires inside the motor. Man I hate doing work twice!

Evan
07-29-2012, 01:30 PM
Don't bother trying to repair the motor. Taking apart a stepper motor destroys the magnets.

John Stevenson
07-29-2012, 02:13 PM
+1 What Evan says, also these old round motors that Denford fitted were state of the art in 1870 but time have moved on, get some square 350 oz in motors on there as replacements.

Every Denford I have converted we have either had to replace motors or drop the acceleration to a point where it's borderline usable.

philbur
07-29-2012, 02:31 PM
Not wishing to flog a dead horse but what about switching wires at the driver?

Phil:)


Unfortunately these motors don't have built-in terminal boxes and that'll be a last resort. I hate splicing wires.

DICKEYBIRD
07-29-2012, 02:33 PM
No, these aren't the ol' round Denford motors, they're some nice square hybrid ones that nice Mr. Sparky_NY sent me.

Turns out the motor's fine; I just found the problem. Good thing I didn't bet money on my splice work. I found a spot where I guess I overheated the heat shrink and one little sharp edge poked through & shorted across 2 wires.

I would've never thought the motor would even work at all with a winding shorted out. X motor now runs smooth as silk.

I'll say it for you John: "Clumsy bastard!":D

DICKEYBIRD
07-29-2012, 02:37 PM
Not wishing to flog a dead horse but what about switching wires at the driver?

Phil:)Thanks Phil, I tried that back on post #10 and then started testing the wiring and found the problem: ME!:D

J Tiers
07-29-2012, 04:42 PM
Turns out the motor's fine; I just found the problem. Good thing I didn't bet money on my splice work. I found a spot where I guess I overheated the heat shrink and one little sharp edge poked through & shorted across 2 wires.

I would've never thought the motor would even work at all with a winding shorted out. X motor now runs smooth as silk.

I'll say it for you John: "Clumsy bastard!":D

I hadn't read this thread, but when I did the FIRST thing I thought was "there's a short or low resistance".....

A stepper driver has some things in common with other "digital" power devices..... So..... when you hear one operating "slowly", giving odd low frequency noises, LOOK FOR SHORTS FIRST.

The slow stepping may simply be the current limit operating.... the current limit usually drops output for a short time, short in human terms, but longer than one or two steps. It may sound like squealing or as in your case, groaning.

We have an old saying "if you can HEAR an SMPS or PWM, it's probably in current limit". Not quite so true with stepper drivers, but still is worth remembering.

Evan
07-29-2012, 08:18 PM
Because Milton is running a half step driver the expected result of a short would be a complete failure to rotate. It must have been an intermittent bad connection probably related directly to vibration of the stepper itself. Half stepping makes a lot of vibration.

J Tiers
07-29-2012, 08:48 PM
Point is that the NOISE described is one which (to me, at least) screams "current limit may be operating!"....... There can be several possible causes for that.

the character of noise made is a very useful diagnostic point.

DICKEYBIRD
07-30-2012, 09:32 AM
Yeah, it was wierd. With a 4 wire motor, me & my limited electronics knowledge would have bet money that there were no 2 wires that could be shorted together that would allow the motor to work at all. It ran in the correct direction at the same feed rate as the good one; it just had less power. It's been hooked up and working for over a week and although I noticed the different sound it was making, I was about to shrug it off as just a difference in the mount/drive belt system. The fault showed up dramatically though when I increased the speed.

The resistance on the "good" motor wiring was Black to Green: 1.3 ohms, Red to Blue 1.3 ohms, open on all other combinations.

On the "bad" motor it was Black to Green: 1.3 ohms, Red to Blue 1.3 ohms, Red to Green 1.4 ohms, Black to Red .2 ohms....I stopped there after realizing somebody must've screwed the pooch on the splice job.;)

Evan
07-30-2012, 12:31 PM
Point is that the NOISE described is one which (to me, at least) screams "current limit may be operating!"....

With chopper stepper drivers current limiting is always operating. There is often a pot which allows adjustment of the current limit. Current is sensed across a low value resistor and used to regulate the maximum current of the driver. During non rotation the current is dropped to about 1/3 of maximum and left on one phase to hold the stepper in place. If the phase was shorted at that time the driver would either cut out due to overheat sensing or just plain fail. The shorting must have been happening in synchrony with the stepping which would explain why it was still rotating and missing steps regularly.

Steppers don't normally miss steps when the system friction is too high. They simply stop rotating and make a lot of noise. They won't restart until powered down because they need an acceleration profile to get up to speed. Loose drive components will cause missed steps as will anything that periodically causes an electrical change in the driver output circuit. That includes bad connections affected by vibration and inductive spiking in the drive output circuit.

outback
08-08-2012, 07:43 AM
Milton;

Glad you got it figured out.

When I did my first retrofit someone told me to use a single cable between the stepper motor and driver board. No splices and avoid terminal blocks. I guess the goal is to minimize chances for loose or bad connections. I was told these conditions were torcher on the driver boards and the stepper motors.
Jim

DICKEYBIRD
08-08-2012, 09:05 AM
Hi Jim,

Yeah, I originally planned to have one piece motor cables but my frugality caused me to come up short. The eBay source I chose for the cable sells in 10 ft. lengths and when I ordered 10 ft. I didn't know at the time I was going to mount all the sparky bits in a slide-out tray. After routing the cable in a manner that allowed free movement with no snags or kinks at the extents of the tray & Z-axis movements, I came up about 18" short. Being a weekend warrior and not wanting to wait another week to hook up the motors, I spliced on the 18" at the driver board end. I used the same gauge wire & twisted them together with a drill to hopefully keep noise out.

So far it works beautifully other than the above mentioned short in the splice at the motor end which is now fixed.