[hitechwaynew]

New kid on the block here. I put together a modest sized 3-axis (4 motor) CNC, and purchased StepperOnline's 4-Axis 269 oz-in (1.4 ft-lb) NEMA 23 kit to power it with.

The problem I am having is that I can't get anything that remotely resembles stability except at extremely low speeds (< 30 IPM) and with single-stepping, and even then the system seems to 'lose' a few pulses.
Each axis runs well solo - quiet, smooth, and w/ no apparent loss of position (unless I jack up any of the parameters - speed, micro-stepping, etc.) but, when all run together there is LOTS of harmonic-sounding noise and even occasional stalling.

Particulars:

Bed size 38 x 48
Gantry & Carriages "rails" are V-Groove bearings - very low rolling resistance (not measured - yet)
Complete gantry assembly weight is probably around 35 lbs
Cross - machine carriage weight = ~ 4 lbs
Vertical carriage w/ router = ~ 5 lbs

Drives are all 10 pitch x 1/2" ACME screws w/ Delrin anti-backlash nuts.
Torque requirements (again not yet measured) seem quite low - I can very easily turn the screws by hand.
Alignment is all very good.

Wiring
All grounded (120VAC, 24VDC, machine)
non-shielded 18 AWG 2 conductor cable used for all motors (each pair feeds one motor winding)
NO interlock wiring has been done yet (limits, e-stop, etc.)

Software / setup
Using Mach3, all config'd as per the driver manufacturers recommendations.

Might I simply be asking too much from NEMA 23's ? They seem to have plenty of torque (gantry movement has damaged a few things rather than stalling)
Is this to be expected from an extremely cheap system ?

I've contacted the seller of the kit - they're not much use (surprise)

I have had plenty of experience years ago with industrial stepper motor systems and never encountered any of the issues I'm having here.

Any and all suggestions / questions are welcome.

Thanks to all in advance

Wayne

[ger21]

Link to the kit?

Have you run drivertest.exe in the Mach3 folder?
If they are cheap drives, you may be getting resonance issues. You should have them set to 1/8 or 1/10 microstepping, but even then you may still be running into resonance issues. I have dampers on my steppers, to get rid of the resonance.

At 24V, you won't see much speed out of that system with 1/2-10 acme. Best case is probably 50-60ipm?

[hitechwaynew]

Link to the kit? https://www.omc-stepperonline.com/cn...60-23hs30.html

Have you run drivertest.exe in the Mach3 folder? ​​Yes. That passes with flying colors.

If they are cheap drives, you may be getting resonance issues. ​​ They're cheap. Extremely so.

You should have them set to 1/8 or 1/10 microstepping, but even then you may still be running into resonance issues. ​​ I can't even st them to 1/2 or 1/4 without all kinds of issues

I have dampers on my steppers, to get rid of the resonance. â€â€‹ I will have to look into this - any examples you might be able t point me to ?

At 24V, you won't see much speed out of that system with 1/2-10 acme. Best case is probably 50-60ipm?

​​ Yes. 600 RPM would be 60 IPM - likely max. Went with the 10 TPI screws to maximize mechanical advantage. Figured screws & nuts are relatively cheap - I could jump to higher pitch (lower TPI) if this worked out................

Thanks for responding !

Wayne

[ger21]

Have you run drivertest.exe in the Mach3 folder? Yes. That passes with flying colors.

Are you relying on the "System Excellent" message? If so, that is meaningless. Make sure you have a perfectly flat line, with a pulse rate that stays steady, and is very close to the kernel speed. At 25Khz, the pulse rate should be right around 25,000, and not fluctuate much at all.

This thread is where I got the idea for the dampers, and I think I have a pic in there somewhere.
https://www.cnczone.com/forums/stepp...er-damper.html

I can't even st them to 1/2 or 1/4 without all kinds of issues

What are the issues? Do you change the steps/unit in Mach3 when you change the microstepping, as well as the accel and velocity?

[hitechwaynew]

Nope - didn't even notice the "System Excellent" bit right off

Line is generally perfectly flat with occasional low-magnitude 'spikes' (far less magnitude than shown in the tutorial video) although with more frequency when it does happen.

Pulse rate is generally at 24995/6 but occasionally drops to 24260-ish

Yes, steps / unit changed when microstepping changes made. Accel, no (been set to 3 ips/s for all) Velocity, no - I haven't tried going below 15 ipm. Above 20 causes real problems regardless of microstep settings


Thanks, again

[rodw]

They seem to be the same drives as I use. But I'm running them at 48 volts with different drivers (Longs Motor DM542A). I've been happy with them. I microstep them at 20 x and can get 750 RPM out of them. I'm not using Mach or a parallel port but am using Linuxcnc and a Mesa 7i76e ethernet card that can step at up to 10 Mhz. I would be looking at your driver and voltage. I got a feeling your choice of driver is pretty appalling.....

[ger21]

After changing the microstep settings, you need to restart, and redo the accel and velocity settings. I thought there was a pop up message that told you this?

Above 20 causes real problems

What exactly does this mean? Do they just run rough? Stall? Have you tried jogging at different speeds, like 10, and 50? Try increasing the accel to 5, or 10, and see if it makes a difference. If it's a resonance issues, different accel rates can help get through resonance points.

[hitechwaynew]

Do they just run rough? Stall? All of the above - noisy, stuttering, stalling
Have tried Jogging at various speeds - haven't tried above 30 bc at 25 and 30 things go 'bad'. Figured faster wouldn't improve things.

Individual Axis' also behave better alone than when running together running code. For example, just jogging any one of the three axis at speeds up to 20 IPM they run very smooth and quiet as opposed to when even just the X&Y are making a coordinated move. Even something as simple as returning to 0,0 X-Y which should be straight-forward, single direction, single speed moves from wherever it may have been at produces a lot of 'harmonic-ey' noise at any speed (including 20 IPM) and tends to result in hesitation and / or stalling above 20 IPM.

I'm fabbing up a set of dampers to see if they provide any benefit...................

[ger21]

If one axis moves smooth by itself, but multiple axis together don't, I would think it's a PC issue. Open your Windows task manager and check CPU usage while it's moving.

[rodw]

If one axis moves smooth by itself, but multiple axis together don't, I would think it's a PC issue. Open your Windows task manager and check CPU usage while it's moving.

Or migrate to LinuxCNC

[hitechwaynew]

If one axis moves smooth by itself, but multiple axis together don't, I would think it's a PC issue. Open your Windows task manager and check CPU usage while it's moving.

Well, not getting much of anywhere with this one. checked CPU/Memory useage while running GCode and CPU useage peaks around 19%, mostly runs around 10%. Memory is only using about 300Mb of the 800 installed.

Built a pair of 3" / 8 ball dampers and installed them on my X & slaved A axis' - no effect at all. As soon as I bump the speed up to 30 IPM from a relatively stable 20 IPM I get stalls as before. Changing accel rates doesn't seem to affect it either

I don't think the motors themselves are inadequate for the job - it takes a lot of force to stop the gantry movement by hand when jogging. I suspect that the el-cheapo drivers may well be a case of "you get what you pay for".

[Ferrr1976]

Hi there,

I've had the exact same problem. Problem was that my 5Volts supply to the Stepperonline mach3 breakoutboard was not connected. In that case the signals are to weak (together). Connected the 5V power supply to the board and all problems were gone....

[qingwokou7]

I also have similar problem, but now have solved. My stepper motor buy from https://www.oyostepper.it/category-6...o-NEMA-17.html and my power supply from:https://www.skysmotor.co.uk/category...er-Supply.html , their customer help me solve the problem.