Hot motors very noisy during certain G codes

Hey guys,

im new to CNC and just wired up a kit and got it running. I have it jogging smoothly, and it will run programs no problem... but, the motors get so hot you cant touch them, and they are very noisey when doing certain things. (eg G02 or going from G53 G0 X0 Y0 Z0 to G54 G0 X0 Y0 Z0)

the motors almost sound like they're creaking when they sound weird.

this is my setup.

Longs-motors

Nema 17
DB420a driver boards

DB25-1205 breakout board

my timing is

---

5050
5050
500
500

40000

---

The dip switches are set to the lowest possible current setting and 200/1 step (all dip switches on)

max speed 4mm/s
accelleration 750mm/s2

like I said, if i jog the machine at up to 240mm/m it runs smooth. but if it goes slow it sometimes makes considerable noise.

Thanks for any help

Hi

This driver came up recently
www.linuxcnc.org/index.php/english/forum...uration-dm542a#60338

Your driver timings are at considerable variance to the recommended ones.

The dip switches are ..... 200/1 step (all dip switches on)

Why?
That just means it takes 40000 pulses to achieve one revolution of the motor, ignoring any gearing on top.

I would recommend taking that down to 8x max.
Microstepping is not gearing, it can only make moves a bit smoother up to a point, beyond that it has rapidly diminishing returns.

If you reduce your acceleration figure by a factor of 10 also, hopefully you will see some immediate benefits.

regards

Hello ArcEye! Thank you so much for your reply.

I don't believe im fully understanding you. So I have dip switches on the side of the driver board

www.longs-motor.com/productinfo/detail_12_80_148.aspx a picture of it here so you can see.

i have it set to 200 pulse/rev and micro 1. are you saying it should be set to 1600 pulse/rev and 8 micro?

The motors im using have 1.8degree so 200 pulse per rev makes sense. Did i miss something?


as for the timing, I used the wantai timing in the linuxCNC wiki. The motors wont move if I move the first 2 numbers below 5050. Adjusting the second 2 numbers doesn't seem to change anything.

Thanks again for your support.

I thought you meant you had 200x micro stepping on and the bottom figure you gave of 40000 was your scale.
Some of the drivers do allow micro-stepping in factors of 10, 100 etc

Nearly all motors are 200 1.8 degree steps, I would set the micro-stepping to 4x (800) which should be a good compromise, improving smoothness without wasting pulses.
(need to multiply your SCALE by 4 or put it back through stepconf again)

The last 2 timing figures can be quite high, not having them so can result in suddenly finding your machine is going the wrong way.

Try setting the first two to 8000.

If your config is from stepconf, you may find it has been ignored anyway and the step reset by the thread frequency.

Moves involving all axes will be noisier than those on a single axis
If your power supply is limited it can also show up in these moves

regards

Also so I just did some messing around. I am using stepconf you are correct.

I changed the dip switches to 3200/rev micro 16

then in stepconf i did

step time 5050
step space 5050
direction hold 10000
direction setup 10000

---

motor steps per rev 3200
driver microstepping 0.66 (Lets talk about this number because I dont fully understand it. I did 0.66 after some trial and error to get proper movement)
pulley teeth 1.0:1.0
leadscrew ptich 1.0
maximum velocity 11
max acceleration 75

---

that gives at the bottom

time to accelerate to max speed 0.1467s
distance to accelerate to max speed 0.8067mm
pulse rate at max speed 23232.8hz
axis scale 2112.0 steps/mm

with this set up my mill runs quite a bit quieter although the strange noise is still there (quite a bit less annoying) and the top speed is a bit slower (which i can live with). I had to set the micro step to 0.66 to get 20mm of movement off a 20mm command. a microstep of 1 gave me 30mm of movement.

I don't use stepconf normally, so I am not completely familiar with the entry fields.
I just fill in the figures directly in the ini / hal files

axis scale 2112.0 steps/mm

That is very high for a metric machine, I don't have one with a figure higher than 800 on any axis.

I would still advise selecting 4x micro-stepping instead of 16x, higher is not better.
That way you will get 4 x more movement from the same number of pulses and should be able to increase max velocity without risking lost steps

motor steps per rev 3200
driver microstepping 0.66 (Lets talk about this number because I dont fully understand it. I did 0.66 after some trial and error to get proper movement)

Motor steps per rev is always 200 (unless you have rare 100 step motors) it is the physical segments in the rotors which set the number.
Driver microstepping is the multiplication factor eg 16

If you have to put 0.66 in here to get correct movement, the likelyhood is that the leadscrew you thought was metric, is imperial or somesuch

Whatever none of these figures matter in their own right, all they are used for is to calculate the axis SCALE figure which appears in the ini file

This figure is the number of pulses required to move one machine unit (mm)

with this set up my mill runs quite a bit quieter although the strange noise is still there (quite a bit less annoying

I would also still advise that you try setting step time and step space to about 8000ns and try that

If the figure is too small it can be noisy and if 5000ns results in motors that will not move at all, then 5500ns is not enough.

The manufacturers figures are guidelines, often based upon the theoretical figures for the actual chip and little to do with real world usage.

You may have to experiment a bit to find figures at which the motors are stable and responsive.

regards

So I changed the dip switches to 1600/rev 8micro

then input

8000
8000
10000
10000
into the timing fields

then into the axis field

200 steps
8 micro

a test program appears to be giving me the correct distances, and the motors aren't running so hot I can't touch them anymore. The noise is still there but I believe its higher pitch so it sounds quieter.

Thank you very much for your help, and I believe I will do some more reading on this subject to better understand it.