Rumba resets after/during movement when two axes are moving

Hi,
after having trouble with a ball bearing and motor cables which became hot (I used Rs232 cables for large motors) I decided to remount everything. Now I have thicker cables. End stop and motor wirings are separated now.
x,y,z and extruder motor work fine when I use them independent.
If I move x and Y for about 2 seconds (in my case around 100mm), suddenly the rumba board resets itself. This is highly reproducable. If I go to the same position by moving first x, than y everything is fine.
I think that rumba restarts because I can see that the display goes blue and it restarts. Repetier Host says "printer reset detected - initializing".
A multimeter on the main power of 28V (no changes betwen 20 an 30 V) showed that power seems to be stable. I will try to check that again using an oszilloscope.

Can someone tell me under which cicumstances the rumba board may reset itself automatically? Or what may be my problem? Or may you help me by using google? I really tried it for more than one hour and found only this: [forum.e3d-online.com]
I do already use the screw mounts. I will check tomorrow if they are damaged (maybe I used too much force an the soldering to the board broke?), but I do not understand why this should create this behaviour.


My config:
- Rumba board with Repetier Firmware 0.92.9
- DRV 8825 Stepper driver
- 2.5 V / 4.5 A steppr motors (Isel milling machine changed to 3D printer); Motor type RS 440-470
- Repetier Host in Win7
- no heated bed connected at the moment

best rgards,
karl.ranseier

Edited 2 time(s). Last edit at 01/04/2017 07:23AM by karl.ranseier.

Are you running 4.5A steppers directly from the Rumba board drivers? It seems, they have too low impedance and require a separate external driver.
Or did you find pololu style drivers, that match the high current steppers?

Thanks for that hint. That's a good thought.
I forgot to mention the driver: I am using the DRV8825 (added it in the first post). I do not drive them with maximum power.

Maybe the high resistance cables compensated some voltage peaks, which appear when the motor stops? Would the Rumbaboard be switched off by those peaks coming from the motor?

The motors do have 4 coils. When I was running them in serial configuration I had more problems with acceleration and speed. I than changed them back to parallel configuration. In that time I was running the machine on 12 V. Increasing the voltage to 24 and 30 V increase the maximum speed (I changed the heater to 24 V; problems accour even with heater switched off). I did not try to put them back in serial after increasing the voltage.

A comment on the serial and parallel configuration of the coils:
P=U*I and U=R*I gives P=R*I². Assuming R=1 Ohm for each coil, parallel would give 0.5 Ohm and serial 2 Ohm. That means that at a constant current defined by the driver could increase the power by a factor of 4, hoping that the torque is proportional to it. Am I correct?

edit: I found out that the impedance is proportional to impedance=sqrt(resistance²+reactance²). If the low impedance is the problem, wiring them in serial configuration should help, or am I wrong?

Edited 2 time(s). Last edit at 01/04/2017 07:31AM by karl.ranseier.

Quote

If the low impedance is the problem, wiring them in serial configuration should help, or am I wrong?

No, you are right, wire them in series.
At 24V you should get enough speed and acceleration.

I have changed my wiring so that I can really fast change between series and parallel configuration (thanks to WAGO).
In series configuration the motors are less noisy. It seems to be good. Speed tests have to be done later, because:

The original problem still exists. The rumba-board sems to switch off during motion with two motors. It is really during the motion and not at the end.

I have only one idea to test: insert at each side of the coils 1 Ohm ( I do have some 20 W resistors, but not 8. So I would need to order some) in order to simulate the old bad cables and if the problem vanishes or if it stems from another part of the machine.

Or do you have another idea?

Solved!
I had set my micro stepping to 1/32 resulting in 1280 steps/mm. At a speed of 40 mm/s that was eventually too much. I do not understand why the board restarts then, but I will accept it :-)

edit: the question is why this did not happen before...

Edited 1 time(s). Last edit at 01/04/2017 01:56PM by karl.ranseier.

I would take a look at the 5V onboard regulator, maybe it "passes out".
Also the 5A polyfuse is the same as on the Ramps boards and they are known to be wacky.

Do you think it will be conctantly <5V or just in the moment when it stops? I will try to measure it, but my oszi is very bad.
With an older rumba board I damaged the regulator when measuring, so I will take more care ;-). A cable connected to 5V was for some ms on a random place of the board and immideatly the regulator smoked and later burned (now I do have a smoke detector!). The positive thing: I do have a spare part regulator now.

Yet I have recognized only the 15A fuse (which in my case looks like one from the car, it is replacable).

So in this pic: [reprap.org] the left yellow part directly at the power connectors is the polyfuse? Thought it is some weird capacitor ;-)

You can measure 5V on the endstop connectors. No need to kill the regulator
Yes, the "senfgelbe" part on the bottom right of the picture is a polyfuse.

Sorry for the long time for the answer.
The voltage was always ok.

Meanwhile I have changed the drivers to Steprockers. The problem is gone. To be honest, I think it was NOT a problem of the drivers. More probably I have done anything else wrong, which is now correct after reassembling everything.

Thanks again for your help.

Edited 1 time(s). Last edit at 03/25/2017 09:28AM by karl.ranseier.