OSE / CNC-Torch-Table-OSE

CNC Torch Table
http://opensourceecology.org/wiki/CNC_Torch_Table_2
11 stars 7 forks source link

Figure out why small movements & ends of movements make motors spazz #6

Open dbenamy opened 10 years ago

dbenamy commented 10 years ago

It seems the system doesn't handle small or slow movements well. The motors start buzzing. This doesn't happen every time but is easy to trigger by making a small movement in Printrun (0.1 or 1). Bigger movements work smoothly but then sometime do this at the end.

I think the big movements start out fast and slow down at the end at that's when this happens.

The controller may be trying to feed them too little current or in some way underpowering them for the small/slow move but they wind up not being to move at all.

Another hypothesis is that there are areas of movement with higher friction and if the motors happen to be trying to make a tiny move with no momentum over those areas, they stall.

When this happens, I click stop motors in Printrun and it stops.

Two obvious things to try are:

dbenamy commented 10 years ago

The pololu data sheet says it "can deliver up to approximately 1 A per phase without a heat sink or forced air flow (it is rated for 2 A per coil with sufficient additional cooling)."

Our long axis motor data sheet is at http://www.xylotex.com/24H290-25-4B.pdf and says they're good for 2.5A per phase. It says they're only good for 5.5V. http://www.indigent-networks.com/pololu-drivers-current-limit-configuration/ says it's safe to overdrive the voltage as long as the current is within limits. Shrug.


Our short axis & height motors are 57bygh317, specs at http://www.circuitspecialists.com/nema_23_stepping_motor_57bygh317.html. Max current is 2.5A.


Unfortunately the vref test point on the pololu boards is half under the heatsink and hard to get to. Fortunately, we should be able to max out the current limit since our motors can handle 25% more current than the pololu max.

To do this we're going to need a fan blowing across the pololu heat sinks.

dbenamy commented 10 years ago

Oh, whoops. I can measure vref on the pot if I need it.

dbenamy commented 10 years ago

Hmm, I rotated the pots fully clockwise which put vref at it's max but don't see much different in the buzzing.

dbenamy commented 10 years ago

For future reference: vref is about 1.7V on the 3 pololu driver boards.

dbenamy commented 10 years ago

Hypotheses to investigate:

It may not actually matter much. While it's cutting it'll be moving and may not happen, and when it's done cutting, we can send a stop motors command or power off the arduino which will stop the buzzing.

dbenamy commented 10 years ago

I set all 4 drivers' current limit pots to about 3/4 to reduce the risk of overheating since max didn't solve the problem.