Re: [CAD_CAM_EDM_DRO] Re: Knee or Quill Drive?

bwrfromuk wrote:

> --- In CAD_CAM_EDM_DRO@y..., Jon Elson <elson@p...> wrote:
>
> Jon,
>
> I measured the voltage across the motor terminals with a scope, on
> raising the load the trough of the voltage falls to 30 volts as the
> bulk capacitor discharges and rises to 32.5 as it charges. The
> supply would benefit from some additonal volts and the bulk capacitor
> need increasing from its present 10milliFarad.

Well, there's your answer. The motor has very little back EMF at
the speed it must be running at, so the 32 V is being dropped
over the armature resistance. Putting 32 V into a stalled motor
will cook it pretty quick. You have to up the belt reduction ratio
by a good margin. I would now recommend going to at least 5:1,
or as high as you can get, up to 10:1 or so. Or, get a motor with
a lot more torque. 32 V at 8 Amps indicates an armature resistance
of 4 Ohms, which is pretty high. I would want a motor with no more
than 1 Ohm resistance in a system like this.

> I do agree with Peter's earlier comments that it is very convenient
> having the long travel available on the knee as the movement is 0.1"
> per revolution and tool changing with R8 tooling more often than not
> involves droping the knee.
>
> You are correct in your assumption that drive is through bevel gears,
> backlash has been suprisingly good at .003" and the limited tests I
> have made shows no significant cyclical errors.

Well, the bevel gearing on my old Bridgeport is a LOT worse than this.
But, the other reason for not CNC-ing the knee is the knee ways
themselves. There is wear in the middle range of travel that allows
the knee to flop front and back by more than .001". I can't snug
up the gib any more or it would bind at the extremes of travel.
I could CNC the knee after re-scraping the knee ways, though.

Jon