RE: [CAD_CAM_EDM_DRO] Big Servo Drives

The tach signal is isolated so as long as twisted pairs are used plus a
screen to help noise shouldn't be an issue as you say.

Not all analogue cards use the tach signal in current torque mode, with some
it is just for velocity mode.

Encoder signals should also be OK if differential pairs are used.



When my mill was in it's analogue control form, the motor and tach wires
were not screened, just the encoder wires and it ran for years without
problems. But you could say it had separate cards, analogue servo cards, and
a huger PID processor in the main CNC system.



Similarly for us the issue where separating cards could be of use is when
you want to add PID step direction control. A separate PID, Encoder, Step
direction card, and separate power bridge card could be advantageous,
possibly for noise, but also for some uniformity during build. A common
control card, and variety of bridge cards.



As for a PWM input card like Jon has yes the little bit extra isn't worth
bothering with.



Dr. Mark Vaughan Ph'D. B.Eng. M0VAU

Managing Director

Vaughan Industries Ltd, reg in UK no 2561068

Water Care Technology Ltd, reg in UK no 4129351

Addr Unit3, Sydney House, Blackwater, Truro, Cornwall, TR4 8HH, UK.
Phone/Fax 44 1872 561288

RSGB DRM111(Cornwall)

_____

From: CAD_CAM_EDM_DRO@yahoogroups.com
[mailto:CAD_CAM_EDM_DRO@yahoogroups.com] On Behalf Of Jon Elson
Sent: 29 May 2007 05:15
To: CAD_CAM_EDM_DRO@yahoogroups.com
Subject: Re: [CAD_CAM_EDM_DRO] Big Servo Drives



roboticscnc wrote:

> Hi Group,
> Forgive my ignorance as I'm not an electrical engineer, but after
> reading the comments on this subject I'm wondering if it would be
> helpful to have an external power board with overkill mosfets and
> anything else necessary to remove back emf, etc. This way you could
> use any servo driver to do the positioning function with the output
> only switching larger mosfets - keeping high voltage / current noise
> away from the encoder signals. The external power board could then be
> designed for the correct power without dictating which servo driver
> you have to buy.

With proper design, you don't have much problem with this sort
of thing. On my original analog servo amp, I did some testing
of smoothness at low speeds. The movement broke into stick-slip
movement at .01 IPM. I calculated the tachometer voltage at
that speed, it was 7 micro-Volts! Obviously, I could not
measure such a low voltage. But, if I shorted out the
tachometer with a screwdriver, the servo amp would fault
immediately, proving that it was receiving that tiny voltage
and that it was enough to keep the servo loop closed. These
signals ran less than 1 inch from the power stage at certain
points, which were switching up to 5 A at 70 V, and switching
fast, under 100 ns. So, that is 70 v / 10^-7 s = 7 x 10^9 V/s,
a pretty high dv/dt. I did spent a lot of time laying out that
board, even bringing sensitive signals in with the PCB
equivalent of a twisted pair cable.

In theory, it would be possible to build a power booster for a
generic PWM servo amp, but you'd need to tap into their current
measuring circuits to allow the current limit and fault sensing
to work at the higher current level.

But, I think practically, if you are going to have to build the
power stages, driver stages, current limit and fault logic, you
might as well add the additional 10% and make it a complete,
stand-alone servo drive.

Jon





[Non-text portions of this message have been removed]