SV: [CAD_CAM_EDM_DRO] Re: breakout boards plus - hard choice

Hi guys,
Here's the message Mariss posted on his Yahoogroup on may 28 2007. I think
it will provide answers to some of your questions.

/Henrik Olsson.

****** Original message from Mariss ********
Hi,

Remember the "unstallable step motor project"? I haven't, have been working
on it and there are new developments.

A month ago I came to the conclusion the approach I had been working on for
two years wouldn't and frankly never could work. It was to tweak the step
and direction inputs on an encoder feedback closed-loop step motor drive to
get the performance the motor can have closed loop.

This revelation suggested the key to the solution is to make the motor drive
self-commutating. This is not in itself an unknown practice though it is a
little obscure. It is obscure only because it in turn requires phase-lag
compensation that no previous implementation has satisfactorily solved. I
have come up with a simple closed-loop solution that is patentable and I
have it running on the bench right now.

What I have is a mild-mannered Clark Kent kind of motor (NEMA-23, 2A per
phase) running at 15,000 RPM with a 24VDC power supply. The drive is self
commutating so it will run the motor up to its maximum possible speed. The
is the 500-line encoder is at its frequency limit.

This is an unharnessed drive. All I can control is the direction of rotation
right now. That in itself is very interesting.

Changing direction results in an initial slow deceleration from 15,000 RPM.
As speed slows and the drive backs up on the motor speed-torque curve into
increasing torque, deceleration increases markedly until the motor actually
jumps on the bench at direction reversal and accelerates away in the new
direction. At all times the motor is delivering maximum possible torque
while doing this and it cannot be stalled.

I have the motor on a dynomometer that I can program an abrupt torque load
(torque discontinuity) to the test motor. The motor does not stall. It's
speed drops abruptly in response to the step function torque load in a
critically damped fashion.

I have also had the motor hit a hard stop. It adapts without ever
desynchronizing (losing steps). The self-commutating circuit is utterly
stable. In fact, when I pull the power on the drive, it decelerates to a
stop using commutated regenerated energy to to keep the drive alive while it
is decelerating without ever stalling (skipping steps). Simply put, it
cannot miss a step or desychronize.

The drive is unharnessed right now. I will be putting a harness on it to
control its speed and position. The method is patentable as well. A cool
feature of the method is with an encoder the drive is closed-loop with all
the features described. Without an encoder it's just another really nice,
ordinary Clark Kent kind of microstep motor drive.

Mariss
**********************************************************************

-----Ursprungligt meddelande-----
Från: CAD_CAM_EDM_DRO@yahoogroups.com
[mailto:CAD_CAM_EDM_DRO@yahoogroups.com] För R Rogers
Skickat: den 20 juli 2007 18:33
Till: CAD_CAM_EDM_DRO@yahoogroups.com
Ämne: Re: [CAD_CAM_EDM_DRO] Re: breakout boards plus - hard choice

Quote: You could be right, the encoder was only for analysis of the
results, but that wasn't my memory of it.

You're right. He's using an encoder for feedback. Not part of the test,
but the intended design. Because it was argued that it's just essentially
the same as a servo from an application standpoint. Drive, motor and
encoder.

If lousy yahoo didn't delete my archive, I could pull up the post. I have
"mail plus" they announced one day I had unlimited storage, the next day it
was empty. I guess thats what they meant...

Ron


Jon Elson <elson@...> wrote:
Mark Vaughan wrote:

> Hi Jon
>
> The G100 just has step direction output, and encoder inputs, it does have
> internal processing power, and it does command the step federate so it

could

> detect and error and slow things down.

Sorry, I didn't mean to imply that a STOCK G100 would be part of
this unstallable stepper scheme. It may, in fact, be possible
to do it without adding hardware to the G100, Mariss hasn't been
totally clear on this because he doesn't want to reveal too much
before he has a product ready to go.
However when released the firmware

> with the G100 had many many fault, Mariss coped out saying he was not a
> programmer, so I very much doubt he has implement a closed loop himself
> within the G100. The guy that now develops the G100 firmware Stephen, is a
> good programmer, but he is generally very vocal when he gets a new trick
> running, and we haven't heard from him regarding this.
>

I am confident that Mariss did, indeed, make the unstallable
stepper work on his bench. That doesn't mean there were not
pieces of laboratory test gear still wired into the system, and
I know he was not controlling it with G-code or Mach(x). It was
entirely a lab demo, not a CNC machine tool, but I know Mariss'
style well enough to know that if he said "it worked", then
indeed it DID work!

> I am still fairly sure Mariss is looking at the current response to a

pulse

> and from that determining impeding failure of a step so he knows when to

cut

> back the feed rate and store pulse errors with his stepper card. It's been

a

> long time since I studied steppers, but if memory serves me right as you
> load a stepper, the current lags and the crest factor changes, just before
> stalling there are massive crest factor changes.

All this should be in the archives of this list or the Gecko
list, I don't remember where the major discussion developed. At
speed, (and maybe that is the only place stalls should occur)
there should be enough back-EMF to know what is happening at the
motor. My concern was that at near-zero speed, there won't be
enough back-EMF to detect in the noisy environment of a stepper
drive, and so a stall could occur without being sensed. This
would require running the machine into an obstacle (a "crash")
but it is a situation that happens all the time in the shop.
You could be right, the encoder was only for analysis of the
results, but that wasn't my memory of it.

Jon





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



Addresses:
FAQ: http://www.ktmarketing.com/faq.html
FILES: http://groups.yahoo.com/group/CAD_CAM_EDM_DRO/files/
Post Messages: CAD_CAM_EDM_DRO@yahoogroups.com

Subscribe: CAD_CAM_EDM_DRO-subscribe@yahoogroups.com
Unsubscribe: CAD_CAM_EDM_DRO-unsubscribe@yahoogroups.com
List owner: CAD_CAM_EDM_DRO-owner@yahoogroups.com, wanliker@...,
timg@...
Moderator: pentam@... indigo_red@... davemucha@...
[Moderators]
URL to this group: http://groups.yahoo.com/group/CAD_CAM_EDM_DRO

OFF Topic POSTS: General Machining
If you wish to post on unlimited OT subjects goto:
aol://5863:126/rec.crafts.metalworking or go thru Google.com to reach it if
you have trouble.
http://www.metalworking.com/news_servers.html

http://groups.yahoo.com/group/jobshophomeshop I consider this to be a
sister site to the CCED group, as many of the same members are there, for OT
subjects, that are not allowed on the CCED list.

NOTICE: ALL POSTINGS TO THIS GROUP BECOME PUBLIC DOMAIN BY POSTING THEM.
DON'T POST IF YOU CAN NOT ACCEPT THIS.....NO EXCEPTIONS........
bill
List Mom
List Owner


Yahoo! Groups Links