RE: cnc,linear encoder design question
Posted by
Deon Styger
on 2002-03-17 23:58:39 UTC
Bill,
Some more comments on your linear encoder questions:
1) A ballscrew rated at 0.004" accuracy is usually much more accurate than
the rating. The rating can be seen as a maximum tolerance value.
2) The error on your ballscrew is not necessarily linear. This means that
certain areas of your ballscrew can be more accurate than others. The
specified tolerance accounts for the maximum inaccuracy over a specified
length.
3) Many (most?) controller software can compensate for the above two points.
The errors on your ballscrew at certain locations can be entered in a table
and compensated for. Measuring (mapping) these errors accurately at various
points can however prove to be a bit tricky. Check the archives for
discussions on this.
4) Apart from backlash being a cause of possible oscillations, there is also
flexing. Mounting your encoder directly onto your motor eliminates all
flexing that might be present in your drive train (belts,etc).
5) Gearing. Mounting your encoder directly onto your motor takes advantage
of the gearing of your drive train. If you have a reduction of 2:1, your
encoder resolution will in effect be doubled. Your motor will rotate twice
as much for the same table movement and the encoder will then also count
twice as many lines. This means that depending on your gearing, you could
get away with cheaper encoders or higher resolution for the same encoder.
6) Machine accuracy. Don't ignore the accuracy of your machine. All encoders
will only measure movement in one axis. Inaccurate bed ways (not
perpendicular to each other) will induce an unwanted movement in one or more
of the other axes. You might find that the biggest source of inaccuracy
might be your machine. This is why some machines are so much more expensive
than others.
Just my 2 cent's worth.
Regards,
Deon Styger.
---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.336 / Virus Database: 188 - Release Date: 11/03/2002
Some more comments on your linear encoder questions:
1) A ballscrew rated at 0.004" accuracy is usually much more accurate than
the rating. The rating can be seen as a maximum tolerance value.
2) The error on your ballscrew is not necessarily linear. This means that
certain areas of your ballscrew can be more accurate than others. The
specified tolerance accounts for the maximum inaccuracy over a specified
length.
3) Many (most?) controller software can compensate for the above two points.
The errors on your ballscrew at certain locations can be entered in a table
and compensated for. Measuring (mapping) these errors accurately at various
points can however prove to be a bit tricky. Check the archives for
discussions on this.
4) Apart from backlash being a cause of possible oscillations, there is also
flexing. Mounting your encoder directly onto your motor eliminates all
flexing that might be present in your drive train (belts,etc).
5) Gearing. Mounting your encoder directly onto your motor takes advantage
of the gearing of your drive train. If you have a reduction of 2:1, your
encoder resolution will in effect be doubled. Your motor will rotate twice
as much for the same table movement and the encoder will then also count
twice as many lines. This means that depending on your gearing, you could
get away with cheaper encoders or higher resolution for the same encoder.
6) Machine accuracy. Don't ignore the accuracy of your machine. All encoders
will only measure movement in one axis. Inaccurate bed ways (not
perpendicular to each other) will induce an unwanted movement in one or more
of the other axes. You might find that the biggest source of inaccuracy
might be your machine. This is why some machines are so much more expensive
than others.
Just my 2 cent's worth.
Regards,
Deon Styger.
---
Outgoing mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.336 / Virus Database: 188 - Release Date: 11/03/2002
Discussion Thread
docholliday01201
2002-03-15 16:47:54 UTC
cnc,linear encoder design question
Dave DIllabough
2002-03-15 18:22:11 UTC
Re: [CAD_CAM_EDM_DRO] cnc,linear encoder design question
mariss92705
2002-03-15 18:53:51 UTC
Re: cnc,linear encoder design question
Deon Styger
2002-03-17 23:58:39 UTC
RE: cnc,linear encoder design question