Re: [CAD_CAM_EDM_DRO] Re: mounting encoder with pulley
Posted by
Stephen Wille Padnos
on 2005-02-18 19:38:43 UTC
caudlet wrote:
initial conditions of the system - you have to initialize your
counter/dividers so that one phase is at 0 counts, and the other is
halfway through the divide ratio, otherwise your two outputs will change
on successive input phase changes, which wouldn't be 90 degrees of
output phase.
The best thing to do in hardware would be to only output one step for
each full cycle instead of 4. Do this by selecting one phase as the
direction signal and the other as clock. You still need a 2-bit counter
and a binary->gray code converter, but those are pretty trivial. What
you lose is the ability to do any ratio other than 1/4 (or 1/4 * some
power of two). Of course, you can chain multiple dividers to get other
division rates.
want to use the internal oscillator), but that's just me. (plus,
they're all flash, so I can make a bunch of mistakes without too much
extra effort :) )
I don't think it's a trivial thing to do this right. USDigital sells an
encoder divider for $39.00 - almost as much as a replacement encoder
from them. (It's called the EDIVIDE, and it looks pretty cool,
actually. This may be perfect for my situation.)
- Steve
>Gee, PAL's and PLA's and all that stuff! I would just grab a CD4015BYou'd run into problems with phase, direction, and jitter. Consider the
>Dual four bit shift register (about 50 cents) and pick a divide by 2,
>4, 8, or 16 output. The encoder just puts out two 90 deg phase
>shifted square waves and turning it into quadrature is the work of
>internal circuits. If your line count was 1000, then each channel
>puts out 1000 pulses. If you divided each by 4 then you end up with
>250 each and back to 1000 in quadrature. Use the divide by 8 and you
>are down to 500 in quadrature. As long as you kept the 90 phase shift
>it should work out. One chip. You might have to use an LS type chip
>if the RPM of the motor was very high.
>
>
initial conditions of the system - you have to initialize your
counter/dividers so that one phase is at 0 counts, and the other is
halfway through the divide ratio, otherwise your two outputs will change
on successive input phase changes, which wouldn't be 90 degrees of
output phase.
The best thing to do in hardware would be to only output one step for
each full cycle instead of 4. Do this by selecting one phase as the
direction signal and the other as clock. You still need a 2-bit counter
and a binary->gray code converter, but those are pretty trivial. What
you lose is the ability to do any ratio other than 1/4 (or 1/4 * some
power of two). Of course, you can chain multiple dividers to get other
division rates.
>Of course you could use a PIC and spend 4.00 bucks and hours writingI'd use an AVR, and spend $1.65 on the CPU and resonator ($1.00 if I
>a program to do the same thing and it would be really neat. I am at
>the stage in my design career where 'really neat' has been replaced
>with 'do it fast and cheap' so I have more time to make chips.
>
>
want to use the internal oscillator), but that's just me. (plus,
they're all flash, so I can make a bunch of mistakes without too much
extra effort :) )
I don't think it's a trivial thing to do this right. USDigital sells an
encoder divider for $39.00 - almost as much as a replacement encoder
from them. (It's called the EDIVIDE, and it looks pretty cool,
actually. This may be perfect for my situation.)
- Steve
Discussion Thread
turbulatordude
2005-02-14 18:54:46 UTC
mounting encoder with pulley
cnc002@a...
2005-02-14 19:11:06 UTC
Re: [CAD_CAM_EDM_DRO] mounting encoder with pulley
cnc_4_me
2005-02-14 19:26:10 UTC
Re: mounting encoder with pulley
R Rogers
2005-02-14 19:34:11 UTC
Re: [CAD_CAM_EDM_DRO] mounting encoder with pulley
turbulatordude
2005-02-14 19:59:52 UTC
Re: mounting encoder with pulley
cnc_4_me
2005-02-14 21:46:17 UTC
Re: mounting encoder with pulley
turbulatordude
2005-02-14 23:18:37 UTC
Re: mounting encoder with pulley
cnc_4_me
2005-02-15 00:31:44 UTC
Re: mounting encoder with pulley
Graham Stabler
2005-02-15 02:59:58 UTC
Re: mounting encoder with pulley
turbulatordude
2005-02-15 06:55:02 UTC
Re: mounting encoder with pulley
Alan Marconett
2005-02-15 23:02:50 UTC
Re: [CAD_CAM_EDM_DRO] Re: mounting encoder with pulley
Lance Hopper
2005-02-16 04:49:13 UTC
Re: mounting encoder with pulley
Andy Wander
2005-02-16 06:44:58 UTC
RE: [CAD_CAM_EDM_DRO] Re: mounting encoder with pulley
Stephen Wille Padnos
2005-02-16 07:33:48 UTC
Re: [CAD_CAM_EDM_DRO] Re: mounting encoder with pulley
Alan Marconett
2005-02-16 12:14:42 UTC
Re: [CAD_CAM_EDM_DRO] Re: mounting encoder with pulley
caudlet
2005-02-16 21:22:03 UTC
Re: mounting encoder with pulley
Alan Marconett
2005-02-17 09:12:43 UTC
Re: [CAD_CAM_EDM_DRO] Re: mounting encoder with pulley
Stephen Wille Padnos
2005-02-17 11:46:06 UTC
Re: [CAD_CAM_EDM_DRO] Re: mounting encoder with pulley
Fred Smith
2005-02-17 12:30:26 UTC
Re: mounting encoder with pulley
Stephen Wille Padnos
2005-02-17 13:47:26 UTC
Re: [CAD_CAM_EDM_DRO] Re: mounting encoder with pulley
Fred Smith
2005-02-17 15:57:36 UTC
Re: mounting encoder with pulley
caudlet
2005-02-17 16:55:50 UTC
Re: mounting encoder with pulley
Stephen Wille Padnos
2005-02-17 19:12:44 UTC
Re: [CAD_CAM_EDM_DRO] Re: mounting encoder with pulley
caudlet
2005-02-18 07:48:52 UTC
Re: mounting encoder with pulley
Alan Marconett
2005-02-18 13:30:27 UTC
Re: [CAD_CAM_EDM_DRO] Re: mounting encoder with pulley
caudlet
2005-02-18 17:06:51 UTC
Re: mounting encoder with pulley
Alan Marconett
2005-02-18 18:01:45 UTC
Re: [CAD_CAM_EDM_DRO] Re: mounting encoder with pulley
Stephen Wille Padnos
2005-02-18 19:38:43 UTC
Re: [CAD_CAM_EDM_DRO] Re: mounting encoder with pulley