Re: [CAD_CAM_EDM_DRO] Quadrature Divider ?
Posted by
Tom Hubin
on 2009-01-06 22:54:53 UTC
Hello Alan,
I considered your question as a challenge to an old digital logic
designer, me.
I would use an LS169 binary up/down counter and an LS83 four bit adder
to input quadrature, divide by eight, and output quadrature.
One encoder signal goes to the counter up/down input and the other
encoder signal goes to the counter clock. The counter 3rd output bit
(Q2) is a divide by eight representation of the counter clock, including
direction.
The three least significant counter output bits (Q0, Q1, Q2) are wired
to one of the inputs of the adder. The other adder input is hard wired
to add 2. The third adder output bit is a representation of the
quadrature phased signal to the counter third output bit.
That takes care of the logic with just two TTL ICs. Now we work on
timing analysis and reliability.
Output signal conditioning needs to be considered. The adder output will
be unstable for a few microseconds just after every input change. So the
adder bit should not be used directly as an encoder output. It needs to
pass through a clocked (same clock as the counter) flip flop. To keep
the two encoder divided outputs in sync you need to pass the counter bit
through a similar flip flop. Then both of these to the outside world via
appropriate single ended or differential drivers. A LS240 is a good
choice if single ended.
The encoder signals coming from in the hardware need to be conditioned
too. Differential or single ended line receivers as appropriate. The
signal that is used to clock the counter must make a clean transition so
input conditioning should include hysteresis. LS240 is a good choice if
single ended.
LS240 is an 8 bit device in a 20 pin package. IIRC there is 6 bit
version of this in a 16 pin package.
If differential drivers and receivers needed I would use AM26LS31 and
AM26LS32 (or maybe it was AM26LS33).
My memory is old and rusty so any of these part numbers could be
incorrect. If you are serious and cannot get data on these chips then I
will dig up the spec sheets for you.
Don't leave any inputs floating. For minimum noise all unused inputs
should be tied high or low.
Tom Hubin
thubin@...
> . a quadrature divider .. maybe a PIC doing a bunch of counting.Hello Alan,
>
> (Here's a question .. would a flip-flop wired as a divide-by-two work
> as a simple divider ? And if so, couldn't a guy cascade three of
> them for a divde-by-eight ?)
> Any thoughts, opinions or circuits gratefully accepted.
I considered your question as a challenge to an old digital logic
designer, me.
I would use an LS169 binary up/down counter and an LS83 four bit adder
to input quadrature, divide by eight, and output quadrature.
One encoder signal goes to the counter up/down input and the other
encoder signal goes to the counter clock. The counter 3rd output bit
(Q2) is a divide by eight representation of the counter clock, including
direction.
The three least significant counter output bits (Q0, Q1, Q2) are wired
to one of the inputs of the adder. The other adder input is hard wired
to add 2. The third adder output bit is a representation of the
quadrature phased signal to the counter third output bit.
That takes care of the logic with just two TTL ICs. Now we work on
timing analysis and reliability.
Output signal conditioning needs to be considered. The adder output will
be unstable for a few microseconds just after every input change. So the
adder bit should not be used directly as an encoder output. It needs to
pass through a clocked (same clock as the counter) flip flop. To keep
the two encoder divided outputs in sync you need to pass the counter bit
through a similar flip flop. Then both of these to the outside world via
appropriate single ended or differential drivers. A LS240 is a good
choice if single ended.
The encoder signals coming from in the hardware need to be conditioned
too. Differential or single ended line receivers as appropriate. The
signal that is used to clock the counter must make a clean transition so
input conditioning should include hysteresis. LS240 is a good choice if
single ended.
LS240 is an 8 bit device in a 20 pin package. IIRC there is 6 bit
version of this in a 16 pin package.
If differential drivers and receivers needed I would use AM26LS31 and
AM26LS32 (or maybe it was AM26LS33).
My memory is old and rusty so any of these part numbers could be
incorrect. If you are serious and cannot get data on these chips then I
will dig up the spec sheets for you.
Don't leave any inputs floating. For minimum noise all unused inputs
should be tied high or low.
Tom Hubin
thubin@...
Discussion Thread
alanrothenbush
2009-01-05 17:33:47 UTC
Quadrature Divider ?
Mariss Freimanis
2009-01-05 23:43:04 UTC
Re: Quadrature Divider ?
Yahoo
2009-01-06 07:06:07 UTC
Re: Quadrature Divider ?
Jon Elson
2009-01-06 09:18:07 UTC
Re: [CAD_CAM_EDM_DRO] Re: Quadrature Divider ?
Alan Rothenbush
2009-01-06 09:43:10 UTC
RE: [CAD_CAM_EDM_DRO] Re: Quadrature Divider ?
Alan Rothenbush
2009-01-06 10:00:12 UTC
RE: [CAD_CAM_EDM_DRO] Re: Quadrature Divider ?
Yahoo
2009-01-06 10:54:03 UTC
Re: Quadrature Divider ?
Alan Rothenbush
2009-01-06 15:24:19 UTC
RE: [CAD_CAM_EDM_DRO] Re: Quadrature Divider ?
Peter Reilley
2009-01-06 20:39:18 UTC
Re: [CAD_CAM_EDM_DRO] Re: Quadrature Divider ?
Tom Hubin
2009-01-06 22:54:53 UTC
Re: [CAD_CAM_EDM_DRO] Quadrature Divider ?
henrikolsson2001
2009-01-07 08:52:51 UTC
Re: Quadrature Divider ?
Peter Reilley
2009-01-07 11:06:43 UTC
Re: [CAD_CAM_EDM_DRO] Re: Quadrature Divider ?
Drew Rogge
2009-01-07 11:17:54 UTC
Re: [CAD_CAM_EDM_DRO] Re: Quadrature Divider ?
wildhorsesoftware
2009-01-08 17:38:31 UTC
Re: Quadrature Divider ?
Tom Hubin
2009-01-14 07:21:56 UTC
Encoder Dithering Specs
Jon Elson
2009-01-14 09:53:20 UTC
Re: [CAD_CAM_EDM_DRO] Encoder Dithering Specs
Roland Jollivet
2009-01-14 15:24:48 UTC
Re: [CAD_CAM_EDM_DRO] Encoder Dithering Specs
Tom Hubin
2009-02-27 14:39:03 UTC
WTB small 3 axis CNC for microscopic 3d inspection
Jack McKie
2009-02-27 15:10:48 UTC
Re: [CAD_CAM_EDM_DRO] WTB small 3 axis CNC for microscopic 3d inspection
Jack Mc Kie
2009-02-27 15:43:21 UTC
Re: WTB small 3 axis CNC for microscopic 3d inspection