Quadrature detection

[David Lesher]

A friend is looking for directional pulse output from a interrupter
disk. He needs a pulse train out of A when going CW, and nothing
out B. CCW rotation gets you the opposite -- pulse output from B.

I recall a design called quadrature detection used in mice, but
that's all I recall. Any pointers to circuit designs?

 

[Allan Herriman]

A friend is looking for directional pulse output from a interrupter
disk. He needs a pulse train out of A when going CW, and nothing
out B. CCW rotation gets you the opposite -- pulse output from B.

I recall a design called quadrature detection used in mice, but
that's all I recall. Any pointers to circuit designs?

http://groups.google.com/groups?oi=djq&as_umsgid=<[email protected]>#1/1
http://jamaica.ee.pitt.edu/Archives/NewsGroupArchives/comp.lang.vhdl/Aug1998/14320.txt

Regards,
Allan.

 

[CBarn24050]

you could program a state machine in a small GAL 16v8 for example.

 

[Tony]

US Digital (www.usdigital.com) has an interface chip, 8 pin, that does just
that. Part #LS7083.

Tony

 

[Jacques Fournier]

A friend is looking for directional pulse output from a interrupter
disk. He needs a pulse train out of A when going CW, and nothing
out B. CCW rotation gets you the opposite -- pulse output from B.

I recall a design called quadrature detection used in mice, but
that's all I recall. Any pointers to circuit designs?

http://www.lsicsi.com
LS7183 (quadrature clock converter)

 

[FSMüller]

this is very easy:
take a D-FlipFlop ( a 7474 e.g.) and connect the clock input to one of
the outputs of the photointerrupter and the other output to the D input.
Now, everytime there is a positive slope on the clock input of the 7474
(from one track signal), the state of the D input is clocked into the
7474. The signal on the Q output of the 7474 indicates the direction of
the encoder, which can be used to switch the clock signal into different
directions. that´s all.

Frank

 

[Roger Hamlett]

http://www.lsicsi.com
LS7183 (quadrature clock converter)

If he has the 'disk', but no detectors, then look at the Honeywell HLC2705.
This is a opto detector system, with the two detectors needed, and the
decoding logic built in, to give a 'count' output, and a 'direction' output.
If you simply 'gate' the count output with the direction signal, you will
have everything needed.
http://content.honeywell.com/sensing/prodinfo/infrared/catalog/Pg_212.pdf

Best Wishes

 

[budgie]

this is very easy:
take a D-FlipFlop ( a 7474 e.g.) and connect the clock input to one of
the outputs of the photointerrupter and the other output to the D input.
Now, everytime there is a positive slope on the clock input of the 7474
(from one track signal), the state of the D input is clocked into the
7474. The signal on the Q output of the 7474 indicates the direction of
the encoder, which can be used to switch the clock signal into different
directions. that´s all.

I was waiting for someone to mention the D flip-flop approach. We used that
over 20 years back on vehicle tripmeters to increment/decrement the distance
register.

 

[R. Steve Walz]

A friend is looking for directional pulse output from a interrupter
disk. He needs a pulse train out of A when going CW, and nothing
out B. CCW rotation gets you the opposite -- pulse output from B.

I recall a design called quadrature detection used in mice, but
that's all I recall. Any pointers to circuit designs?

-------------------
Here:

http://www.armory.com/~rstevew/Public/Motors/Quadrature/QuadDecoderCounter.gif

The top half of the circuit...

-Steve

 

[David Lesher]

{Many replies..}

Thanks to all...

 

[Ben Bradley]

I was waiting for someone to mention the D flip-flop approach. We used that
over 20 years back on vehicle tripmeters to increment/decrement the distance
register.

Hmm, I would have thought that application should be increment
only. I recall an old Perry Mason TV episode where Perry put his hand
on the hood of a car that allegedly hadn't been driven, but the hood
was warm. The antagonist had put the back wheels on blocks and ran the
car in reverse for an hour (dunno how he didn't get carbon monoxide
poisoning) to take some miles off the odometer as an alibi.

 

[Jim Thompson]

Hmm, I would have thought that application should be increment
only. I recall an old Perry Mason TV episode where Perry put his hand
on the hood of a car that allegedly hadn't been driven, but the hood
was warm. The antagonist had put the back wheels on blocks and ran the
car in reverse for an hour (dunno how he didn't get carbon monoxide
poisoning) to take some miles off the odometer as an alibi.

"Ferris Bueller's Day Off" ?

...Jim Thompson

 

[budgie]

Hmm, I would have thought that application should be increment
only.

(snip)

This was a tripmeter for rally cars. If you overshoot a junction, you don't
always reverse back - sometimes you flick turn and drive forwards to return to
the junction. Either way, it was considered desirable to be able to get the
trip-meter to decrement back to the reading it would_have_been at the junction,
so there was a +/- direction switch as well as the separate increment/decrement
register inputs. I hope that clarifies the application.