Electronic camshaft


I would like to find a piece of electronic equipment that I think the best description would be an electronic camshaft. In the old days a shaft was hooked to a piece of equipment that had several cams that made and broke switches at various points in a revolution.

I would like something very similar to that, only electronic, that could communicate with a computer to set the on and off degrees of each output, but would turn the outputs on and off autonomously, without having to go through the computer. That always takes too long. I'm looking for 500-600 Hz. ... you'd think that'd be slow and easy but it ain't so.

I'm thinking a 360 ° encoder hooked to a counter whose output was the address buss to a 360 X 16 bit memory.
The memory locations would have the ons and offs programmed into them by the computer and then the data buss could just turn 16 outputs on and off as the counter incremented and enabled each location. Seems to me the speed of that would be pretty much limited only by the IC's.

The closest thing I could find is THIS

It's very close, but there's not a lot of info here and they have not responded to inquiries. Also it seems a bit much for the application, but the diagram is exactly what I have in mind.

Actually, I think there should be something available at the level of a chip or just a few chips on a board, but I can't find it.

Any help ????

How fast would the shaft be turning, what resolution do you need, and how many outputs? What about the physical characteristics of the shaft?

Maybe a 4-bit encoder and a PIC or BASIC Stamp is all you need.
How fast would the shaft be turning, what resolution do you need, and how many outputs? What about the physical characteristics of the shaft?

Maybe a 4-bit encoder and a PIC or BASIC Stamp is all you need.

Assuming the encoder is already physically connected to the shaft....

The speed of the encoder would be 600 Hz max, so it would count from 0 to 360 in approx 0.6 sec.

I would like 16 outputs.( or more (more is always better))

It would have to be capable of counting encoder pulses and changing the state of the outputs in less than 1.6 mS. so it doesn't miss any pulses. That's where the rub always comes. If there is any computing done, it always seems to take too long.

There are some embedded boards that would probably do it, no distributed I/O of which I am aware.

What I would really like is something on a small board level that would communicate with a computer serially or ethernetically (new word? ) to set the on/ off degrees of the outputs, and then count the encoder pulses autonomously

Sort of a 360 X 16 bit memory where the encoder count would put the contents of 360 memory locations on the data buss and the data buss would light up the outputs. The memory locations would have been pre programmed by the computer serially or otherwise. The computer would not need to be involved in polling the encoder or anything else while that happened.

I am unfamiliar with the Basic stamp, but I have been looking into it the last several days.

So, you are saying that you already have a 360 ppr encoder that turns no more than 36 rpm? Can it ever go in reverse?

What kind of outputs does it have? TTL, open collector, differential? Is it quadrature (A, B)?

1.6 mS is a lot of time for hardware or something embedded.

Using an 8/16-bit counter to drive a memory address bus is a simple, old-school way, as long as you have a way to program the memory device. Back then, you would use an EPROM programmed with your table.

As soon as you say serial/Ethernet, though, you are talking about an embedded processor. It could be a PIC, an 8-bit micro (still kicking!), or an AVR. These devices can be used in existing platforms (i.e. STAMP, Rabbit) to ease development.
I have done this very thing with a rabbit 3300 (I think) board. It has ether and serial on board as well as a quadrature decoder. If you buy the eval board there is space for lots of proto 'stuff'. Read the data sheet for more details. Free C compiler with the eval board. I forget the cost but it is for sure higher than a stamp or other solution. Maybe 400 bucks?