Help please with stepper driver step buffer

E

Eric R Snow

Jan 1, 1970
0
Greetings All,
I'm trying to build a stepper motor driven indexer. The goal is to use
the inherent accuracy of a stepper motor and encoder to index a wheel.
US Digital sells an IC that uses the output from an encoder to
generate step and direction pulses. These pulses can then be used with
a stepper motor driver IC to drive a stepper motor. It works well
until the stepper motor recieves pulses faster than it can move and
steps are lost. If it's easy enough, and cheap enough, I would like to
build some type of buffer that will accept the signals from either an
encoder or the step and direction IC and output them at a certain rate
that doesn't exceed the stepper motor's maximum step rate. Is this
something that can be easily done by someone who is good at following
directions, can solder well, but really doesn't know much about
electronics? Any help would be great.
Thanks,
Eric R Snow
 
R

Rich Grise

Jan 1, 1970
0
Greetings All,
I'm trying to build a stepper motor driven indexer. The goal is to use the
inherent accuracy of a stepper motor and encoder to index a wheel. US
Digital sells an IC that uses the output from an encoder to generate step
and direction pulses. These pulses can then be used with a stepper motor
driver IC to drive a stepper motor. It works well until the stepper motor
recieves pulses faster than it can move and steps are lost. If it's easy
enough, and cheap enough, I would like to build some type of buffer that
will accept the signals from either an encoder or the step and direction
IC and output them at a certain rate that doesn't exceed the stepper
motor's maximum step rate. Is this something that can be easily done by
someone who is good at following directions, can solder well, but really
doesn't know much about electronics? Any help would be great. Thanks,
Eric R Snow


The obvious answer is, slow down the pulses you're sending to the stepper.

How are you generating them now?

Thanks,
Rich
 
E

Eric R Snow

Jan 1, 1970
0
The obvious answer is, slow down the pulses you're sending to the stepper.

How are you generating them now?

Thanks,
Rich
The IC that generates the step and direction recieves the output from
an encoder that is spun by hand. So what happens is that since the
encoder divides each revolution into discrete steps these discrete
steps can then be used to index a disc, driven by a stepper motor. If
the person turning the encoder too fast spins it too fast steps may be
lost. I could somehow limit the speed at which the encoder can be
spun, but would rather buffer the signals somehow in order to have an
easily spun encoder, with 100 detented positions.
Eric
 
R

Rich Grise

Jan 1, 1970
0
The obvious answer is, slow down the pulses you're sending to the stepper.

How are you generating them now?
The IC that generates the step and direction recieves the output from
an encoder that is spun by hand. So what happens is that since the
encoder divides each revolution into discrete steps these discrete
steps can then be used to index a disc, driven by a stepper motor. If
the person turning the encoder too fast spins it too fast steps may be
lost. I could somehow limit the speed at which the encoder can be
spun, but would rather buffer the signals somehow in order to have an
easily spun encoder, with 100 detented positions.
Eric[/QUOTE]

Then you'll need some kind of FIFO - when the step rate from the knob
gets faster than the stepper motor can take, store up the steps, and clock
them out at whatever rate the motor can handle. Actually, it sounds like
kind of a fun project - I've driven steppers with a Z80, and I've written
a FIFO with an 8748 or so - if you have a micro development system, it's
just software. ;-)

You _could_ do a hardware FIFO, but it'd be a nightmare, unless you have
a CPLD/FPGA development system. ;-)

Good Luck!
Rich
 
E

Eric R Snow

Jan 1, 1970
0
Then you'll need some kind of FIFO - when the step rate from the knob
gets faster than the stepper motor can take, store up the steps, and clock
them out at whatever rate the motor can handle. Actually, it sounds like
kind of a fun project - I've driven steppers with a Z80, and I've written
a FIFO with an 8748 or so - if you have a micro development system, it's
just software. ;-)

You _could_ do a hardware FIFO, but it'd be a nightmare, unless you have
a CPLD/FPGA development system. ;-)

Good Luck!
Rich
Thanks Rich,
I was hoping for a single IC solution. It looks like I'll have to
limit the speed mechanically. Maybe. The ultimate goal is to come up
with a design that amateur machinists can make. The cost needs to be
low and the electronics need to be simple. I'm not averse to using a
microcontroller myself but if I went this route I'd have to supply
pre-programmed microcontrollers. I just want to publish the info and
make it available for free non-commercial use. There are many hobby
machinists who are retired and have limited incomes. They also, as a
group, tend to not be real familiar with more than basic electronics.
So I need to be able to provide them with good instructions and a
parts list with common devices. In other words, directions suitable
for me. There is one option that I've considered and that's position
feed back from the stepper motor. This would require another encoder,
LED display, and another IC. But as long as the counts from the
manually turned encoder match the counts from the stepper motor all is
well. Another option is to use the Z pulse from the encoder and a cam
operated switch on the stepper shaft to turn on a couple LEDs. This
would only verify position once every revolution. But it's the high
count indexing where people will be in a hurry so it may be the best
option.
Eric
 
R

Rich Grise

Jan 1, 1970
0
Then you'll need some kind of FIFO - when the step rate from the knob
gets faster than the stepper motor can take, store up the steps, and clock
them out at whatever rate the motor can handle. Actually, it sounds like
kind of a fun project - I've driven steppers with a Z80, and I've written
a FIFO with an 8748 or so - if you have a micro development system, it's
just software. ;-)

You _could_ do a hardware FIFO, but it'd be a nightmare, unless you have
a CPLD/FPGA development system. ;-)
Thanks Rich,
I was hoping for a single IC solution. It looks like I'll have to
limit the speed mechanically. Maybe. The ultimate goal is to come up
with a design that amateur machinists can make. The cost needs to be
low and the electronics need to be simple. I'm not averse to using a
microcontroller myself but if I went this route I'd have to supply
pre-programmed microcontrollers. I just want to publish the info and
make it available for free non-commercial use. There are many hobby
machinists who are retired and have limited incomes. They also, as a
group, tend to not be real familiar with more than basic electronics.
So I need to be able to provide them with good instructions and a
parts list with common devices. In other words, directions suitable
for me. There is one option that I've considered and that's position
feed back from the stepper motor. This would require another encoder,
LED display, and another IC. But as long as the counts from the
manually turned encoder match the counts from the stepper motor all is
well. Another option is to use the Z pulse from the encoder and a cam
operated switch on the stepper shaft to turn on a couple LEDs. This
would only verify position once every revolution. But it's the high
count indexing where people will be in a hurry so it may be the best
option.
Eric[/QUOTE]

Well, you could build a board with some micro on it, and preprogram
each one, for an investment of a few hundred dollars and a lot of your
time to program the thing, and just sell it turnkey.

For positional feedback, I'd seriously recommend linear encoders on
the bed itself - it totally eliminates slop. :)

What is it overall you're trying to accomplish?

Thanks,
Rich
 
J

jasen

Jan 1, 1970
0
Thanks Rich,
I was hoping for a single IC solution.

only single IC solution I can think of is some sort of microcontroller
that track the position of the input and dose its best top put the output in
the same place as fast as it can.
It looks like I'll have to
limit the speed mechanically. Maybe. The ultimate goal is to come up
with a design that amateur machinists can make. The cost needs to be
low and the electronics need to be simple.
I'm not averse to using a microcontroller myself but if I went this
route I'd have to supply pre-programmed microcontrollers.

I know there are some micros that can be just hooked up to a USB port (etc)
to program them.
 
E

Eric R Snow

Jan 1, 1970
0
Well, you could build a board with some micro on it, and preprogram
each one, for an investment of a few hundred dollars and a lot of your
time to program the thing, and just sell it turnkey.

For positional feedback, I'd seriously recommend linear encoders on
the bed itself - it totally eliminates slop. :)

What is it overall you're trying to accomplish?

Thanks,
Rich
Rich,
A rotary indexer. Use might be indexing a dial in order to engrave
equally spaced lines. Another use could be drilling a hole pattern.
With a powerful enough stepper it could even be used for milling arcs.
But the main point is that the high accuracy that the encoders and
steppers are made to enables precision indexing. The discrete steps
would be very useful for indexing gears when milling them. And this
high accuracy is available for cheap. The electronics in between the
encoder and stepper need to be cheap too.
ERS
 
Top