Re: Jog wheel

["marc" <dfs-at-xmission.com>]

----- Original Message -----
From: "William Scalione" <scalion-at-dellepro.com>
To: "Multiple recipients of list" <emc-at-nist.gov>
Sent: Monday, June 03, 2002 3:49 PM
Subject: Re: Jog wheel

It does not have to be that big of deal.
Less than 5 bucks will get you a "PIC" or Atmel uController that would
easily
do this. All you'd need to do is add a MAX232 TTL<->RS-232 Level converter
for $1.50 and you'd be set add all the switches indicators and handwheels
you
want quite cheaply.

As to the rotary encoders I have simply printed absolute position Gray Code
encoders disks with 512 positions/rev.

A 100 count incremental would be a no-brainer.

It is possible as-is to read a relative position from a stepper motor
windings
by watching the pulses from the windings as you turn it.

100-200-400 steps / rev are common.

All of the ones I have have at least a nub of a shaft on the back, I could
easily attach
the disk to or the front shaft long enough to mount the disk between the
motor body
and the face of the control and leave plenty of shaft to connect the knob.

As to the detent strength, simply energize a set of windings.
Take a 2 pole 200 step motor connect plain old DC power to a pair of
windings,
adjust current to get the "Feel" you want.
The 1.5" square stepper I have in my hand is rated 4V .95 Amp.
4 Volts is too tight for me, 1V -at- .1 Amp feels about right to me.
All it would take is a variable resistor mounted on the board to adjust for
feel.

BTW, take a stepper apart, and it will demagnetize and be junk as a stepper
at least.
>
> >
> > I'm way out of my league here and others are welcome to trash this idea
> but
> > it seems to me that rs232 serial ports are going to be around for a
while
> > with linux.  If we write the code to connect a serial port into iosh we
> > should be able to make almost any kind of device we want out there and
> > connect it with a rather simple cable.  USB powered would be easier but
we
> > can hang a little connector alongside the db9 to carry power to the
> device.
>
> US Digital has a quadrature to RS232 adapter, but it's $125.00 each.
> Probably
> just a LS7166 chip, a uart, max232, and some glue logic. If someone can
come
> up
> with a circuit, I will hardwire a few together and send them off to you
> software
> experts. They also have a USB model that will do 4 encoders and some
digital
> inputs, but it's pricey. Then there is also the Dan Mauch Encoder board
that
> would
> work if you are not using it for encoder feedback, but that is ISA bus.
> Probably
> not the best for future expansion.
>
>
> > > a single button to step through the axis
> > > select and one to enable. The axis select/enable button would
> > > be a way of ensuring this is the axis I want to move.
> >
> > I like this idea if the enable could be under or alongside one of the
> fingers
> > that you use to hold the device.  One nightmare of mine used a 1" or so
> > diameter tube about 4 inches long.  The wire would exit one end and the
> > handwheel would cover the other end.  Most of the handwheel would be
> covered
> > around the outer edge so that you could thumb it from one side but not
run
> > the wheel on your arm or leg.  The top of the wheel would have a
> depression
> > so that you could spin it with one finger of your other hand.  That way
> you
> > could use it as either a traditional pendent, or one hand could hold the
> > cylinder and thumb the wheel slowly while your other hand slid a paper
> that
> > you are using as a gage to touch off a part.
> >
> > I went looking once for small 100 tooth gears that I could use for a
> detent
> > but couldn't find any.  But if we turn out the knobs ourselves from AL
we
> > should be able to center drill or mill100 detents around the outer edge
of
> > the bottom and use a spring and ball from below.  Writing a g-code
program
> to
> > do this would be a good task for CP-1.
>
> I did the same today, and found some NEMA 17 stepper motors with 3.6
degree
> per step. None with dual shafts to mount an encoder to the back side, and
> probably
> not much detent torque (is that the proper term?) either. But if you took
> the rotor out
> would it have 100 teeth on it?

Marc Christensen