This would work but I'm not real keen on having a device to physically turn
a knob.
I've got a whacky idea that I'd like to run past the EE types in the group.
I have a large number of unused Opto 22 digital outputs
http://www.opto22.com/site/pr_details.aspx?cid=4&item=G4ODC5
For sake of argument, lets keep the numbers simple. Get four resistors, 100,
200, 400, 800 ohm in series in place of the pot. In parallel with each
resistor put an opto output. Then using base two, any resistance from 0 to
1500 ohms is possible. The switch time on these outs is 100 microseconds -
I'm thinking a pot to control power level on the plasma cutter couldn't
possibly notice the switch time. I don't know if these devices will work at
0 volt between the otpo 22 output connections.
Or is there another easy way to do a pot with a CNC control?
Karl
There are several semiconductor companies who make digital pots
and board level digital pots are available too.
Google "digital potentiometer".
Art
What you suggest would work if the pot is a two-terminal variable
resistance. If it is a three-terminal voltage divider you need to
understand and describe the possible CNC outputs and pot circuit more
thoroughly.
FETs have an ON resistance well below 1 Ohm without an offset at zero
volts.
http://en.wikipedia.org/wiki/Digital-to-analog_converter
jsw
Its 0 - 24 volt, 2K ohm pot. Don't know how to measure impedance, but i
assume high.
karl
If it's two terminal, then it is a Rheostat, not a Potentiometer.
> If it is a three-terminal voltage divider you need to
> understand and describe the possible CNC outputs and pot circuit more
> thoroughly.
>
> FETs have an ON resistance well below 1 Ohm without an offset at zero
> volts.
>
> http://en.wikipedia.org/wiki/Digital-to-analog_converter
>
> jsw
--
Lead free solder is Belgium's version of 'Hold my beer and watch this!'
Ya, see your point. Its three treminal. To do this, I'd have to double the
complexity and have resistors above and below the wiper. Pot turner looks
better now.
karl
> There are several semiconductor companies who make digital pots
> and board level digital pots are available too.
> Google "digital potentiometer".
> Art
I've done this search and only find ICs for circuit board type stuff. Not
something to just replace a pot in a machine. Maybe it exists but all the
descriptions are written in EE not english.
Karl
Disconnect the pot wiper wire and connect a milliamp meter in series,
then record the current at both end settings.
jsw
With all the electrical noise a plasma cutter generates I think you
are right.
jsw
Karl there are dc powered, motorized pots made by Alps and others for
remote volume controls. but you could use a R/C servo motor to turn the
pot, and a 555 timer IC to generate the PWM control signals. This was
common on the old C-band feed horns to select the polarity, and trim the
alignment as you changed channels.
Digital pots are not intended as drop in replacements. They are low
voltage, low power devices that usually need microprocessor control.
Any noise in the control wiring or supply rail corrupt the settings.
They were intended to replace pots inside a piece of equipment, to allow
ATE calibration of equipment.
Depending on the current required at that input, a DAC and OpAmp
might be usable. If the DAC is parallel input you could store several
preset levels, and select them by changing the status of the address
lines of a EEPROM or EPROM. A lower tech method would be the use of
diode arrays to control the DAC inputs.
BTW, we used DACs for volume controls on telemetry receivers. We fed
the low level audio to the reference pin, then used the output to feed
the audio amplifier.
There are motorized potentiometers like some stereo's use with remote volume
control. There is also a digital pot that acts like a motorized pot. I've
seen the equivalent circuits for digital pots, you could make an equivalent.
For example, make a 2K ohm resistance using 10ea 200 ohm resistors in
series, the voltage output depends on the junction you select with an analog
switch. It may be a better solution to use some sort of digital pot, if you
can't find a digital pot to work with 24V you could use an op-amp to change
the voltage you can get from the pot (or DAC for that mater) to the 0-24V
range.
Parallax has some digital pots on their website, complete with datasheets
and source code for the Basic Stamp.
But for CNC control, I would probably just use a 0-10V analog output, like I
do for my spindle speed, and amplify it to 0-24V. If you wanted you could
control your pot output with your S command.
RogerN
A lot of the stuff I've used (dc motor controls, vfd's) that has a pot
also has another way to control output, such as analog signal input
terminals. I suppose you've exhausted the possibility of such for
your plasma torch?
Pete Keillor
I'm terrible at circuit board work and useless at EE design, but I can have
kits done.
How about this one? My control can output 0-10 volt for analog control.
http://www.kitsrus.com/pdf/k102.pdf
And then get a servo motor.
Karl
You've ALMOST got it right. Google R/2R or "Resistor Ladder" to get
what you're looking for.
Awfully elaborate. Easier would be to run three wires from the box,
(usually a potentiometer has three terminals) to an interface that
contains multiple pots just like the original, and a bunch of relays
to switch one or the other into the circuit. How many 'states' do you
need to have the digital gizmo control? Fast-medium-slow-off?
If one terminal is ground, wire those all together and put a
double-pole relay on each channel. The digital gizmo just
selects one of the preset pots.
As you will undoubtedly find, 'digital' potentiometers have low
voltage
and current limits, and are best designed in at the factory.
>You've ALMOST got it right. Google R/2R or "Resistor Ladder" to get
>what you're looking for.
BINGO!
Just what I want to do.
Thanks
Karl
Errr.... maybe not. If the wiper of the pot is a variable voltage,
and feeds a high impedance circuit, then fine.
But, the wiper might be +24V and it proportions current in two legs,
or the wiper might be a current source into a low-Z input, or
the pot might be a nonlinear taper type... there's LOTS of
potentiometer circuits, besides the one that has +V on
one leg and GND on the other. DAC and amplifier is
the same as R/2R ladder in most cases (and it's easier
to wire up, with less parts).
It looks like it should work for you. That kit will handle up to
four servos, too. You would want it well shielded when using around
your welder or plasma cutter. If you do use it, ground the unused
inputs, and put a 1K resistor from each unused output to ground.
That would work.
You could also use a couple of your opto outputs to step a stepper
motor, but the RC servo is a better solution because it contains
position feedback -- so the pot will always be driven to the same
position for given analog voltage input from the controller.
What is the point of this "pot turner"? You seem to imply that you're
trying to control the cut current of a plasma cutter from the CNC, and
per the folks who make the plasma cutters, including the hi-def ones,
controlling the current during CNC cutting provides little to no
benefit.
The plasma cutters don't have such a control input (at least the typical
air plasma units don't), because there is no need for them.
Many units do have a control input for "arc on" and an output for "arc
good" and an arc voltage tap as standard equipment, because those are
the control signals required for CNC control, but arc current is not
needed as a dynamic control.