Maker Pro
Maker Pro

Stepper motor control with a single wire

M

mook johnson

Jan 1, 1970
0
Stepper motor gurus,

I have an application that requires stepper motor control with the following
constraints.
1) only a single wire w/return is available
2) must have some feedback or assurance that a step completed
(knowing/anticipating the motor position is required)
3) no local electronics near the stepper motor (transformers and possibly
diodes are allowed)


From what I know about steppers, they are pretty good about holding if you
keep the coil energized. They are relatively whimpy when moving from one
position to another. This would allow the motor to miss a step if there is
a load on the output.

any ideas on how the above can be accommodated?
 
R

Roger Hamlett

Jan 1, 1970
0
mook johnson said:
Stepper motor gurus,

I have an application that requires stepper motor control with the following
constraints.
1) only a single wire w/return is available
2) must have some feedback or assurance that a step completed
(knowing/anticipating the motor position is required)
3) no local electronics near the stepper motor (transformers and possibly
diodes are allowed)


From what I know about steppers, they are pretty good about holding if you
keep the coil energized. They are relatively whimpy when moving from one
position to another. This would allow the motor to miss a step if there is
a load on the output.

any ideas on how the above can be accommodated?
Basically, it can't.
To give intelligent feedback (which is implied by the fact that you are
going to have to send the 'feedback' along the same wire as the commands),
you would have to put some processing (even if just logic), at the motor.
At the minimum (even without any feedback), you will need logic to develop
to steps for the motor.

Best Wishes
 
I

Ian Stirling

Jan 1, 1970
0
mook johnson said:
Stepper motor gurus,

I have an application that requires stepper motor control with the following
constraints.
1) only a single wire w/return is available
2) must have some feedback or assurance that a step completed
(knowing/anticipating the motor position is required)
3) no local electronics near the stepper motor (transformers and possibly
diodes are allowed)

Why 3?
 
R

Rich Webb

Jan 1, 1970
0
Stepper motor gurus,

I have an application that requires stepper motor control with the following
constraints.
1) only a single wire w/return is available
2) must have some feedback or assurance that a step completed
(knowing/anticipating the motor position is required)
3) no local electronics near the stepper motor (transformers and possibly
diodes are allowed)

The obvious answer to "where's the stepper shaft?" is an off-the-shelf
encoder. http://www.usdigital.com/products/optical-encoders.shtml has a
representative selection. They're robust, they work, they're reasonably
priced.

It might help if you would go back a step and explain where the "only a
single wire w/return" (i.e., two wires?) requirement comes from.

A crude-but-effective method would be a variable resistor driven by the
shaft. Anticipate issues with resolution and noise.

You could also convert an AB quadrature encoder from the binary 00, 10,
11, 01 pattern to four voltage levels and squirt that down the one wire.
Could probably be done with "transformers and possibly diodes" (allowing
for both LEDs and photodiodes) but why?
 
R

Roger Hamlett

Jan 1, 1970
0
Rich Webb said:
The obvious answer to "where's the stepper shaft?" is an off-the-shelf
encoder. http://www.usdigital.com/products/optical-encoders.shtml has a
representative selection. They're robust, they work, they're reasonably
priced.

It might help if you would go back a step and explain where the "only a
single wire w/return" (i.e., two wires?) requirement comes from.

A crude-but-effective method would be a variable resistor driven by the
shaft. Anticipate issues with resolution and noise.

You could also convert an AB quadrature encoder from the binary 00, 10,
11, 01 pattern to four voltage levels and squirt that down the one wire.
Could probably be done with "transformers and possibly diodes" (allowing
for both LEDs and photodiodes) but why?
The encoder itself will contain electronics.
Generally, either an optical sensor, or hall effect switch. Even the purely
'passive' types, will require electronics at this end to decode the
position.
This is why I don't think the problem as stated, has a solution. However I
agree with you. The third requirement is 'silly', and with this removed, the
number of solutions rises to almost infinite. :)

Best Wishes
 
M

mook johnson

Jan 1, 1970
0
Stepper motor will be in a pressurized furnace where the temperature can
reach 350F. Only one electrical penetration is there and we want to
retrofit this motor into that system. Getting a high temperature motor or
transformer or diodes is not a big issue. Integrated circuits and
transistors are a reliability risk at those temperatures and pressures.


Ian Stirling said:
Greenbank
 
I

Ian Stirling

Jan 1, 1970
0
mook johnson said:
Stepper motor will be in a pressurized furnace where the temperature can
reach 350F. Only one electrical penetration is there and we want to
retrofit this motor into that system. Getting a high temperature motor or
transformer or diodes is not a big issue. Integrated circuits and
transistors are a reliability risk at those temperatures and pressures.

Why do you want the stepper motor?

Can you use the casing of the furnace as a ground?

With only two wires, it's annoying.
Take a number of series resonant circuits rectified with diodes.

You can do this with only diodes, capacitors and resistors.

To turn apply positive current to phase 1, apply 100Khz.
To apply negative, 150, positive phase 2 200, ...

This can in principle give a large number of controlled voltage sources.
Also, if you measure the impedance at each frequency, you can measure
the output of sensors connected similarly.
 
S

Spehro Pefhany

Jan 1, 1970
0
Ian Stirling said:

Suppose you had a unipolar single phase stepper, something along the
lines of a solenoid driving a pawl. Now consider a diode or a zener
in series with the coil, and a pot (with a series current-limiting
resistor) driven by the "stepper" shaft. The pot could parallel the
diode and you could determine the position by measuring the resistance
with a
relatively low voltage, and power the solenoid by applying a
relatively
high voltage.

Best regards,
Spehro Pefhany
 
N

N. Thornton

Jan 1, 1970
0
Basically, it can't.


Hi

I'll say it can be done, but whether its practical is another matter.
Here's an example method:

The motor is in series with an inductor inside the furnace. On the
motor shaft we have a large rotary multiposition switch, and spot
welded to that we have various resistors.

Now, for very short pulses the motor will be nearly o/c due to the
series inductor. Thus one can briefly measure those resistors and
determine the position.

I dont like the idea of a pack of switches in a furnace, but
technically its not impossible.


I think you'd do better to look at borehole electronics, as used down
oil drilling holes. That runs at high temp high pressure and is
reliable. Or easier, make another hole in your furnace.

Hmm, I dont suppose a bit of glass in the furnace wall is acceptable?
If so, one could read a patterned disc from the outside with a low
power laser.

Regards, NT
 
Top