Looking for a sanity check on a stepper driver circuit

[Don Bruder]

I've got a radio-controlled step-sequencer/driver built, tested, and
functional for driving a single steper motor.

As it sits, it controls one stepper motor - Call that one "Stepper A."
The four separate outputs of the sequencer are each as shown in the
"Driver" drawing below. Each connects to a "Phase #" line in the "Motor
side" drawing below that. The line of "========" in the "Motor side"
drawing is the demarcation between the existing circuit (above the line)
and proposed additions (below the line.)

My goal is to control a second stepper - Call this one "Stepper B" -
separately from the first. At any given moment, Stepper A might be
active, OR Stepper B might be active, or both might be INACTIVE, but
there will (or at least, according to plans, there *SHOULD*) never be a
time when both A and B are active simultaneously.

I've been doing some thinking, and it seems as though I should be
able to do what I'm looking for WITHOUT building a second sequencer
and the associated circuitry that goes with it. Existing circuitry
on the R/C board I'm already controlling "Stepper A" with can easily
provide me with an A and a B signal, to drive points A and B in the
"Proposed additions" section of the drawing.

What I'm looking for is someone to confirm or deny the sanity involved
in my thinking.

Anybody see anything blatantly wrong with my "brainstorm" as drawn
below? (ignore "cosmetics" such as lack of any back-EMF diodes in the
drawing - it's much simpler for me to draw it without them, with the
understanding that they are needed, and actually present, in the circuit)

I *AM* wondering if I ought to toss some "steering" diodes in between
Stepper A and Stepper B, say at the locations marked "D?" though...
Good idea, or unneccesary "overkill"?


Driver
(four separate circuits, all identical)

,------O Phase "X"
|E (To "Motor Side" phases, below)
B |
Control "X" O-----------K NPN (2N6387)
(from sequencer logic |C
via optocoupler) |
|
---
-



Motor side
Phase 1 O--+-------D?---UUU---, ---,
| | |
Phase 2 O--(-+-----D?---UUU---+ |
| | +------, |--- Stepper A
Phase 3 O--(-(-+---D?---UUU---+ | |
| | | | | |
Phase 4 O--(-(-(-+-D?---UUU---' | ---'
| | | | |
| | | | +------------O +12VDC
| | | | | Present stepper power connection
| | | | | (would be deleted for the
| | | | | second motor/selector additions)
| | | | |
========================================================================
| | | | |
| | | | |
| | | `------UUU---, | ---,
| | | | | |
| | `--------UUU---+ | |
| | +---, | |--- Stepper B
| `----------UUU---+ | | |
| | | | |
`------------UUU---' | | ---'
| |
| | Proposed motor "selector"
,-------------------' | and second stepper motor
| | additions
B |C |
B O----------K NPN (2N6387) |
|E |
| |
+12VDC O------+ |
|E |
B | |
A O----------K NPN (2N6387) |
|C |
| |
`----------------------'

 

[Chuck Harris]

I've got a radio-controlled step-sequencer/driver built, tested, and
functional for driving a single steper motor.

As it sits, it controls one stepper motor - Call that one "Stepper A."
The four separate outputs of the sequencer are each as shown in the
"Driver" drawing below. Each connects to a "Phase #" line in the "Motor
side" drawing below that. The line of "========" in the "Motor side"
drawing is the demarcation between the existing circuit (above the line)
and proposed additions (below the line.)

My goal is to control a second stepper - Call this one "Stepper B" -
separately from the first. At any given moment, Stepper A might be
active, OR Stepper B might be active, or both might be INACTIVE, but
there will (or at least, according to plans, there *SHOULD*) never be a
time when both A and B are active simultaneously.

I've been doing some thinking, and it seems as though I should be
able to do what I'm looking for WITHOUT building a second sequencer
and the associated circuitry that goes with it. Existing circuitry
on the R/C board I'm already controlling "Stepper A" with can easily
provide me with an A and a B signal, to drive points A and B in the
"Proposed additions" section of the drawing.

What I'm looking for is someone to confirm or deny the sanity involved
in my thinking.

Anybody see anything blatantly wrong with my "brainstorm" as drawn
below? (ignore "cosmetics" such as lack of any back-EMF diodes in the
drawing - it's much simpler for me to draw it without them, with the
understanding that they are needed, and actually present, in the circuit)

I *AM* wondering if I ought to toss some "steering" diodes in between
Stepper A and Stepper B, say at the locations marked "D?" though...
Good idea, or unneccesary "overkill"?


Driver
(four separate circuits, all identical)

,------O Phase "X"
|E (To "Motor Side" phases, below)
B |
Control "X" O-----------K NPN (2N6387)
(from sequencer logic |C
via optocoupler) |
|
---
-



Motor side
Phase 1 O--+-------D?---UUU---, ---,
| | |
Phase 2 O--(-+-----D?---UUU---+ |
| | +------, |--- Stepper A
Phase 3 O--(-(-+---D?---UUU---+ | |
| | | | | |
Phase 4 O--(-(-(-+-D?---UUU---' | ---'
| | | | |
| | | | +------------O +12VDC
| | | | | Present stepper power connection
| | | | | (would be deleted for the
| | | | | second motor/selector additions)
| | | | |
========================================================================
| | | | |
| | | | |
| | | `------UUU---, | ---,
| | | | | |
| | `--------UUU---+ | |
| | +---, | |--- Stepper B
| `----------UUU---+ | | |
| | | | |
`------------UUU---' | | ---'
| |
| | Proposed motor "selector"
,-------------------' | and second stepper motor
| | additions
B |C |
B O----------K NPN (2N6387) |
|E |
| |
+12VDC O------+ |
|E |
B | |
A O----------K NPN (2N6387) |
|C |
| |
`----------------------'

Well, it certainly isn't going to work with NPN transistors wired
as shown above.

 

[Jasen Betts]

I've got a radio-controlled step-sequencer/driver built, tested, and
functional for driving a single steper motor.

As it sits, it controls one stepper motor - Call that one "Stepper A."
The four separate outputs of the sequencer are each as shown in the
"Driver" drawing below. Each connects to a "Phase #" line in the "Motor
side" drawing below that. The line of "========" in the "Motor side"
drawing is the demarcation between the existing circuit (above the line)
and proposed additions (below the line.)

My goal is to control a second stepper - Call this one "Stepper B" -
separately from the first. At any given moment, Stepper A might be
active, OR Stepper B might be active, or both might be INACTIVE, but
there will (or at least, according to plans, there *SHOULD*) never be a
time when both A and B are active simultaneously.

I *AM* wondering if I ought to toss some "steering" diodes in between
Stepper A and Stepper B, say at the locations marked "D?" though...
Good idea, or unneccesary "overkill"?

you'll need them on both motors.

without them curent will flow forwards through half of the "unselected" motor
and backwards out the other half of it... then though the "undriven"
wuindings of the "powered" motor.

so your selector won't work, and torque will be reduced.

even with the diodes

you'll find that the deselected motor will have less torque than when it was
powered.

switching between motors when the sequencer is in a different configuration
than before the last switch could prove interesting too.

 

[default]

Steering diodes are necessary on both motors - all windings, and those
should be PNP transistors if you planning on switching +12 connected
to the emitters.

Other than that I don't see why it wouldn't work

 

[Don Bruder]

you'll need them on both motors.

without them curent will flow forwards through half of the "unselected" motor
and backwards out the other half of it... then though the "undriven"
wuindings of the "powered" motor.

I figured that would probably be the case. Diodes it is, then, and on
both motors.

so your selector won't work, and torque will be reduced.

even with the diodes

you'll find that the deselected motor will have less torque than when it was
powered.

I'm assuming you mean holding torque (Which is pretty much irrelevant to
my project)? Obviously, I'm going to lose <diode drop> worth of voltage
when a given coil is powered, but I can't see any other reason for loss
of power - assuming the coil *IS* powered.

switching between motors when the sequencer is in a different configuration
than before the last switch could prove interesting too.

On that topic, I already expect that I'll see a "twitch" as the newly
selected motor powers up and the mag-fields "line up" to match whatever
state the sequencer is in. This is non-critical to the project - In
fact, due to the combination of small step-size (both motors are 1.8
deg/step units) and the fairly high-ratio reduction gearheads on them
(According to "count the teeth and do the math", one is a 51.5:1
reduction, and the other is a 62.3:1 reduction - assuming I didn't
mis-count gear teeth) that "twitch" might not even be measurable without
resorting to much more precise tools than I have, or have any interest
in using - Worst case I can come up with is the "new" motor powering up
180 degrees out of phase from the current sequencer state, causing a 2
step jump as it synchs up - Barely noticable, and completely irrelevant
for the project.

 

[Don Bruder]

Well, it certainly isn't going to work with NPN transistors wired
as shown above.

AUGH! That's what I get for making the drawing from memory of assembling
the circuit... <sigh> OK, I've got my transistors wired backwards in the
drawing - Good catch. Other than that little brain-fart, any other
obvious problems?

 

[Don Bruder]

Steering diodes are necessary on both motors - all windings, and those

Which confirms what I already suspected - A second time, even

should be PNP transistors if you planning on switching +12 connected
to the emitters.

Yah... Somebody else caught that little brain-fart, too... Swap E and C
on all the transistors, and you've got what I actually meant.

Other than that I don't see why it wouldn't work

Great - I'm *NOT* totally crazy!

 

[default]

Yah... Somebody else caught that little brain-fart, too... Swap E and C
on all the transistors, and you've got what I actually meant.

Or use a PNP it will drop a lot less voltage as a high side switch.
NPN as a high side switch drops four volts and PNP drops point six.

 

[Jasen Betts]

I'm assuming you mean holding torque (Which is pretty much irrelevant to
my project)? Obviously, I'm going to lose <diode drop> worth of voltage
when a given coil is powered, but I can't see any other reason for loss
of power - assuming the coil *IS* powered.

yeah that's what I meant.

On that topic, I already expect that I'll see a "twitch" as the newly
selected motor powers up and the mag-fields "line up" to match whatever
state the sequencer is in.

that's ok then.

 

[Rich Grise]

I'm assuming you mean holding torque (Which is pretty much irrelevant to
my project)? Obviously, I'm going to lose <diode drop> worth of voltage
when a given coil is powered, but I can't see any other reason for loss
of power - assuming the coil *IS* powered.

If holding torque is not an issue, then just diode it up, and have fun!
This can be expanded to as many motors as you want, as long as you're sure
that you understand that you can only drive one at a time. ;-) I once
modified some other guy's Z80 stepper driver's timing loops, so instead of
going "CLACKETY-CLACKETY-CLACKETY etc," the actuators went,
"bvvvvvvvvvvvp." And the client and all of his minions went, "Ooh! Aah!"
It was very gratifying.

Cheers!
Rich

 

[Rich Grise]

Or use a PNP it will drop a lot less voltage as a high side switch.
NPN as a high side switch drops four volts and PNP drops point six.

I agree - I also wouldn't use an emitter follower in this app, because it
exposes the poor little E-B junction to the inductive kickback, which
with a stepper, is not a bad thing (the kickback, that is - the high E-B
reverse voltage is a bad thing in any transistor circuit), if you want any
speed out of it. ;-) (the higher you can let the spike go, the faster the
magnetic field at that pole will decay.) So, you protect your transistors
by giving them a high BVcbx and shunting them with zeners. (or tranzorbs,
if you're switching insanely high amounts of power. ;-) )

Good Luck!
Rich

 

[Rich Grise]

Or use a PNP it will drop a lot less voltage as a high side switch.
NPN as a high side switch drops four volts and PNP drops point six.

NPN high side switch drops point six volts, PNP drops point two or less. ;-)

Cheers!
Rich

 

[default]

NPN high side switch drops point six volts, PNP drops point two or less. ;-)

I think you have it backwards. the common type of self biased
collector input npn power supply requires a full four volts higher
across it to stay in regulation - a switch should do the same.

But just the base needs a greater voltage on it so one way to keep the
output or efficiency up is to return the base bias to a higher voltage
power supply than the collector input.