Audio circuit to drive solenoid

[Doug B.]

Hello all,

I am looking for some design help on a circuit to control a rotary
solenoid using an audio signal. First, some context:

This is for a musician/artist friend who is building a prototype of a
new musical instrument. The instrument consists of a series of nitinol
(MuscleWire) strings that are plucked by a wooden stick attached to a
rotary solenoid. He currently has eight strings+solenoid combinations
going. When it's played, the instrument sounds like a chorus of
detuned cats...

To drive a solenoid under his current setup, my friend simply connects
it across the positive and negative terminals of one channel of a car
stereo amp. He uses either an analog synthesizer or a digital music
application on his computer to generate pulse trains of low-frequency
signals. I guess you could call it a primitive form of PWM. He
manually tweaks the signal shape, period and volume until he gets the
solenoids to activate and pluck the strings at the rates he wants.

All this works (remarkably) well except for two problems: 1) The
solenoids get really hot after a short time. No wonder, as they spend
a good part of the time uselessly activated. I haven't actually hooked
up a scope, but I suspect that only the first 10-20 milliseconds of
the audio pulses are going to actually activate the solenoid; the rest
of the energy is going to heat his apartment. 2) Each solenoid takes
up a precious channel of an amp that is rated to 50 watts while actual
power needs are probably only a watt or two.

So what I'd like to do is design a circuit that would take the line-
level audio output from his synth or soundcard (1 V rms on the Macbook
he uses) and use it to trigger a pulse that is just long enough to
drive the solenoid. I don't have the specs on the solenoid, but I
believe it needs about 10 milliseconds for activation, then another 60
- 100 ms to run its course to 45 degrees. The solenoid is rated at 12
volts, but I'm not sure whether that is the 100% duty cycle rating or
something less. In any case, my friend uses 13.8 V regulated power
supplies for his setups, so I'd like to use that to power the circuit.

My first idea for this was to half-wave rectify the audio signal and
use it to drive a power transistor, but I don't see how to limit the
resulting pulse length, then shut off, regardless of how long the
audio note plays. RC circuit maybe?

I welcome all ideas, great and small. Thanks!

doug beeson
montreal,canada

 

[Tim Williams]

So you're looking for a circuit which produces a one-shot pulse, of specific
(and presumably adjustable) amplitude(s) and timing(s)?

If he's truely PWMing the solenoid, it could be getting hot from the AC.
Solenoids are often big hunks of steel, and big hunks of steel get hot when
you put AC into them.

Since this is a musical instrument, I'm going to bet it would be much more
convienient to implement a MIDI interface. Not something I've done, but you
can probably find a chip to do the hard part.

Tim

 

[J.A. Legris]

So you're looking for a circuit which produces a one-shot pulse, of specific
(and presumably adjustable) amplitude(s) and timing(s)?

If he's truely PWMing the solenoid, it could be getting hot from the AC.
Solenoids are often big hunks of steel, and big hunks of steel get hot when
you put AC into them.

Since this is a musical instrument, I'm going to bet it would be much more
convienient to implement a MIDI interface.  Not something I've done, but you
can probably find a chip to do the hard part.

Tim

Good idea. You can get some inspiration at the following web page that
describes a MIDI-controlled trumpet interface.

http://www.logosfoundation.org/instrum_gwr/korn.html

Regarding the overheating: for continuous sting-plucking your
solenoids should be rated for continuous duty.

 

[Doug B.]

So you're looking for a circuit which produces a one-shot pulse, of specific
(and presumably adjustable) amplitude(s) and timing(s)?

Yes, I think that sums it up. I know of the classic 555 for one-shot
pulses. Is there an alternative?

If he's truely PWMing the solenoid, it could be getting hot from the AC.
Solenoids are often big hunks of steel, and big hunks of steel get hot when
you put AC into them.

Since this is a musical instrument, I'm going to bet it would be much more
convienient to implement a MIDI interface. Not something I've done, but you
can probably find a chip to do the hard part.

Tim

I have thought of a MIDI interface. Problem is, my friend is really
not interested in using anything new than his standard analog
interface so...not this time.

 

[J.A. Legris]

Yes, I think that sums it up. I know of the classic 555 for one-shot
pulses. Is there an alternative?




I have thought of a MIDI interface. Problem is, my friend is really
not interested in using anything new than his standard analog
interface so...not this time.

Maybe he needs some convincing. For example, it's a simple matter to
extract the velocity parameter from each MIDI note-on message, which
can be used to vary the intensity or repetition rate of your string
pluckers, i.e. dynamic modulation. Come to think of it, the present
technique of driving solenoids with the audio amps is probably already
providing dynamic response. Your friend may be disappointed by the
flat response of one-shots.

 

[RFI-EMI-GUY]

Hello all,

I am looking for some design help on a circuit to control a rotary
solenoid using an audio signal. First, some context:

This is for a musician/artist friend who is building a prototype of a
new musical instrument. The instrument consists of a series of nitinol
(MuscleWire) strings that are plucked by a wooden stick attached to a
rotary solenoid. He currently has eight strings+solenoid combinations
going. When it's played, the instrument sounds like a chorus of
detuned cats...

To drive a solenoid under his current setup, my friend simply connects
it across the positive and negative terminals of one channel of a car
stereo amp. He uses either an analog synthesizer or a digital music
application on his computer to generate pulse trains of low-frequency
signals. I guess you could call it a primitive form of PWM. He
manually tweaks the signal shape, period and volume until he gets the
solenoids to activate and pluck the strings at the rates he wants.

All this works (remarkably) well except for two problems: 1) The
solenoids get really hot after a short time. No wonder, as they spend
a good part of the time uselessly activated. I haven't actually hooked
up a scope, but I suspect that only the first 10-20 milliseconds of
the audio pulses are going to actually activate the solenoid; the rest
of the energy is going to heat his apartment. 2) Each solenoid takes
up a precious channel of an amp that is rated to 50 watts while actual
power needs are probably only a watt or two.

So what I'd like to do is design a circuit that would take the line-
level audio output from his synth or soundcard (1 V rms on the Macbook
he uses) and use it to trigger a pulse that is just long enough to
drive the solenoid. I don't have the specs on the solenoid, but I
believe it needs about 10 milliseconds for activation, then another 60
- 100 ms to run its course to 45 degrees. The solenoid is rated at 12
volts, but I'm not sure whether that is the 100% duty cycle rating or
something less. In any case, my friend uses 13.8 V regulated power
supplies for his setups, so I'd like to use that to power the circuit.

My first idea for this was to half-wave rectify the audio signal and
use it to drive a power transistor, but I don't see how to limit the
resulting pulse length, then shut off, regardless of how long the
audio note plays. RC circuit maybe?

I welcome all ideas, great and small. Thanks!

doug beeson
montreal,canada

I think using a rectifier bridge to create a DC solenoid bias is the
first order solution, then return the solenoid to ground through an FET
modulated by a 555 timer or other circuit to lower the duty cycle.


--
Joe Leikhim K4SAT
"The RFI-EMI-GUY"©

"Treason doth never prosper: what's the reason?
For if it prosper, none dare call it treason."

"Follow The Money" ;-P

 

[Doug B.]

I think using a rectifier bridge to create a DC solenoid bias is the
first order solution, then return the solenoid to ground through an FET
modulated by a 555 timer or other circuit to lower the duty cycle.

--
Joe Leikhim K4SAT
"The RFI-EMI-GUY"©

"Treason doth never prosper: what's the reason?
For if it prosper, none dare call it treason."

"Follow The Money" ;-P

Thank you for your suggestion.

When you say "a DC solenoid bias", do you mean using the audio signal
to switch on the solenoid? If so, what would you recommend to do that?
FET, Darlington, something else? Also, I presume that I could
simultaneously start the charge cycle on the 555 so that it will shut
down the solenoid at the end of the RC period, correct?

doug beeson

 

[GregS]

Thank you for your suggestion.

When you say "a DC solenoid bias", do you mean using the audio signal
to switch on the solenoid? If so, what would you recommend to do that?
FET, Darlington, something else? Also, I presume that I could
simultaneously start the charge cycle on the 555 so that it will shut
down the solenoid at the end of the RC period, correct?

doug beeson

I am going to do a similar thing but with a sliding mechanism, using a gearmotor.
I will send a control TTL pulse through a TA7288P then using a 555 to cutout the power
to the motor using a Mosfet. You could use a comparator driven with the rectified
audio to drive a Mosfet for the solenoid. You might be able to slip another "and"
control for the timer.


greg

 

[JosephKK]

Hello all,

I am looking for some design help on a circuit to control a rotary
solenoid using an audio signal. First, some context:

This is for a musician/artist friend who is building a prototype of a
new musical instrument. The instrument consists of a series of nitinol
(MuscleWire) strings that are plucked by a wooden stick attached to a
rotary solenoid. He currently has eight strings+solenoid combinations
going. When it's played, the instrument sounds like a chorus of
detuned cats...

To drive a solenoid under his current setup, my friend simply connects
it across the positive and negative terminals of one channel of a car
stereo amp. He uses either an analog synthesizer or a digital music
application on his computer to generate pulse trains of low-frequency
signals. I guess you could call it a primitive form of PWM. He
manually tweaks the signal shape, period and volume until he gets the
solenoids to activate and pluck the strings at the rates he wants.

All this works (remarkably) well except for two problems: 1) The
solenoids get really hot after a short time. No wonder, as they spend
a good part of the time uselessly activated. I haven't actually hooked
up a scope, but I suspect that only the first 10-20 milliseconds of
the audio pulses are going to actually activate the solenoid; the rest
of the energy is going to heat his apartment. 2) Each solenoid takes
up a precious channel of an amp that is rated to 50 watts while actual
power needs are probably only a watt or two.

So what I'd like to do is design a circuit that would take the line-
level audio output from his synth or soundcard (1 V rms on the Macbook
he uses) and use it to trigger a pulse that is just long enough to
drive the solenoid. I don't have the specs on the solenoid, but I
believe it needs about 10 milliseconds for activation, then another 60
- 100 ms to run its course to 45 degrees. The solenoid is rated at 12
volts, but I'm not sure whether that is the 100% duty cycle rating or
something less. In any case, my friend uses 13.8 V regulated power
supplies for his setups, so I'd like to use that to power the circuit.

My first idea for this was to half-wave rectify the audio signal and
use it to drive a power transistor, but I don't see how to limit the
resulting pulse length, then shut off, regardless of how long the
audio note plays. RC circuit maybe?

I welcome all ideas, great and small. Thanks!

doug beeson
montreal,canada

On the face of it, i recommend switching the interface to the
instrument to MIDI. It just seems so much more natural for the
interface to me.

 

[JosephKK]

Yes, I think that sums it up. I know of the classic 555 for one-shot
pulses. Is there an alternative?




I have thought of a MIDI interface. Problem is, my friend is really
not interested in using anything new than his standard analog
interface so...not this time.

Crikey. Point out your friend that the sound card already has a
duplex midi interface.

 

[RFI-EMI-GUY]

Thank you for your suggestion.

When you say "a DC solenoid bias", do you mean using the audio signal
to switch on the solenoid?

Yes; rectify the AC and apply the DC component to the positive side of
the solenoid.

If so, what would you recommend to do that?
FET, Darlington, something else?

You can use a full wave diode bridge rectifier.

Also, I presume that I could

simultaneously start the charge cycle on the 555 so that it will shut
down the solenoid at the end of the RC period, correct?

If the use of DC bias alone does not solve your heating problem, you
could use a free running 555 timer to produce a 50% duty cycle square
wave at some multiple of the highest modulating frequency. This timer
could drive a Mosfet switch on the negative end of the solenoid.

Make sure all of this is filtered well on the supply lines so that
nothing feeds back into the ammplifiers.

doug beeson

--
Joe Leikhim K4SAT
"The RFI-EMI-GUY"©

"Treason doth never prosper: what's the reason?
For if it prosper, none dare call it treason."

"Follow The Money" ;-P

 

[[email protected]]

Hello all,

I am looking for some design help on a circuit to control a rotary
solenoid using an audio signal. First, some context:

This is for a musician/artist friend who is building a prototype of a
new musical instrument. The instrument consists of a series of nitinol
(MuscleWire) strings that are plucked by a wooden stick attached to a
rotary solenoid. He currently has eight strings+solenoid combinations
going. When it's played, the instrument sounds like a chorus of
detuned cats...

To drive a solenoid under his current setup, my friend simply connects
it across the positive and negative terminals of one channel of a car
stereo amp. He uses either an analog synthesizer or a digital music
application on his computer to generate pulse trains of low-frequency
signals. I guess you could call it a primitive form of PWM. He
manually tweaks the signal shape, period and volume until he gets the
solenoids to activate and pluck the strings at the rates he wants.

All this works (remarkably) well except for two problems: 1) The
solenoids get really hot after a short time. No wonder, as they spend
a good part of the time uselessly activated. I haven't actually hooked
up a scope, but I suspect that only the first 10-20 milliseconds of
the audio pulses are going to actually activate the solenoid; the rest
of the energy is going to heat his apartment. 2) Each solenoid takes
up a precious channel of an amp that is rated to 50 watts while actual
power needs are probably only a watt or two.

So what I'd like to do is design a circuit that would take the line-
level audio output from his synth or soundcard (1 V rms on the Macbook
he uses) and use it to trigger a pulse that is just long enough to
drive the solenoid. I don't have the specs on the solenoid, but I
believe it needs about 10 milliseconds for activation, then another 60
- 100 ms to run its course to 45 degrees. The solenoid is rated at 12
volts, but I'm not sure whether that is the 100% duty cycle rating or
something less. In any case, my friend uses 13.8 V regulated power
supplies for his setups, so I'd like to use that to power the circuit.

My first idea for this was to half-wave rectify the audio signal and
use it to drive a power transistor, but I don't see how to limit the
resulting pulse length, then shut off, regardless of how long the
audio note plays. RC circuit maybe?

I welcome all ideas, great and small. Thanks!

doug beeson
montreal,canada

I might have the solution you're looking for. Please contact me.