Maker Pro
Maker Pro

Piezo Pulse Amplifier

J

J. McMillan

Jan 1, 1970
0
I am working on a system that will drive a piezoelectric inkjet head.
A microcontroller and a digital to analog converter is being used to
produce a controlled pulse shape (a specific rise time, hold time, and
fall time).

This pulse then needs to be amplified from 5 volts to the 15 to 30 volt
range. The pulses will have rise and fall times in the 1 us to 20 us
range.

I have seen a solution that uses the Apex line of high voltage/high slew
rate opamps but I think those might be overkill and I am looking for
something cheaper.

Does anyone have any implementation suggestions?

Sincerely,
James McMillan
 
R

Robert Baer

Jan 1, 1970
0
J. McMillan said:
I am working on a system that will drive a piezoelectric inkjet head.
A microcontroller and a digital to analog converter is being used to
produce a controlled pulse shape (a specific rise time, hold time, and
fall time).

This pulse then needs to be amplified from 5 volts to the 15 to 30 volt
range. The pulses will have rise and fall times in the 1 us to 20 us
range.

I have seen a solution that uses the Apex line of high voltage/high slew
rate opamps but I think those might be overkill and I am looking for
something cheaper.

Does anyone have any implementation suggestions?

Sincerely,
James McMillan

Why not a direct drive system. where the driver is turned on and off,
the supply voltage being what you need, and use series and shunt
resistors to control rise and fall times (the piezoelectric head being
the capacitance)?
If need, parallel a capacitor to the head.
 
B

Bill Sloman

Jan 1, 1970
0
J. McMillan said:
I am working on a system that will drive a piezoelectric inkjet head.
A microcontroller and a digital to analog converter is being used to
produce a controlled pulse shape (a specific rise time, hold time, and
fall time).

This pulse then needs to be amplified from 5 volts to the 15 to 30 volt
range. The pulses will have rise and fall times in the 1 us to 20 us
range.

I have seen a solution that uses the Apex line of high voltage/high slew
rate opamps but I think those might be overkill and I am looking for
something cheaper.

Does anyone have any implementation suggestions?

Check out the Texas Instruments (was Burr-Brown) OPA552

http://focus.ti.com/lit/ds/symlink/opa551.pdf

It seems to offer the voltage swing you need, and slew rate
(24V/usec).

It isn't stable for closed loop gains of less than five, but you seem
to need a gain of six, so that shouldn't be a problem. Farnell list
them for about $5 in small quantities, which seems reasonable
(certainly wehn comapred with Apex prices).
 
J

James Meyer

Jan 1, 1970
0
I am working on a system that will drive a piezoelectric inkjet head.
A microcontroller and a digital to analog converter is being used to
produce a controlled pulse shape (a specific rise time, hold time, and
fall time).

This pulse then needs to be amplified from 5 volts to the 15 to 30 volt
range. The pulses will have rise and fall times in the 1 us to 20 us
range.

I have seen a solution that uses the Apex line of high voltage/high slew
rate opamps but I think those might be overkill and I am looking for
something cheaper.

Does anyone have any implementation suggestions?

Sincerely,
James McMillan

I'd be tempted to look at adding a single transistor, emitter follower
with gain, to the output of almost any fast op-amp and closing the feedback loop
around the whole thing. You did say the rise and fall times are already being
generated by the micro, didn't you?

Jim
 
Top