Hi all,
I am embarking on a project to "recreate" the sound of a 1990s Drumfire sound board. Somebody posted a plan on Yahoo. I noticed a couple of parts -- op amps with leading circles -- that Google displayed as OTA, operation transconductance amplifiers. I had never heard of that. The model in the circuit, the CA3080, is unobtanium but I found on Digikey a couple (very few) of replacement OTA parts that resembled the CA3080.
Was this OTA part a fad? What is its real-world use today?
thanks,
-- Douglas Beeson
ound board. Somebody posted a plan on Yahoo. I noticed a couple of parts -- op amps with leading circles -- that Google displayed as OTA, operation tr ansconductance amplifiers. I had never heard of that. The model in the circ uit, the CA3080, is unobtanium but I found on Digikey a couple (very few) o f replacement OTA parts that resembled the CA3080.
** The LM13700 dual OTA is still made:I've seen it used as a VCA in compressor /limiters and a few other voltage controlled jobs.
I still have a small quantity of RCA CA3280s ( improved dual 3080s) which a re similar but also obsolete now.
.... Phil
ound board. Somebody posted a plan on Yahoo. I noticed a couple of parts -- op amps with leading circles -- that Google displayed as OTA, operation tr ansconductance amplifiers. I had never heard of that. The model in the circ uit, the CA3080, is unobtanium but I found on Digikey a couple (very few) o f replacement OTA parts that resembled the CA3080.
The OTA is an important amplifier type. CA3080 was retired due to technolo gy changes, but LM13700, NE5517, are similar; OPA861, MAX435 are upgrades.
Hans Camenzind's book has a chapter (Ch. 10) on these; you can download it from
Big benefits: program current sets the gain ...and the power dissipation near rail-to-rail output range you can wire-add the outputs differential inputs output is a controlled current source/sink with high compliance
program current sets the gain
Big problems: Current sets the bandwidth (from glacial to merely very slow) Horrible nonlinearity unless input diodes are used (the Gilbert configurati on) Low and nonlinear input impedance when input diodes are used
OTAs in theory can be pretty useful, but the ones still made are either slo w as molasses (LM13700) or not true OTAs (OPA860). About 25 years ago, you could get VT713s (from VTC) which were a good 100 times faster than the LM1
3700. I used to use those sometimes. I've considered designing in the LM136 00/LM13700 several times, but it never seems to make the cut. The OPA860 is more of an improved BJT than an OTA, despite the title on the datasheet.Also sometimes important: the current path goes through cascaded current mi rrors without emitter degeneration, so the output current has several times full shot noise.
Cheers
Phil Hobbs
They were a minor fad for a while. They weren't used much and are rarely used now.
-- John Larkin Highland Technology, Inc lunatic fringe electronics
As Phil notes, it was used in a lot of voltage-controlled circuits. In particular, a drum synth uses a rapidly-changing pitch to give that distinctive "drum-like" sound. OTAs would be used for the frequency sweep with a VCO, and also as a voltage-controlled filter (VCF) acting on a white noise source for snare sounds. In addition, the overall amplitude envelope of the sound would use an OTA as a voltage-controlled amplifier (VCA). All of these would be controlled by envelope generators and/or the strike detector.
Since we're talking about the creation of sound and not reproduction, some of the limitations of OTAs for other audio applications don't really apply here.
If you don't care about creating an exact clone of the Drumfire, note that (IMHO, anyway) you can get pretty good drum sounds from pretty cheesy circuits. (PAIA used biased diodes for voltage controls in a lot of their stuff... can't get much cheesier than that, but they were really popular.)
Best regards,
Bob Masta DAQARTA v9.00 Data AcQuisition And Real-Time Analysis
and one other Temperature-dependent gain
The 'nonlinearity' is large-signal distortion; that's avoidable.
Hanbury-Brown's correlator used the same input circuitry, potentially had the same nonlinearity.
That too. Didn't want to appear too negative all at once. ;)
Getting anywhere near the maximum slew rate of those very very slow parts requires running large input error voltages, though.
But his got rid of the input offset by commutating the output of the multiplier, and his incoming SNR was so low that the extra noise caused by running the transistors at a very low level wasn't a problem.
I really like OTAs in principle. For instance, back in the late '80s I built a PLL with a 100:1 frequency range, whose damping was constant and whose loop bandwidth was proportional to the input frequency. It used a charge pump to supply the bias current to OTA active filters, and a VCO with an exponential V->F characteristic. (I was going to build a 1 Hz
- 50-MHz analogue lock-in.)
If you could still get VT713s, I'd probably use some, but not LM13700s or OPA860s.
Cheers
Phil Hobbs
-- Dr Philip C D Hobbs Principal Consultant ElectroOptical Innovations LLC Optics, Electro-optics, Photonics, Analog Electronics 160 North State Road #203 Briarcliff Manor NY 10510 hobbs at electrooptical dot net http://electrooptical.net
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.