OPAMP ASTABLE

John Larkin · 2013-09-10T05:29:06+00:00

I was thinking about building a small, low power dc/dc converter, +5 to maybe -25 volts, to bias a photodiode. I only need maybe 0.5 mA, but I want the converter to be very quiet, which means no sharp switching edges or shoot-throughs. After futzing with a few ideas, I figured that a pair of r-r opamps, with reasonable slew rates, could do it. So, the opamp astable multivibrator. The complementary squarish waves are net 10 volts p-p, so could drive a 3:1 stepup transformer into a full-wave doubler or something. The opamp inputs are swinging past the rails, so I don't know if the opamp model is correct. You'd expect ESD diodes or something. Anyhow, that can be fixed. Version 4 SHEET 1 880 680 WIRE 176 -96 80 -96 WIRE 352 -96 256 -96 WIRE 240 -32 208 -32 WIRE 256 -32 240 -32 WIRE 208 0 208 -32 WIRE 80 16 80 -96 WIRE 176 16 80 16 WIRE 352 32 352 -96 WIRE 352 32 240 32 WIRE 144 48 128 48 WIRE 176 48 144 48 WIRE 208 80 208 64 WIRE 672 96 544 96 WIRE 720 96 672 96 WIRE 752 96 720 96 WIRE 672 128 672 96 WIRE 176 144 128 144 WIRE 352 144 352 32 WIRE 352 144 240 144 WIRE 544 176 544 96 WIRE 672 224 672 208 WIRE 720 224 672 224 WIRE 752 224 720 224 WIRE 672 256 672 224 WIRE 80 272 80 16 WIRE 176 272 80 272 WIRE 352 272 240 272 WIRE 544 304 544 256 WIRE 672 368 672 336 WIRE 128 384 128 144 WIRE 128 384 80 384 WIRE 176 384 128 384 WIRE 352 384 352 272 WIRE 352 384 256 384 WIRE 240 448 208 448 WIRE 256 448 240 448 WIRE 208 480 208 448 WIRE 80 496 80 384 WIRE 176 496 80 496 WIRE 352 512 352 384 WIRE 352 512 240 512 WIRE 144 528 128 528 WIRE 176 528 144 528 WIRE 208 560 208 544 FLAG 208 80 0 FLAG 208 560 0 FLAG 240 -32 +5 FLAG 240 448 +5 FLAG 144 48 V2 FLAG 144 528 V2 FLAG 544 304 0 FLAG 672 368 0 FLAG 720 96 +5 FLAG 720 224 V2 SYMBOL Opamps\LT1498 208 -32 R0 WINDOW 0 63 91 Left 2 WINDOW 3 38 123 Left 2 SYMATTR InstName U1 SYMBOL Opamps\LT1498 208 448 R0 WINDOW 0 68 91 Left 2 WINDOW 3 66 121 Left 2 SYMATTR InstName U2 SYMBOL res 272 -112 R90 WINDOW 0 -6 80 VBottom 2 WINDOW 3 -32...

G

George Herold 12 years ago

Oh! I didn't know such things existed. Hmm for ~$20 I could get a 100V supply. That might be fun... don't think of it as low current, it's a built in current limiter!

George H.

Vote

J

John Larkin 12 years ago

We'll never get along. Our definitions of "horrendous spikes" differ by 7 or 8 orders of magnitude.

John Larkin Highland Technology Inc www.highlandtechnology.com jlarkin at highlandtechnology dot com Precision electronic instrumentation Picosecond-resolution Digital Delay and Pulse generators Custom timing and laser controllers Photonics and fiberoptic TTL data links VME analog, thermocouple, LVDT, synchro, tachometer Multichannel arbitrary waveform generators

Vote

J

John Larkin 12 years ago

What sets the frequency?

John Larkin Highland Technology Inc www.highlandtechnology.com jlarkin at highlandtechnology dot com Precision electronic instrumentation Picosecond-resolution Digital Delay and Pulse generators Custom timing and laser controllers Photonics and fiberoptic TTL data links VME analog, thermocouple, LVDT, synchro, tachometer Multichannel arbitrary waveform generators

Vote

J

Jim Thompson 12 years ago

Depends on your definition of "quiet".

You "design" crap, I design elegance ;-) ...Jim Thompson

Vote

J

Jim Thompson 12 years ago

Those with half a brain ?>:-} ...Jim Thompson

Vote

B

bloggs.fredbloggs.fred 12 years ago

Frequency is 1/2 the output slew rate divided by the total output voltage excursion. Output slew rate is just Vin/(Ri x Ci) of the integrator and output voltage excursion is your choice.

Vote

J

John Larkin 12 years ago

Makes no sense to me. Got a numerical example? Or Spice model?

John Larkin Highland Technology Inc www.highlandtechnology.com jlarkin at highlandtechnology dot com Precision electronic instrumentation Picosecond-resolution Digital Delay and Pulse generators Custom timing and laser controllers Photonics and fiberoptic TTL data links VME analog, thermocouple, LVDT, synchro, tachometer Multichannel arbitrary waveform generators

Vote

K

Klaus Kragelund 12 years ago

Nice device. Could be used for driving isolated FET, with no need for auxiliary supply

Cheers

Klaus

Vote

K

Klaus Kragelund 12 years ago

The TLP3902 is a lot cheaper:

formatting link

Cheers

Klaus

Vote

G

George Herold 12 years ago

Yeah, that looks like the 'intended' application. I'll have to order a handful to play with.

George H.

Vote

J

John Larkin 12 years ago

I've done that; use my zero-standby-power opto-totem circuit, too.

formatting link

That only makes 5 uA. Probably why it's cheaper... less silicon.

John Larkin Highland Technology Inc www.highlandtechnology.com jlarkin at highlandtechnology dot com Precision electronic instrumentation Picosecond-resolution Digital Delay and Pulse generators Custom timing and laser controllers Photonics and fiberoptic TTL data links VME analog, thermocouple, LVDT, synchro, tachometer Multichannel arbitrary waveform generators

Vote

G

George Herold 12 years ago

(There's also no stock in the US)

Hey, the TLP590B is a dip! I ordered 10 from mouser. I wonder what the noise is like on the output. With all the junctions in series I bet there's a bit of a temp coef. And then the shot noise will all add. (only in quadrature of course.)

George H.

Vote

J

Jim Thompson 12 years ago

I have none on-hand, but I wonder what you get if you take a 4N25, tie output device collector to emitter, call that cathode, and the base as anode... how much current do you get? ...Jim Thompson

Vote

G

George Herold 12 years ago

I've got a tube of those... and the answer is,

I LED I sc (mA) (uA) ________________

10 47 20 96 30 143 40 188

Hey if I un-short the collector and emitter I get a the same short circuit current out of one, but much less out of the other only 20uA for 10mA of LED current.

Care to guess, and then tell me why. If I hadn't done the measurement I would have guessed that I'd get less current out of each leg.

George H.

Vote

P

Phil Hobbs 12 years ago

Not a lot. The efficiency is roughly CTR/beta.

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

Vote

J

Jim Thompson 12 years ago

I'm no expert in photo-induced currents in junctions, but my understanding is that most of it is in the base-emitter path.

(Perhaps because that junction is closest to the surface, but I'm just SWAG-ing :-)

Load it with Larkin's 60K and see what voltage you get. ...Jim Thompson

Vote

J

John Larkin 12 years ago

A few ma of supply rail spike with a few us risetime is quiet. A few amps in a few ns is noisy.

This is horrible, amps of shoot-through current with sub-ns rise:

formatting link

You called 9 mA of slow spike "horrendous". Why?

I doubt that you could have designed any of my products. Well, maybe one of the trivial accessories.

John Larkin Highland Technology, Inc jlarkin at highlandtechnology dot com http://www.highlandtechnology.com Precision electronic instrumentation Picosecond-resolution Digital Delay and Pulse generators Custom laser drivers and controllers Photonics and fiberoptic TTL data links VME thermocouple, LVDT, synchro acquisition and simulation

Vote

J

Jim Thompson 12 years ago

There you go with the snot-nosed crap again. Why don't you grow up?

You expressly specified quiet, then posted a distinctly NOT-quiet schematic.

Could you design this...

Not a chance.

I'll be publishing the specifications, some 21 pages, once I cleanse it of any NDA issues.

I doubt you will be able to understand much more of it than the power supply consumption >:-} ...Jim Thompson

Vote

J

John Larkin 12 years ago

I tried a Cree T1-3/4 white LED and a Vishay VBPW34S photodiode. With

20 mA / 3 volts on the LED, carefully aligned, I get from the pd 0.52 volts open-circuit, 1.6 mA short-circuit current. That's a CTR of 8% and power transfer of, I'm guessing, around 1%.

John Larkin Highland Technology, Inc jlarkin at highlandtechnology dot com http://www.highlandtechnology.com Precision electronic instrumentation Picosecond-resolution Digital Delay and Pulse generators Custom laser drivers and controllers Photonics and fiberoptic TTL data links VME thermocouple, LVDT, synchro acquisition and simulation

Vote

P

Phil Hobbs 12 years ago

If you use a longer wavelength LED, you should do better for power transfer, and maybe even CTR.

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

Vote

OPAMP ASTABLE

Join the Discussion

Didn't find your answer?