Phototransistor + Base???

[phector2004]

Hi guys,

So I've been experimenting with some of the random components I bought a while back and plugged an LTV4N35 into my breadboard.

Basically, this is a 6-pin photocoupler IC with the following pinout:

1: LED+
2: LED-
3: NC

4: Emitter
5: Collector
6: Base

What I'm wondering is what purpose the base has in a photocoupler?

I've set up an LPB-1 driving the LED portion, and set the phototransistor with the same resistor values as the LPB-1. I have no problems getting sound out of the chip without any signal entering the base, but when I tried taking off the other base-connected components (i.e. 1M transistor from 9V to B), it shut off completely.

Does it play a role in biasing the transistor? What else is it involved in?

[Earthscum]

I have a H11AA1, same pin assignment, but the LED is actually a pair, back to back like what we use for clippers... convenient.

From what I've gathered (I haven't really tried this chip yet), you basically think of the base as you normally would, except that whatever "signal" that is coming in is being transmitted through the LED to the base. This means, essentially, that you can set it up as a pulse trigger, amplifier, etc, etc. I was looking at trying mine in a compressor setup, since I get the same action no matter whether my signal is going positive or negative. Then I should be able to put filtering on the base to smooth out ripple. It's an idea I've been stewing in my head for awhile.

[Hides-His-Eyes]

Could phototransistors be used for vactrol-like purposes? E.g modulation, compression..?

[jasperoosthoek]

They could be used that way but it is not straight forward. Only where you could use a normal bipolar in such a circuit they can replace the bipolar. But as a simple variable resistor they don't work. The collector will always be a diode.

What you can do is connect them to an OTA chip such as a CA3080. That can be done in more ways.

[Hides-His-Eyes]

Is anyone still making OTAs?

[jasperoosthoek]

AFAIK they are readily available. Just go to small bear's shop.

[phector2004]

I have a H11AA1, same pin assignment, but the LED is actually a pair, back to back like what we use for clippers... convenient.

From what I've gathered (I haven't really tried this chip yet), you basically think of the base as you normally would, except that whatever "signal" that is coming in is being transmitted through the LED to the base. This means, essentially, that you can set it up as a pulse trigger, amplifier, etc, etc. I was looking at trying mine in a compressor setup, since I get the same action no matter whether my signal is going positive or negative. Then I should be able to put filtering on the base to smooth out ripple. It's an idea I've been stewing in my head for awhile.

Thanks for the info!

I don't know if this is bad design, but I didn't put a DC-blocking cap between the LPB-1 and the LED to get the entire signal through the LED. Seems like your chip is more convenient though!

Cheers

Could phototransistors be used for vactrol-like purposes? E.g modulation, compression..?

I've gotten some compression with my current setup, but nothing near an actual compression pedal. You could probably substitute this chip in place of a vactrol, but not as straight forward as making your own!

[Earthscum]

I remember now... the best way I could find to use these is to think of envelope followers that have a half-wave recti feeding a transistor that takes care of the filtering. Then you use the "control voltage" output to control your fet. I did hook it up (I found my notes on it... they sucked back then, lol). I had apparently confirmed using it in a feedback loop on a TL082, but had it set up similar to a muff stage with a 10u cap to allow clipping of all signal. You could easily adjust what signals get "clipped" within the feedback of the Op amp. If I can hunt down that darn chip, I'm gonna board it up again. I didn't have anything in my notes about filtering, so I assume it wasn't a spectacular result due to my lack of knowledge at the time of signal conditioning.

BTW, Electronic Goldmine and Jameco both carry the H11AA1. Jameco has the cheaper one at $0.59 compared to EG's $0.75.

[Hides-His-Eyes]

So simplest idea; connect the LED to an LFO, use the transistor as a buffer; would you have a tremolo?

[jasperoosthoek]

That wouldn't work. All you get is more DC current in the transistor, very similar to leakage current. It does not change the gain/hfe of the transistor. The gain of (clean) transistor circuits is normally set by the components around it. High and low gain transistors will sounds almost the same.

[Hides-His-Eyes]

Makes sense. Ah well, now I know why nobody uses 'em.

[amptramp]

On military equipment that had to retain functionality, I always used to connect the phototransistor base to the emitter through a 750K resistor.  That guaranteed that the leakage current at high temperatures would not turn the transistor on.  It also reduced the sensitivity to LED current where driver leakage was an issue.  Besides, a floating base pin is just asking to pick up interference unless there is some resistance to the emitter.

[mac]

I tried to use a 4n27 as in the Rock Face... noise, RF, etc
I connected the base to gnd, vcc with resistors or caps...
I couldn't get it to work properly in the circuit I was testing, which is more unstable than a Fuzz Face.

mac

[amptramp]

Connecting the base directly to gound will shut it off, or more correctly, turn the phototransistor into a photodiode with much less current gain from LED to output..  A large value resistor will keep it operating.