CdS Photocells

My accurate fit of the R vs I curve on the Silonex datasheet yielded two exponents...

-0.930992 and -0.00573652

...Jim Thompson

This could just be just the way illumination and timing is specified.

Simple CDS sensors (dimmers etc) are roughly 50mS reducing R (on),

For coupled sensors, turn-on can be an order of magnitude less, while turn-off time may be x10 to x100 greater (depending on load impedance), as the actual illumination levels and sensitivity are removed from consideration or control.

Dark current readings usually demand >5sec darkness, before measurement.

RL

Donati's "Photodetectors" has a section (5.6) on photoconductors. (I couldn't get google to kick up a few pages.)

If you don't find anything else, let me know I can click a few pics and post on dropbox... (The reference section has no (obvious) photo conductor references.)

George H.

when you think about what politicians actually do know, it is scary that we let them lead us.

They pass so many stupid laws, most are ignored. I asked a British engineering team if our design would need to be CE tested, and they all broke up in laughter. One then kindly explained to this dumb Yank that CE means Can't Enforce.

Why not test some?

Three letters - AGW.

I'm sure it is. time constant was north of a second on the ones I used a long time ago. It varies hugely depending on R, which varies hugely depending on lux.

NT

Shouldn't be hard to measure. Measure C in total darkness, and R in total darkness & sunlight, and take it from there.

I might be wrong but I'd expect mfrs to be wary of providing anything exact since they're not exact devices.

NT

They could provide RANGES.

I suspect that manufacturers don't provide anything exact because they don't know what the hell they're making.

I took Donati to lunch. Photocondcutros are typically doped (making traps) for more gain but less HF response. CdS is typically doped with CdCl2 and CuCl2, copper is an acceptor and contributes trap states with a lifetime of ~50 ms.

CdSe is simialr to CdS but with a trap time of ~10 ms.

George H.

Thinking about the physics I suspect the time constant is a non-linear function of the resistance, slowing rapidly as the resistance rises in response to removal of the light. ...Jim Thompson

Lab equipment generally meets CE standards by default.

AFAIK, the only limitations they put on that sort of thing include operation within a shielded enclosure, filtered and surge-protected wiring, and stuff like that. Really it's your own fault if your test site sucks, and blows up your test equipment (or vice versa!).

So such equipment isn't required to meet e.g. IEC 61000-4 and such.

Tim

"Jim Thompson" wrote in message news: snipped-for-privacy@4ax.com...

No, I said R vs. t. In other words something suggestive of diffusion.

Tim

Ah, cool!

Tim

The TC (according to Donati) is the lifetime of the trapped state. Different impurites have different trapping times..

George H.

I'd be interested in seeing some data (to see if I can fit it ;-) ...Jim Thompson

-- | James E.Thompson | mens | | Analog Innovations | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | San Tan Valley, AZ 85142 Skype: Contacts Only | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at

The problem is, there's quantum effects in the sensitivity and turnoff, and the carrier lifetime (of the promoted electrons) as well as the dark current and noise, depend on doping/contaminants/aging. The recombination of those charge carriers is potentially very complicated, and if you want something more than a simple model, the thinking cap has to come off the shelf (or the library card out of the wallet).

I'd expect the charge carriers to build at the same exponential-decay curve on turnon as at turnoff. Maybe, if you only care about the onset of turnon, you get the fastest rate because that depends on building up the charge carriers, while you get a slower rate at turnoff because there's more than one recombination time constant, from more than one conduction band? There might, for instance, be a metastable state related to dopants.

In any case, the manufacturer's data doesn't constitute a complete picture.

What do you think's going on besides varying R and fixed C?

NT