Maker Pro
Maker Pro

LDR reads low (10K) in darkness?


Bill Bowden

Jan 1, 1970
I repaired an outdoor dusk/dawn LED lamp that had a defective LDR used to switch the LED on/off. Problem was the LDR resistance read low (10K)in darkness, so I replaced it with one I had laying around and got it working. Since then (don't know how long) the same problem has returned with the LDR resistance again reading low in darkness. I checked a few LDRs in my spare parts and found one that also read low in darkness, which had never been used.Do these things go bad over time, and do they deteriorate sitting in the sun all day?


Bill Bowden

Jan 1, 1970
They absolutely most certainly do go bad just sitting around. They
are incredibly sensitive to moisture. Potting in polyester resin,
epoxy etc. or using hermetically sealed photocells is the only way to
get around the problem. Humidity will kill them while they sit in a
parts bin

The cells with just a thin clear coating of epoxy don't deal with
water very well at all.

Thanks for that. The cell I replaced was encased in silicon rubber or something like that. I have one cell that appears to be encased in epoxy, so I might use that. I have some clear roofing sealer that is somewhat elastic when dry. Might try encasing a bare cell in some of that. The circuit switches at a resistance of about 27K ohms and the dusk reading of the cell is about 77K ohms. So, I was thinking of adding a 15K resistor in series with thecell so if it goes bad and only reaches 15K at dusk, the thing should still work?


Bill Bowden

Jan 1, 1970
On Tue, 19 Nov 2013 17:54:37 -0800 (PST), Bill Bowden

I think you'd just be delaying the inevitable by padding the

Pay attention to sealing where the leads enter the cell in the back,
as well as the surface,

It's been awhile since I used a CDS cell, but I seem to remember there
was a visible change in the moisture damaged cells versus new ones.
Take a magnifying glass to it... you may be able to see where the
moisture entered - comparing it to a new one. The silvery stuff turns

Providing you don't need a lot of current, a photo transistor can
replace a CDS cell in a DC circuit. Not good for operating relays
directly - but if your switch point is 27K you probably aren't using a
relay. For the hell of it I put a photo transistor across my
ohmmeter. Total darkness is >2 meg ohms, the light from my computer
screen ~8K, holding it close to a desk lamp ~100 ohms. I wonder that
someone doesn't already make CDS replacements with a couple of back to
back parallel photo transistors.

Back in the germanium age, we would hack the metal covers off power
transistors, pour in epoxy and make photo transistors that would
operate (sensitive) relays directly.

BTW Bill Bowden were/are you the same guy on the TC mailing lists some
time in the 90's?

No, I've never been on any mailing lists.

I encased a small LDR in epoxy yesterday. I mixed a glob of glue and rolledthe LDR in it and then dried it slowly by rotating it every half hour or so. Came out almost a perfect sphere like a little marble. Now, all I have to do is mount it through the top surface so water doesn't get into the mainassembly.



Jan 1, 1970
Frnak McKenney said:
While not ignoring the idea of a photodiode/phototransistor, I'm
curious as to whether anyone could comment on what kind of protective
coating could be applied to an existing (e.g.) CdSe LDR to protect it
and lengthen its useful lifespan.

Am I right that CdSe cells don't like DC operation? I seem to recall
something from long ago about them working better long term with AC. Aside
from that, these things were supposed to go extinct with RoHS, weren't they?