CdS sensor in lightmeter (Pentax 1/21 analog) response time

[Murray@uptowngallery]

Hello:

I have heard about pointing a CdS meter at a bright light source before reading a dim source to speed up the response.

What do I do if I am eyeing an illuminated church at night but only get about an EV5 off the highlights? I only want the highlights and some detail of the bell tower, but the bright source isn't enough to 'pump up' the photocell.

I notice a few-to-several second drifting of my analog Pentax 1/21 spotmeter on low levels, but I'm happy with the stability after this period of time. I haven't waited 3 minutes to see what happens, as the manual mentions.

Does one need to compensate in some way with such a meter and low levels, knowing readings will be inaccurate? (Ignoring reciprocity failure which isn't mandatory at EV 4-5).

Thank you

Murray

 

[glbeas]

Carry a flashlight with you and shine it into the lens to pump the CdS cell.

 

[Murray@uptowngallery]

Hello:

That's what I was going to plan on...but wondered what happens if you don't... you get a prematurely inaccurate reading? Mine seems to drift downward over that several second period, so it might read higher than actual if I'm impatient...

That drift gives a whole lot of confidence in the batteries, regardless of how new they are!

Thanks

Murray

 

[Dave Mueller]

This doesn't make any sense from an electronics viewpoint. CdS cells operate on light, not heat (ok, at the molecular level, the energy in the photons is dissipated as heat, and all resistors change value according to temperature, but I don't think your meter has a temperature correction circuit!). Exposing them to bright light to speed up response can't possibly work. The cell is the only light sensitive part in the meter, and if you could add light to make it "faster", that would also affect the reading since you would be measuring the sum of the two lights. Small variations in readings at low light levels could be caused by flare from nearby sources and a host of other things. I think a controlled experiment is in order...

 

[glbeas]

It's not the heat that the CdS cell is responding to, it's the free electrons that are knocked loose from the crystalline structure by the light that allows the electrical current to flow.Therefore a flash of extra light can have an effect on later exposure to light briefly. It's like there is a threshold level of light to get the current flowing, similar to the way film responds to light. How much effect this is will need to be experimented with, though textbooks say CdS cell do have a "memory" that fades over time.

 

[Jorge Oliveira]

It's been a long time I've worked with CDS cells as such, but AFAIR there was no mention of light flashing the cell for low light reading.

The CDS cell's resistence increases as light decreases (that's why less current flows through them) and, for low light levels this resistence increase is a quite slow process indeed. That's one of the reasons they were replaced by other technologies.

So, it doesn't makes sense to me to flash it with light - any reading in this condition could be false.

Do a test: in a dark place, read patiently the light of an even surface until it stablizes. Then flash it and read again after stabilization.
If the flashed reading indicates more light, I woul feel this is a fake reading.

Jorge O

 

[Ed Sukach]

Odd ... If CDS meters "drifted" with time, I would imagine another system would be required for "Flash" meters. After all, the maximum flash duration I get from my DynaLites is 1/450th of a second. The Gossen UltraPro I use hasn't shown **any** drifting that I can see.
There are minor - very minor - variations in readings - to be expected - those variations are "precision" errors, and SOME error is unavoidable in all measuring systems - not only photometer - "Light meters".

 

[Jorge Oliveira]

Ed

Flash meters do not use CdS cells, they use silicon cells (photodiodes), a different breed and, nowadays, fast to boot. CdS ones become obsolete by the mid 80's.

Photodiodes are used, for instance, in fiber optics communications where 'flash' time is of 1/1,000,000,000 of a second duration - or less..

I believe the fist camera manufacturer to use them was Fuji (silicon blue cells) by the late 70's.

Jorge O

 

[Murray@uptowngallery]

That guy looks familiar...doesn't he belong to Mensa and advises Dilbert on occasion?

Murray