Pentax Spotmatic CdS Light Dependent Resistors

Steve Roberts · Jan 24, 2019

I can't resist the challenge of buying and restoring a Spotmatic if I find one for not much money at a car boot sale, charity shop, etc.. Most of the faults are relatively easily cured, but increasingly I'm finding that one or both of the CdS sensors has died or that its sensitivity has diminished. This is to be expected, as even the youngest SP1000s are pushing 45 years old. There are two common responses to this issue on forums; a) that the Spotmatic will operate OK with just one CdS sensor (it will operate, but not well as the reading taken is lop-sided), and b) to forget the metering and either estimate the exposure or use a separate hand-held meter.

The Spotmatic appeared in 1964, when electronic components were rather basic compared to today. With the wealth and range of components available today (and probably made to higher specs) I would be very surprised if there isn't a light dependent resistor out there that would replace Pentax's original item.

The primary parameters that need to be taken into account are physical size, light resistance, dark resistance, linearity and spectral response. These are freely available in great detail for modern day replacements and I can make reasonably accurate measurements of a 'known good' original CdS sensor, but before I go to the trouble of exploring all the options, I wonder whether anyone else has yet investigated and arrived at a modern CdS substitute? There are plenty of cheapo LDRs available but their sensitivity and repeatability are questionable. Also, I am aware that the SP-F, K1000 and KM use significantly different CdS cells, but they are younger and not (yet) typically suffering the same issues, so I'm not too worried about them at this stage.

Any thoughts or experience would be appreciated.

Steve

David Lyga · Jan 24, 2019

Somehow, someway, the late competitor of the Spotmatic, the Minolta SR-T series (albeit introduced only after the Spotmatic was a few years old), managed to keep its two CdS cells intact. I have never met an SR-T that did not have a working meter, but I have met many a Spotmatic that did have a faulty meter. Mechanically, they are both robustly built, but the Achilles' Heel for the Pentax was that meter. Also, the Nikkormat usually has a working meter. Maybe both the SR-T and the Nikkormat had meters which were better built and less apt to be negatively influenced by the deleterious effects of constant, strong light. - David Lyga

David Lyga · Jan 24, 2019

Steve Roberts said:
The Spotmatic appeared in 1964, when electronic components were rather basic compared to today. With the wealth and range of components available today (and probably made to higher specs) I would be very surprised if there isn't a light dependent resistor out there that would

It may have appeared for sale to the public in 1964 but it was 'introduced' at Photokina back in 1960, four years earlier. The four year hiatus was taken up with ironing out the bugs. Too bad the CdS bug did not die. - David Lyga

Kino · Jan 24, 2019

You might find some answers here:

Selenium Meter Repair on theOlympus Auto Eye II
http://www.thermojetstove.com/Autoeye/meter.htm

bernard_L · Jan 24, 2019

Steve Roberts said:
The Spotmatic appeared in 1964, when electronic components were rather basic compared to today. With the wealth and range of components available today (and probably made to higher specs) I would be very surprised if there isn't a light dependent resistor out there that would replace Pentax's original item.

Not so, unfortunately. Some technologies are just abandoned on the roadside. Try to buy a Se photocell! Or a germanium transistor. Not quite as bad for CdS photoresistors, also known as LDR (light dependent resistor). The main electronic parts supplier where I live (Farnell) lists 13 LDR's versus 300+ photodiodes; and of the 13LDR's, 7 are out of stock or discontinued. Turning to a major US distributor, the ratio is 60 LDRs versus 1200+ photodiodes.

Specifying an LDR is non-trivial. LDRs have a linear relation between log(illumination) and log(R). See graph below. The slope in log-log coordinates appears under various names, here "sensitivity", slightly misleading. Typical values are less than 1; that is advantageous for the designer, because a large (ratio) range of illumination is compressed into a smaller range of electrical resistance. But this also means the following: while a photodiode is fully spec'ed by he current at say, 10 lux, because the current-illumination relation is linear, for an LDR, the slope (called "sensitivity" below) must also be correct.

Worse. Look at the graph. The two straight lines represent the extremes of behavior (R versus illum) for the same part number: Advanced Photonix PDV-P9008. So, if you were to order that component, you should expect a 20:1 spread in R values at 10lux. Presumably, Pentax, Minotla, etc, had negotiated tight specs from their CdS cell supplier, and you are not in the same position.

Bottom line: I suggest you look for a donor body, having a major fault other than in the meter circuit, to cannibalize the whole meter circuit.

Steve Roberts · Jan 24, 2019

[QUOTE="Bottom line: I suggest you look for a donor body, having a major fault other than in the meter circuit, to cannibalize the whole meter circuit.[/QUOTE]

Thanks Bernard. I appreciate all that you say, but the thrust of my original post was that I would like to find a modern replacement rather than have to go in search of dwindling reserves of Spotmatics that may or may not have good LDRs and cannibalising them. Were the original components supplied to such exacting specifications, I wonder? With exposure meter accuracy typically only to within half a stop (and at what level of illumination?) and the ability to adjust the Spotmatic meter's reading over a vast range via the potentiometer, perhaps the components supplied via the manufacturers' own S.O.T. (Select On Test) procedures weren't so special. Like you, I have purchased components from Farnell and many other companies and I think there's scope out there for some experimentation.
Best wishes,
Steve

oreston · Jan 24, 2019

David Lyga said:
It may have appeared for sale to the public in 1964 but it was 'introduced' at Photokina back in 1960, four years earlier. The four year hiatus was taken up with ironing out the bugs. Too bad the CdS bug did not die. - David Lyga

I don't know that I'd call it a "bug" per se. I think poor low light performance is the one thing that really marks out the Spotmatic meter (even a properly working one) as "vintage" technology. Even so, CdS photoelectric cells were a huge leap forward from Selenium and in everyday situations they can still be good enough.
The CdS sensors in the Spotmatic certainly can and do fail on occasions but from experience I'd say that, while the pool is gradually diminishing, there's still a fairly high proportion of cameras out there with basically viable meters. Many of the ones I've encountered with "faulty" metering (either temperamental or not working at all) have simply had a poor or dirty connection between the contact on the reverse of the battery chamber (ie on the inside of the base plate) and the one inside the camera body - which is a simple fix. The variable resistor on the circuit can enable quite tight calibration, assuming the sensors aren't too degraded.
But yes, a confirmed source for replacement LDRs of an appropriate specification, if such a thing exists, would be very good news indeed.

John Koehrer · Jan 25, 2019

I hate to display my lack of knowledge on this but I'm curious if a pair of like CDS cells
would work because the circuit's a balanced bridge?

bernard_L · Jan 25, 2019

@ Steve Roberts
At least my technical arguments fell in receptive ears. One possibility is that the actual spread is much better (tighter) than the spec, especially if bought from the same batch. And, remaining in a hopeful mood, maybe there are mini-pots (at least two) to cope with the deviation of a matched pair from the nominal design spec. Good luck.
Bernard

Steve Roberts · Jan 25, 2019

bernard_L said:
@ Steve Roberts
At least my technical arguments fell in receptive ears. One possibility is that the actual spread is much better (tighter) than the spec, especially if bought from the same batch. And, remaining in a hopeful mood, maybe there are mini-pots (at least two) to cope with the deviation of a matched pair from the nominal design spec. Good luck.
Bernard

Always pleased to receive constructive observations or suggestions! As oreston points out, the low light level performance of Spotmatic meters was never that special anyway, so if I can achieve a reasonable result over the range of, say 1/30th at f2.0 to 1/500th at f11, that would cover most eventualities. I'm no stranger to disappointment in the field of experimentation with electronic and mechanical things, so won't be throwing myself off the Tamar Bridge if I don't have any success. I'll report back (but don't hold your breath!)
Best wishes,
Steve

AgX · Jan 25, 2019

I must admit this is the first time I learn of an old CdS resistor failing.

ic-racer · Jan 26, 2019

AgX said:
I must admit this is the first time I learn of an old CdS resistor failing.

Same here. How does he know the CDS is bad?

Steve Roberts · Jan 28, 2019

ic-racer said:
Same here. How does he know the CDS is bad?

The most basic clue may be that the camera's TTL metering needle has a very limited, sluggish response that delivers an incorrect reading. (Obviously basic checks such as testing/replacing the battery, checking battery compartment contacts, meter switch, wiring and so on and so forth should be carried out along with investigation of the meter trim potentiometer, looking for an open circuit or noisy carbon track. Having two CdS sensors, one each side of the v/finder, sometimes the giveaway is that the meter is far more responsive to a light source on one side than it is on the other. Suspected faulty CdS sensors can easily be checked by substitution of known good ones from another camera. Once removed from the camera, the state of the sensor can be roughly gauged using an ohmmeter whilst a) sticking a finger over the sensor's active surface to give a 'dark' reading and then b) holding it close to a light source. Failed/failing sensors normally (but not always) exhibit an excessively high resistance when in the 'exposed to light state', when their resistance should fall dramatically.

ic-racer · Jan 28, 2019

Just that the manual had the warning:

Also remember that the two CdS cells installed have to be replaced at the same time, as it is extremely difficult to determine which one of the CdS cells is defective.

In your favor, it reads as if the circuit can be calibrated with any of the three CdS cell sensitivities available for replacement based on the trimming of variable resistor 10008 (which in the schematic looks incorrectly shown as a non-variable resistor).
The service manual, however, does not explain how to set trimmer 10008. It seems as if it is fixed at the the factory and marked with paint to match the color of the CdS cell wires. Probably why it is shown as a fixed resistor in the manual; they don't intend you to mess with it.

With no mercury battery and without knowing the specs of the CdS cells; Sometimes the simplest circuits are the most complex to get working correctly.
Two things we do know, the resistance of the galvometer is 3k ohms and at a 3uA current, the needle should point exactly between the plus and minus in the finder.

AgX · Jan 28, 2019

Very interesting about these triple-sets of replacement CDS resistors. And as Pentax made such spare sets they likely expected them to fail. Well, somewhere I got a philips booklet from the 60s on their CDS resistors, maybe that will make me wiser on their longevity.

ic-racer · Jan 28, 2019

AgX said:
Very interesting about these triple-sets of replacement CDS resistors. And as Pentax made such spare sets they likely expected them to fail. Well, somewhere I got a philips booklet from the 60s on their CDS resistors, maybe that will make me wiser on their longevity.

That they had three grades of CdS cells I think is from them testing batches of the CdS cells at the camera factory and separating them into three groups. Like the way resistors are made. Resistors are not made to a specific resistance, batches of resistors are made and they are sorted and colored based on how they test at the factory.

This grading of photocell resistance is also seen in the Polaroid 250 repair manual:

Steve Roberts · Jan 30, 2019

"Also remember that the two CdS cells installed have to be replaced at the same time, as it is extremely difficult to determine which one of the CdS cells is defective."

I can't really think of a scenario where one would replace only one CdS cell except perhaps because of limited availability, e.g. scavenging from another body. That would be akin to replacing half the spark plugs in your engine at a service! As for identifying which cell is defective (if only one), 50 years down the line from manufacture they tend to be fairly terminally non-responsive. Of course the Pentax manual was written when the cameras would have been much newer and presumably Pentax were anticipating that repairers then were looking for a much more subtle difference than would typically be seen today. I have a Hewlett Packard programmable precision PSU which serves to replicate the original mercury cell or a modern substitute (or pretty much anything) as necessary. By the way, I see thread has appeared on a similar subject regarding the OM-1n.
Steve

ic-racer · Jan 30, 2019

Very similar circuits. You should be able to get it working with some experimentation of resistor values. In the 1970s I build a darkroom exposure meter with the same circuit.

ic-racer · Jan 31, 2019

I don't know if you saw the OM1 adjustment procedure for the resistors in that schematic. May help you zero in on correct values for the Pentax:

zanet57 · May 4, 2019

ic-racer said:
Just that the manual had the warning:

In your favor, it reads as if the circuit can be calibrated with any of the three CdS cell sensitivities available for replacement based on the trimming of variable resistor 10008 (which in the schematic looks incorrectly shown as a non-variable resistor).
The service manual, however, does not explain how to set trimmer 10008. It seems as if it is fixed at the the factory and marked with paint to match the color of the CdS cell wires. Probably why it is shown as a fixed resistor in the manual; they don't intend you to mess with it.

With no mercury battery and without knowing the specs of the CdS cells; Sometimes the simplest circuits are the most complex to get working correctly.
Two things we do know, the resistance of the galvometer is 3k ohms and at a 3uA current, the needle should point exactly between the plus and minus in the finder.

Steve Roberts said:
"Also remember that the two CdS cells installed have to be replaced at the same time, as it is extremely difficult to determine which one of the CdS cells is defective."

I can't really think of a scenario where one would replace only one CdS cell except perhaps because of limited availability, e.g. scavenging from another body. That would be akin to replacing half the spark plugs in your engine at a service! As for identifying which cell is defective (if only one), 50 years down the line from manufacture they tend to be fairly terminally non-responsive. Of course the Pentax manual was written when the cameras would have been much newer and presumably Pentax were anticipating that repairers then were looking for a much more subtle difference than would typically be seen today. I have a Hewlett Packard programmable precision PSU which serves to replicate the original mercury cell or a modern substitute (or pretty much anything) as necessary. By the way, I see thread has appeared on a similar subject regarding the OM-1n.
Steve

In fact it's not so difficult to determine which cell is defective. Usually their resistance goes to 0 kohms and resistance does not vary with light. A working Cds cell has its resistance vary from 2 kohms at daylight to more than 100 kohms at night. You have to open the top of the camera and desolder wires.

ciniframe · May 4, 2019

Wow! Never knew how complicated it was to have a built in meter reading off the screen. Always the guy to head for the simplest (and cheapest) solution I have no problem in ignoring in-op meters in all mechanical cameras, and grabbing my trusty Gossen Pilot for most daylight work outdoors.
Of course with my OM-4Ti that is not an option but I have plenty of OM-1 bodies also.
Besides, all this electronics talk is way WAY above my understanding.

Pentax Spotmatic CdS Light Dependent Resistors

Privacy & Transparency

Privacy & Transparency