Is there a standard method to test the calibration of a handheld spotmeter? I got a Pentax Digital Spotmeter and I want to figure out if the readings are accurate and, if not, if there a way to calibrate the meter.
Is there any readily available benchmark for that matter?
Procure a Kodak or equivalent "gray card". This is a battleship gray placard with a surface that reflects 18% of the ambient light. You can test such a surface for accuracy. Perhaps a local photofinisher will help. Most such shops routinely use an instrument called a densitometer. Such an instrument reads film by transmitted light and prints by reflected light. The gray card should read 0.75 density units.
Load black & white film in your camera. Use the spot meter to measure the gray card. The meters output will be the shutter speed and aperture. Set the meter to the box ISO speed of the film. Shoot a series of exposures of the gray card. I suggest you use 1/3 f-stop increments. Also, shoot close-ups of the gray card i.e. the entire frame is an image of the gray card. Develop the film per specifications of the developer used.
Have the shop measure each frame plus the a reading from a clear film area (D-minimum). The frame that reads 0.75 + the clear film reading will be the correct exposure.
Assume the clear film (D-minimum) reads 0.12, then the frame with the correct exposure will read 0.75 + 0.12 = 0.82.
Lots of pitfalls when calibrating a meter.
0.05 = 1/6 f-stop
0.10 = 1/3 f-stop
0.15 = 1/2 f-stop
0.20 = 2/3 f-stop
0.30 = 1 f-stop
I test meters by finding a flat even-coloured / evenly-illuminated surface which is large enough to cover the field of view of whatever lens I care about (I use a white wall: I'd be wary of something that had any significant colour), and then pointing a camera I trust (so a recent digital camera) at it so it fills the field of view, taking a reading from it (fixing the ISO and aperture gives you a single number), and then pointing whatever I want to test at it from the same place and with an equivalent focal-length lens (or just metering the centre of the field with a spotmeter) and taking an equivalent reading from that to compare. Do this several times with both to check you're getting a stable reading.
You want to do this in some reasonably sane lighting conditions: a white wall under direct sunlight is not reasonably sane, a white wall at night is also not, but a white wall on an overcast day should be fine so long as the light is not changing. A white wall under pretty bright electric light with no natural light is also OK, and has the advantage that the light is stable: this is what I use.
Expect a variation of half a stop for anything old: nothing is really much more accurate than that (and B/W film barely notices half a stop), although the Pentax spotmeters are pretty good in my experience (I have an older one which I use for LF). There is also differing light loss in lenses of course: there's just no point in worrying about tiny variations.
This tests a single point in the response curve of the meter, so you don't know whether the meter is linear or not. You can iterate under different conditions to get a feel for that (when comparing cameras just try at different apertures: you can't do this with the meter), but things going nonlinear is almost never a problem in my experience: if something is out by n stops it will be out by n stops over its whole useful range.
If doing this for a spotmeter you can meter off a bunch of bits of the wall to get a feel for what the variation across the field-of-view of the reference camera / lens is: it should be small for this to work.