Bronica AE-II prisms inaccurate

[jgboothe]

I manage two AE-II prism finders for use on ETRSi cameras. Both give inaccurate exposures (at least in terms of the reported shutter speed). One is around 1 1/2 stop over and one is around 1 1/3 stop under, when compared against a reliable sekonic light meter. These inaccuracies are the same when I use the prisms on a different body or with different lenses, so I'm pretty sure the problem lies with the prisms themselves.

Although I have not been able to check this fully, the inaccuracy appears to be linear - ie. they under or over expose by the same amount in different conditions, from dim to bright, and can therefore be used as long as you remember to compensate with the film speed dial. However, I manage these cameras on behalf of students and the discrepancy confuses them and leads to problems, so I would like to fix them if possible (in case the world ever gets back to normal again!). These issues existed when we bought the units secondhand around a year ago.

Does anyone know the likely cause of this inaccuracy? Is it likely to be the base sensitivity of the diodes or some other problem in the circuit? Could it be something as simple as the film speed dials being re-assembled incorrectly? Any ideas for fixing it?

 

[M-88]

Hello

Two years ago I had AE-II prism and inaccuracy was non-linear. That said, in daylight it was two stops off, while in low light it was five stops off. At least your one has linearity and you can use it to your advantage. When I disassembled mine, it was full of early 1980's electronics. I didn't even bother myself with checking the circuitry, because photodiodes (AE-II has two of them) had some sort of mold on them. I didn't even try to clean those because diodes were cemented in transparent material and would require ungluing.

"Could it be something as simple as the film speed dials being re-assembled incorrectly?" - if the ISO dial goes through the full range correctly including exposure compensation steps, then it is unlikely. I've read somewhere that AE and AE-II prisms are more likely to be non-working, rather than working. While AE-III are arguably better in terms of reliability and they offer some cool features as well. But they are also priced accordingly.

 

[wiltw]

If the target that you aim at does not average out all areas to 'midtone grey' (or 18% grey), your reflected light meter will NEVER equal an incident meter reading!!!

So you need to start with a target wall whose brightness difference from 18% gray (or the incidet meter reading) is known.
If you get yourself an 18% gray card target, and compare a reading (with the card filling the viewfinder) taken with that vs. the incident light meter readings, the difference should be darned close to each other. Any deviation would indicate the amount of error inherent to that specific metering prism.

Unlike the newer AE-III prism, there is no 'fixed compensation' adjustment switch which applies to all readings.

 

[jgboothe]

Hello

Two years ago I had AE-II prism and inaccuracy was non-linear. That said, in daylight it was two stops off, while in low light it was five stops off. At least your one has linearity and you can use it to your advantage. When I disassembled mine, it was full of early 1980's electronics. I didn't even bother myself with checking the circuitry, because photodiodes (AE-II has two of them) had some sort of mold on them. I didn't even try to clean those because diodes were cemented in transparent material and would require ungluing.

"Could it be something as simple as the film speed dials being re-assembled incorrectly?" - if the ISO dial goes through the full range correctly including exposure compensation steps, then it is unlikely. I've read somewhere that AE and AE-II prisms are more likely to be non-working, rather than working. While AE-III are arguably better in terms of reliability and they offer some cool features as well. But they are also priced accordingly.

Ok. Doesn't sound too promising, although if your inaccuracy wasn't linear and mine is, then maybe it isn't the same problem. You are probably right about the ISO dial. It does stop in the correct place at the extremities, so it's probably unlikely to be that. If no-one offers any experience of a successful repair then I will probably just accept the inaccuracy and tell the students to use a hand-held meter. Trouble is, half of them don't know how to use one!

 

[btaylor]

...just accept the inaccuracy and tell the students to use a hand-held meter. Trouble is, half of them don't know how to use one!

Sounds like a great time to learn!

 

[M-88]

Ok. Doesn't sound too promising, although if your inaccuracy wasn't linear and mine is, then maybe it isn't the same problem. You are probably right about the ISO dial. It does stop in the correct place at the extremities, so it's probably unlikely to be that. If no-one offers any experience of a successful repair then I will probably just accept the inaccuracy and tell the students to use a hand-held meter. Trouble is, half of them don't know how to use one!

Unless you are apt in repairs of electronics, you'll have to pay more for servicing those prisms than they are worth, unfortunately. So I'd say now is a good time for students to learn how to use a handheld meter.

Actually a lot of MF users employ handheld meters anyway.

 

[Deleted member 88956]

I manage two AE-II prism finders for use on ETRSi cameras. Both give inaccurate exposures (at least in terms of the reported shutter speed). One is around 1 1/2 stop over and one is around 1 1/3 stop under, when compared against a reliable sekonic light meter. These inaccuracies are the same when I use the prisms on a different body or with different lenses, so I'm pretty sure the problem lies with the prisms themselves.

Although I have not been able to check this fully, the inaccuracy appears to be linear - ie. they under or over expose by the same amount in different conditions, from dim to bright, and can therefore be used as long as you remember to compensate with the film speed dial. However, I manage these cameras on behalf of students and the discrepancy confuses them and leads to problems, so I would like to fix them if possible (in case the world ever gets back to normal again!). These issues existed when we bought the units secondhand around a year ago.

Does anyone know the likely cause of this inaccuracy? Is it likely to be the base sensitivity of the diodes or some other problem in the circuit? Could it be something as simple as the film speed dials being re-assembled incorrectly? Any ideas for fixing it?

Stating how you arrived at these differences between meter and prisms would help. If this was done in tightly controlled test so for meter saw what prisms did, then it does appear prisms may need adjustment. I assume you have checked battery contacts to ensure they have no sign of oxidation. Also, are prisms reading same difference in A and M modes ?

Do both prisms show clear view or are there black horizontal lines visible? This may have an effect on metering as some have deteriorated quite a bit.

 

[jgboothe]

If the target that you aim at does not average out all areas to 'midtone grey' (or 18% grey), your reflected light meter will NEVER equal an incident meter reading!!!

So you need to start with a target wall whose brightness difference from 18% gray (or the incidet meter reading) is known.
If you get yourself an 18% gray card target, and compare a reading (with the card filling the viewfinder) taken with that vs. the incident light meter readings, the difference should be darned close to each other. Any deviation would indicate the amount of error inherent to that specific metering prism.

Unlike the newer AE-III prism, there is no 'fixed compensation' adjustment switch which applies to all readings.

Thanks, but when I checked the readings from the prism against the light meter, I used the light meter in reflective mode, and pointed them both at the same evenly lit area. I wouldn't expect them to agree perfectly, but over a stop is too much, and both prisms are out in different directions - roughly three stops apart.

 

[jgboothe]

Stating how you arrived at these differences between meter and prisms would help. If this was done in tightly controlled test so for meter saw what prisms did, then it does appear prisms may need adjustment. I assume you have checked battery contacts to ensure they have no sign of oxidation. Also, are prisms reading same difference in A and M modes ?

Do both prisms show clear view or are there black horizontal lines visible? This may have an effect on metering as some have deteriorated quite a bit.

I'm pretty sure I did a visual check on the battery contacts and they looked clean, so I didn't pursue this any further. When I get back in work I'll check them again to be sure.

I hadn't noticed any black lines or degradation within the prism view - they are both pretty clear. Each prism was reading the same in A and M modes.

 

[wiltw]

Thanks, but when I checked the readings from the prism against the light meter, I used the light meter in reflective mode, and pointed them both at the same evenly lit area. I wouldn't expect them to agree perfectly, but over a stop is too much, and both prisms are out in different directions - roughly three stops apart.

OK, so all your tests were 'three meters reading reflected light', and not comparing two reflected light vs. one incident meter.
Puzzling that one would be more then +1EV and the other is more than -1EV in error.

 

[choiliefan]

Do these prisms have adjustable pots?

 

[jgboothe]

Do these prisms have adjustable pots?

What do you mean by adjustable pots?

 

[abruzzi]

Pots = potentiometers, basically variable resistors that would let you adjust the sensitivity.