PDA

View Full Version : How to Measure T-Stop Accurately



Bob Renalds
03-24-2009, 09:50 AM
My first post is a question.

How to you verify a T-stop setting. I know you can measure the focal length of a lens (exactly) and also measure the entrance pupil (with a profilometer) and calculate F-Stop. You could then measure thr transmission and get a T value by dividing by the square root of transmission.

I know some rental houses have T-stop machines and I was curious to how they worked. I was also interested in what spectral range they operated to calculate the value.

Finally, how repeatable are the machines for measure to measure and also machine to machine.

If one lens measure T/1.945 I assume it can be called T-1.9. If it measures T/1.96, it's got to be a T/2.0. So the measurement seems critical to those marking the lenses.

I could not find much in the way of measurement methods and machine specs on the web.

Dominic Jones
03-24-2009, 09:59 AM
Whilst I'm not up on the exact equipment used, what I can tell you is that you've got the methodology a little backwards!

You don't measure a stop and say "Oh, it's T2.675" or whatever - you find the precise point that the aperture passes through T2.8, 4, 5.6, etc and mark those positions on the lens. It's usually done by the manufacturer, of course...

I would imagine the widest aperture is a rounded a little, but should be fairly consistent when talking about lenses from quality manufacturers.

I'm sure someone can chime in on the exact equipment use, but that's the theory.

HTH,
Dom.

Stephen Williams
03-24-2009, 10:14 AM
I would imagine the widest aperture is a rounded a little, but should be fairly consistent when talking about lenses from quality manufacturers.
HTH,
Dom.

Hi,

Some of the Zeiss SupersSpeeds come to mind.

Stephen

chuck colburn
03-24-2009, 04:18 PM
The device for calibrating stops is called an intergrating sphere. Various companys involved in optics have them. I use to use the one at Century Optics before they were bought by Schneider.

Daniel Browning
03-24-2009, 05:13 PM
If one lens measure T/1.945 I assume it can be called T-1.9. If it measures T/1.96, it's got to be a T/2.0.

Ha! You must not work in marketing. T/1.99 is reason enough to call it T/1.9 according to the ethics of most businesses.

Of course, photography is filled with all sorts of little inaccuracies. For example, to any rational mind, f/5.65685 would be rounded to f/5.7. Yet f/5.6 is used anyway.