If you set "1/60" shutter speed, what exposure time do you expect?

Here's a diagram of what I'm talking about.

Exit pupil is at the top. sensor at the bottom. Shutter in between.

Thankyou.

Yes, I get all that. I have diagrams like that for EFCS. AFAICT, EFCS wants to know where the exit pupil is to make exposure consistent across the frame.

If I understand correctly, the first and second curtains run in separate tracks, seperated by some distance in the 0.5mm ballpark. With the curtains about 5mm from the sensor plane.

I have some limited confidence in the 5mm from the sensor plane being in the right ballpark, but not much confidence in my estimate of the seperation of the shutter curtains.

Because the curtains are likely to be accelerating for roughly the first half of their travel across the sensor, it makes a difference whether the first or second curtain is nearer the sensor. Without thinking too hard, it seems like it would be better if the first curtain was nearer the sensor.

the main worry seems to be the clash between mains frequency and pictorial outcome, If not near 1/60 or 1/50 will a small deviation really matter?

p-

Inspiring thought Jim !

John's summary shows that exposure of a D7200, mechanical shutter, goes up steadily to 15% higher than proportional exposure at 1/8192s as compared to 1/1024s.

I confirmed that general behavior for several Nikon cameras using mechanical shutter. My D850 which showed so precise exposures with fully electronic shutter, was the only one deviating from a steady increase over the proportional exposure. Details on measurement procedures were given in my post higher up in this thread. Also the D850 shows higher than proportional exposure for shorter exposure times. The rms1 and rms2 columns are mostly due to jitter of the mechanical shutter, staying below 2% at 1/8192s. The higher than proportional exposure goes up to about 10% at 1/8192 as compared to 1/1625.

I have also tested the mechanical shutter of my Z7II with an f/4 lens at f/4 120mm in the speed range 1/161s to 1/2048s , which has the shutter fully open for a moment at least to 1/256s . At the lower speeds shutter jitter impacts barely the exposure time. At the higher speeds here, jitter stays below 1/% . The exposure excess goes up to 7.5% at 1/2048s as compared to 1/161s.

How can that be, if I assume that the mechanical shutters are sent on their spring driven path with high electronic clock precision.
The geometry of the mechanical shutter, as learned from my shutter replacement on a D7200
is like this exit pupil illuminating the pixels is about 100mm before the sensor plane, The first shutter curtain is closer than 2mm to the sensor plane, maybe just 1mm. The second shutter curtain is another half to less than 1 mm further away from the sensor plane. The sensor patch measuring exposure is centered around the optical axis.
Sure, since the shutter is not in the sensor plane, there is a lengthening of exposure into the time of fractional shading.
But, the on axis beam illuminating the central pixel will be chopped by the timing of the shutter curtains, which I assume to be precise except for mechanical jitter. A beam from the pupil towards oncoming shutter curtains will be chopped a bit earlier as a consequence of the geometry, but a beam on the opposite side will be chopped a bit later by the same amount. So at the moment, I do not see yet how this could lead to exposure excess on the order of 10%.

I now used a similar setup with a diffuser covering the lens and using a single LED panel to illuminate it. My D7200 with a replaced shutter has a stronger apparent stretch at short exposure times than yours.

I fact it is about twice as bad as yours. It is roughly like that the second curtain starts late by some amount, independent of shutter time. This matters most at high shutter speed. For my D7200, with replaced shutter, this would amount to about 40 micro seconds late, while your D7200 would have the 2nd curtain about 20 micro seconds late.
I have used the Z7 II with mechanical shutter and the matte screen before lens over a wider time range, taking 1/203s as reference:

not much stretch appears up to about 1/1024s shutter speed. The 19% stretch at the highest speed would imply a late second shutter by about 25 micro seconds.
Curiously, the shutters are marked with what may be calibration notes by Nikon. In my picture at nikongear you can make out the first number -45 on the old shutter. Possibly, a late (or early) second shutter could be corrected in firmware adjustments. - I try to remember to ask a question to Nikon service, if there is such an adjustment possibility, if I get a chance to.
To rule out other odd non-linearities, I reran the test with silent electronic shutter for the Z7II :
like the D850 this is quite linear. Curiously, it is not quite as clean as my previous test with two LED panels illuminating white paper, Z7 II electronic shutter:

Of course, even a 30% exposure time stretch at 1/8000s is only about a third stop above target. This is not a problem for action photography, where 1/8000 might be used, and only occasionally. Also of course, it is the mechanical tolerances, which make exceedingly short exposures with this type of variable speed shutter a technical challenge. Understandable that camera makers seem to agree to 1/8000 as the fastest for (near) focal plane mechanical shutters.

I don't see how that's possible. The recent Nikon sensor stack thickness is between 1.1 and 2.2 mm.

If my theory is correct, the effective shutter speed at high settings will vary with f-stop.

What Jim is saying is that the effective exposure times we're seeing depend on the Numerical Aperture, the location of the exit pupil, the height of the curtains above the sensor, the seperation of the curtains in a direction normal to the sensor, and the speed versus time of the shutter curtains - as well as the target shutter speed.

We have a similar situation to EFCS, but not as bad. I'd thought a bit about the problems of implementing EFCS, but it hadn't occurred to me that similar effects were significant with a mechanical "focal plane" shutter.

We can expect exposure variation along the direction of sensor travel too, at high shutter speeds.

From examining various images of cut-away cameras on the web, the mechanical shutters usually seem to be about 5mm from the sensor, though the space inbetween isn't all air.

A detail I'm missing is an estimate of the seperation of the planes of the shutter curtains.

I did some more tests on D7200 at 150mm f/5.6, early this morning, but I haven't processed them yet. If what Jim is alluding to is correct, the increased f/number may give more even exposures.

I've also photographed 5kHz LEDs to measure the D7200 shutter speed profile.

More in a bit.

good point Jim ! I meant to 1 surface of sensor.
However, if my argument on integrating the exposure contribution to a pixel from the exit pupil, as chopped by the shutter, is correct, then the distance of the shutter from the sensor should not matter near the optical axis.
Still, I have been thinking how to take accurate measurements with slow apertures, to see clearly if there is a dependency.

I do see some powers of 2 in the "TargetExposureTime" tab in ExifTool for my Canon. I do not know if it is in the EXIF actually, or the tool creates it, etc. I see 1/256, 1/2048, etc. I see also 1/20, 1/25.4, 1/161, which seems to be 1/3 stops rounded.

You were using an f/1.4 lens at f/1.4. I was using an f/2 lens at f/2. The difference in Numerical Apertures might go some way towards explaining the differences in our results.

First, I must apologise for my late reply.

Most of the world's nominal GDP is in 50Hz regions, so perhaps 50Hz flicker-minimising shutter speeds should get more attention than 60Hz flicker-minimising shutter speeds.

In 50Hz regions, we might want shutter speeds:
1/100.0, 1/50.0, 1/33.3, 1/25.0, 1/20.00, 1/16.67, 1/12.50, 1/10.00, ...

Rather than the 60Hz shutter speeds:
1/120.0, 1/60.0, 1/40.0, 1/30.0, 1/24.0, 1/20.00, 1/15.00, 1/12.00, 1/10.00, ...

Some of those don't conflict:
1/120.0, 1/100.0, 1/60.0, 1/50.0, 1/40.0, 1/20.00, 1/10.00

But it's a mathematical accident that the conventional stop-based (EV-based) scale of shutter speeds (see here) provides shutter speeds which are quite effective at attenuating 50Hz mains-related lighting flicker (though not as effective as they could be).

The key mathematical fact here is that, for actual shutter speeds which are close to sub-multiples of the (fundamental) flicker frequency the remaining flicker (for all flicker harmonics) is very nearly equal to the relative error of the actual shutter speed, compared to a submultiple of the flicker frequency.

Alternatively, the flicker attenuation is very nearly equal to 1/(relative shutter speed error).

Depending on your Engineering/Maths background, that may have been a bit hard to digest, so I'll give some practical examples:

If you want to mess about with this stuff in a spreadsheet, this might help:

Function SINC( ByVal x  As Double )  As Double
   x = Abs(x)
   If x > 1.0E-8 Then
      If x = Int(x) Then
         SINC = 0
      Else
         SINC = Sin(Pi * x) / (Pi * x)
      End If
   else
      SINC = 1
   End If
End Function

your hypothesis does not pass test.

A three stops faster lens f/1.4 against f/4 before leads to the same 19 % stretch of exposure for the fastest shutter speed of my Z7II. A linear dependence on f-stop ratio would have your hypothesis expect a 2.8 fold increase of exposure stretch with the fast lens. The idea of late second curtain still stands. It will be tested by John's direct shutter time measurements.

It is a shutter specimen individual thing. All my cameras have different exposure stretch at 1/8192s with mechanical shutter:
my D7200 29% yours 15% , my D500 11% , D800 13% , D850 10%, Z7II 19% (f/1.4 and f/4).
My reply to Jim showed that there is no f ratio dependence for this exposure stretch of significance.

Well, I guess that disproves my hypothesis. Good job.

This is what I got on D7200 with a 150mm lens at f/5.6:

The lens is Sigma 50-150 f/2.8 DC OS. At 150mm it's truly telephoto. Bill Claff doesn't have a model for it, but I'm guessing that the exit pupil is 10cm+ from the focal plane.

I halved the height (but not width) of my patch size to give 512x256, as well as using a larger f/ number.

That reduced the "exposure stretch" on my D7200 to 11%.

This makes me more confident about some surprising electronic shutter speed measurements I made on Panasonic S5 (more later) - in these measurements, the electronic shutter speeds have this kind of "exposure stretch". Perhaps to match the mechanical shutter?

In any case, if we're not seeing anything close to 1/8192 over a tiny patch at the centre of the frame, when are we supposed to see it ?

You would have to look into the statistics of your averaged 16 measurements to know about the significance. At least the stretch is "monotonic" roughly doubling when halving the shutter speed. This is consistent with a systematic, constant, lateness of the second curtain. It may be telling that all our cases see lateness. Maybe that is intentional: 50% early would be a full stop underexposed: about 1/16000s, 50% late is roughly half a stop overexposed: the tolerance is more favorable to the late side.

After our results, I do not expect seeing 1/8192s with a mechanical shutter. What we see is less than a third stop off the mark, which is perfectly acceptable for a maximum speed in practical action photography. We can see more accurate shutter speeds with the electronic shutter, where I confirmed it for the D850 and Z7II. One can understand that Nikon only provides an electronic shutter with top models Z9 and Z8, now that they can make a reasonable flash sync speed of 1/200s with electronic shutter.

BTW, I use a 200x200 patch which gives 10'000 pixels in each of the four Bayer subsets. And the 10'000 pixels of the set let expect a 100 fold improved SNR for the DN average as compared to the single pixel DN.

My argument for independence of exposure stretch against aperture, may imply a stretch dependence along the shutter movement direction. One would have to recheck this and perhaps set up a measurement with small patches along the way.

By lack of time, I loose interest for more detail on exposure stretch with mechanical shutters. Surely there are further subtleties and catches with electronic first shutter curtain. Exposure stretch is not a practical problem, even for for my repaired D7200. I learnt, that I can get precise exposures with electronic shutter up to the highest speeds. One might use this to see how consistent stopping down can be with specific lenses: preliminary results taken along with my MTF measurements suggest fairly accurate for electronically controlled apertures.

Mechanical shutters do not move at constant speed, so there may be a lot more complexity to this. On two FF cameras that I've recently seen photographing strobes, the roll of the curtains varies about 2x down the frame! That means that the camera must also be changing the width of the opening as it moves down the frame, to perfectly equalize exposure down the frame. Perhaps this could explain some of the variance. Also, light is not coming perfectly perpendicular to the sensor, so the width of the opening looks different at the top and the bottom of the frame, since the curtains are staggered in height. I don't know if this is controlled for by f-ratio, or just an accepted inconsistency.