Half black photo with Flash in Auto mode Nikon D3100

[wfooshee]

I have never heard of adjusting the speed to accomadate orientation. I would venture a guess that any deviation in shutter speed from landscape vs portrait is significantly less than the 1/3 stop adjustment available to us.

As for which half of the image is black, that CAN be changed by synching rear curtain vs. the standard front curtain. If the flash is triggered with the normal front curtain sync, it goes off when the front curtain is fully open. With top-to-bottom shutter travel that will show the bottom of the image dark, as the second curtain has already covered the top of the sensor, and the inverted image puts the dark area at the bottom. With rear curtain sync, the flash triggers the instant the rear curtain is ready to start. With too high a speed that means the front curtain has not fully opened, so in this case the rear sync is actually earlier than the front sync. The top of the image will be dark.

It is possible that the shutter blades are hanging or not traveling freely which would mess up the timing. That may not even affect exposure too much if the blades are traveling slower than expected but evenly slow. The gap between them will be the right timing, just the blades aren't clear of the frame when they're supposed to be. If one blade travels differently than the other, than exposure gets adjusted from the intended exposure. If one blade sticks and continues, you'll have a lighter band at that point of the frame. I had that with a Canon AE-1 that would occasionally show a slightly lighter band about 2/3 the way across the frame. That was a horizontal cloth shutter, and the rear curtain was sticking just a bit, and where it stuck got more exposure.

I still haven't seen what flash is being used. If it's a Nikon flash, or a 3rd party designed for Nikon, it should limit the maximum shutter automatically just by being attached and powered on. A cheaper 3rd-party flash may not have the smarts to do that, and you can set whatever shutter speed you want. If it's a proper flash that the camera should communicate with and you get shutter interference with the flash, then the shutter is probably not working right, most likely slow blade travel.

 

[The_Traveler]

I was doing a search for 'whirled pees" and this thread came up.

 

[WayneF]

As for which half of the image is black, that CAN be changed by synching rear curtain vs. the standard front curtain. If the flash is triggered with the normal front curtain sync, it goes off when the front curtain is fully open. With top-to-bottom shutter travel that will show the bottom of the image dark, as the second curtain has already covered the top of the sensor, and the inverted image puts the dark area at the bottom. With rear curtain sync, the flash triggers the instant the rear curtain is ready to start. With too high a speed that means the front curtain has not fully opened, so in this case the rear sync is actually earlier than the front sync. The top of the image will be dark.

Your thought out logic sounds good, but if you actually go try it, the camera does NOT work that way. At least the Nikon D300 and D800 cameras do Not change the banding with either Rear or Front curtain sync. The banding is identical either way (different from each other, but the same with respect to Rear/Front Curtain sync). Just actually try it.

I've often wondered about how Rear Curtain sync can handle the timing. The full duration of a speedlight at maximum power level is normally pretty near the maximum sync speed. It can be a slight issue for front curtain sync cutting off the trailing tail of fading flash at maximum sync speed. Holding shutter back to maybe 1/160 second is often seen better for seeing all of the speedlight maximum power level. Not a big difference, not saying very significant, but the numbers are pretty close.

So how would rear curtain do this? It cannot wait to trigger until the rear curtain is ready to close, since then there won't be any time left for a long flash duration. So surely it has to trigger early by some nominal time to allow a nominal flash duration? Maybe it even backs off about by Maximum sync speed to allow most of a maximum power flash before it closes? Which would mean that maximum sync shutter would not be Rear curtain at all? This would not matter for a very slow shutter (as is often the point of Rear), and Rear Sync is pointless anyway at a fast shutter, when it doesn't matter what it does. So that is my made-up theory (also totally lacking evidence).

 

[wfooshee]

Your thought out logic sounds good, but if you actually go try it, the camera does NOT work that way. At least the Nikon D300 and D800 cameras do Not change the banding with either Rear or Front curtain sync. The banding is identical either way (different from each other, but the same with respect to Rear/Front Curtain sync). Just actually try it.

So I dig out my old Canon 300TL (which goes with my T90 film camera) and set my shutter to 1/500. Sure enough the black band is at the bottom of the image on both modes. I think I know why....

Just guessing, but maybe the rear sync isn't actually tied to the curtain itself, but still to the front curtain, plus a delay for the shutter duration. A fast enough shutter means too small a duration to affect the timing.

I've often wondered about how Rear Curtain sync can handle the timing. The full duration of a speedlight at maximum power level is normally pretty near the maximum sync speed.

According to my SB-600's manual, the duration at max power is 1/900 second, less than a third of the sync speed shutter duration (1/250 on my camera.) 1/64 power is only 1/25000 second. Studio lights are different in how their flash tubes work, and for some their duration is shortest on highest power, and actually gets longer with reduced power, resulting in the possibility of closing the shutter during the lighting if you aren't careful.

This would not matter for a very slow shutter (as is often the point of Rear), and Rear Sync is pointless anyway at a fast shutter, when it doesn't matter what it does. So that is my made-up theory (also totally lacking evidence).

I agree that rear sync is useless unless you're talking long shutter durations, but as long as we're doing thought experiments I'll point out that my hypothesis in my earlier post was that his shutter was malfunctioning by "dragging" the blades. The gap between curtains may be timed correctly (giving a correct shutter speed) but if the blades are moving too slowly, timing of events such as flash sync get screwed up. It may be that the shutter is indeed set to 1/200, and forced to that setting by the flash, but slow-moving blades are resulting in the curtain not being "out of the way" in time when the flash goes off.

 

[WayneF]

So I dig out my old Canon 300TL (which goes with my T90 film camera) and set my shutter to 1/500. Sure enough the black band is at the bottom of the image on both modes. I think I know why....

Just guessing, but maybe the rear sync isn't actually tied to the curtain itself, but still to the front curtain, plus a delay for the shutter duration. A fast enough shutter means too small a duration to affect the timing.

It never seems worded that way, but it seems a good guess. The second curtain is just a timed delay from the first curtain, so I suppose they can easily add 5 seconds shutter duration to it. Otherwise, they need a way to trigger the flash awhile before the rear curtain moves, which seems unnecessarily complex.

According to my SB-600's manual, the duration at max power is 1/900 second, less than a third of the sync speed shutter duration (1/250 on my camera.) 1/64 power is only 1/25000 second. Studio lights are different in how their flash tubes work, and for some their duration is shortest on highest power, and actually gets longer with reduced power, resulting in the possibility of closing the shutter during the lighting if you aren't careful.

Actually, speedlight maximum power level is the same concept as monolights. But the monolights simply reduce the voltage to set lower power (which is slower, less energetic), where the speedlight always sets full voltage, but truncates the duration with IGBT for lower power levels (which is of course faster). The monolights become more red, and the speedlights become less red (chops off the cooler tail).

The ISO standard for flash duration is the standard t.5 spec, which is the time between the half power points (which is a standard thing for engineers measuring difficult things that trail off with no definite end point). This is true of speedlight maximum power too. Then the universal rule of thumb is that the t.1 time (the 90% power above the 10% power points) is three times t.5, which would be more useful for photographers. So the t.5 1/900 second is actually 1/300 second t.1 duration. And of course, 10% is not zero, so there is still some more too.

Top end Nikon camera models specify flash sync at 1/250 second, but they offer 1/320 second sync and can do it. They clearly do 1/320 second fine, but Nikon adds a footnote saying "the range might be a bit less" at 1/320. That is just Nikon speak, and my guess is they imply at maximum power, due to chopping off a long flash duration. Not a factor at lower power (for speedlights), and the later manuals seem to be dropping that now (but they still spec 1/250 and offer 1/320).

t.5 is why the speedlight 1/2 power duration is usually near the full power duration... Because full power is measured to the half power points, and 1/2 power cuts off the flash tube at the half power point. Some like the SB-600 are a bit further from being equal, 1/900 and 1/1600 second. This also depends on rise time, and that lower power flashes are faster than higher power flashes. But the SB-800 is 1/1050 full and 1/1100 half.

I agree that rear sync is useless unless you're talking long shutter durations, but as long as we're doing thought experiments I'll point out that my hypothesis in my earlier post was that his shutter was malfunctioning by "dragging" the blades. The gap between curtains may be timed correctly (giving a correct shutter speed) but if the blades are moving too slowly, timing of events such as flash sync get screwed up. It may be that the shutter is indeed set to 1/200, and forced to that setting by the flash, but slow-moving blades are resulting in the curtain not being "out of the way" in time when the flash goes off.

The focal plane shutter is not blades, but is two very thin roll-up titanium curtains, both driven by the same motor. That is the beauty, shutter accuracy is controlled by a simple quartz delay starting the second curtain.

I think we need to hear from the OP first, to at least hear the actual shutter speed, and maybe the flash type and situation (off camera?) before we scare him telling him the shutter is bad. He said "problem only occurs with flash". Any thing is possible, but that probability seems pretty low at this point,

 

[The_Traveler]

What a great and useful thread.

 

[wfooshee]

The focal plane shutter is not blades, but is two very thin roll-up titanium curtains, both driven by the same motor.

Roll-up???? Where's a roll-up titanium shutter? The old cloth focal-plane shutters rolled on spools, but the modern shutters are blades. Four blades for each "curtain."

Naked shutter blades found with a Google search

Titanium can't roll onto a spool. Making the curtain out of blades makes it stowable in the space above and below the image sensor.

I think we need to hear from the OP first, to at least hear the actual shutter speed, and maybe the flash type and situation (off camera?) before we scare him telling him the shutter is bad. He said "problem only occurs with flash". Any thing is possible, but that probability seems pretty low at this point,

My busted-shutter theory is speculation, but from having seen it before, a mechanical issue like dragging curtains can throw off the timing. It would be nice for the OP to come back and answer some of the questions.

 

[WayneF]

Roll-up???? Where's a roll-up titanium shutter? The old cloth focal-plane shutters rolled on spools, but the modern shutters are blades. Four blades for each "curtain."

Naked shutter blades found with a Google search

Titanium can't roll onto a spool. Making the curtain out of blades makes it stowable in the space above and below the image sensor.

OK, my blunder then. Here is more detail:

Nikon s Honeycomb-pattern Titanium Shutter - Index Page

I was going too fast, sorry. Thin metal can roll up, and cloth curtains always did, so I didn't pause to question it. But yes, the rule is, question everything.

 

[KmH]

For example, the D300 and D800 shutters run in opposite direction in regard to top or bottom banding.

Huh? You lost me with that statement.
A D800 shutter runs from bottom to top for an exposure? Because a D300 shutter moves from the top to the bottom to make an exposure.
Does Nikon say that somewhere? I've looked at quite a few web/sites pages (including Nikon) and can't find anything about that.
Sounds interesting. Weird, but interesting.
How about some links?

OK, try here: Maximum Shutter Sync Speed for Flash

No thanks.
I prefer an independent source of your information.

As far as the titanium shutter info in the article you linked to:

the famous Titanium shutter was finally replaced with an aluminum alloy version for the FM2n in 1989.

So Nikon abandoned the titanium shutter some 20+ years ago?

 

[WayneF]

No thanks.
I prefer an independent source of your information.

Suit yourself. Yeah, for all you know, I might have screwed up and mounted one camera upside down on the tripod. Then maybe gravity messed it up. LOL

As far as the titanium shutter info in the article you linked to:

the famous Titanium shutter was finally replaced with an aluminum alloy version for the FM2n in 1989.

So Nikon abandoned the titanium shutter some 20+ years ago?

Says it did in the FM2n, which is actually quite well known.

It was only YOU that extended it to all Nikon.

 

[Avishekh Gautam1]

So I dig out my old Canon 300TL (which goes with my T90 film camera) and set my shutter to 1/500. Sure enough the black band is at the bottom of the image on both modes. I think I know why....

Just guessing, but maybe the rear sync isn't actually tied to the curtain itself, but still to the front curtain, plus a delay for the shutter duration. A fast enough shutter means too small a duration to affect the timing.

It never seems worded that way, but it seems a good guess. The second curtain is just a timed delay from the first curtain, so I suppose they can easily add 5 seconds shutter duration to it. Otherwise, they need a way to trigger the flash awhile before the rear curtain moves, which seems unnecessarily complex.

According to my SB-600's manual, the duration at max power is 1/900 second, less than a third of the sync speed shutter duration (1/250 on my camera.) 1/64 power is only 1/25000 second. Studio lights are different in how their flash tubes work, and for some their duration is shortest on highest power, and actually gets longer with reduced power, resulting in the possibility of closing the shutter during the lighting if you aren't careful.

Actually, speedlight maximum power level is the same concept as monolights. But the monolights simply reduce the voltage to set lower power (which is slower, less energetic), where the speedlight always sets full voltage, but truncates the duration with IGBT for lower power levels (which is of course faster). The monolights become more red, and the speedlights become less red (chops off the cooler tail).

The ISO standard for flash duration is the standard t.5 spec, which is the time between the half power points (which is a standard thing for engineers measuring difficult things that trail off with no definite end point). This is true of speedlight maximum power too. Then the universal rule of thumb is that the t.1 time (the 90% power above the 10% power points) is three times t.5, which would be more useful for photographers. So the t.5 1/900 second is actually 1/300 second t.1 duration. And of course, 10% is not zero, so there is still some more too.

Top end Nikon camera models specify flash sync at 1/250 second, but they offer 1/320 second sync and can do it. They clearly do 1/320 second fine, but Nikon adds a footnote saying "the range might be a bit less" at 1/320. That is just Nikon speak, and my guess is they imply at maximum power, due to chopping off a long flash duration. Not a factor at lower power (for speedlights), and the later manuals seem to be dropping that now (but they still spec 1/250 and offer 1/320).

t.5 is why the speedlight 1/2 power duration is usually near the full power duration... Because full power is measured to the half power points, and 1/2 power cuts off the flash tube at the half power point. Some like the SB-600 are a bit further from being equal, 1/900 and 1/1600 second. This also depends on rise time, and that lower power flashes are faster than higher power flashes. But the SB-800 is 1/1050 full and 1/1100 half.

I agree that rear sync is useless unless you're talking long shutter durations, but as long as we're doing thought experiments I'll point out that my hypothesis in my earlier post was that his shutter was malfunctioning by "dragging" the blades. The gap between curtains may be timed correctly (giving a correct shutter speed) but if the blades are moving too slowly, timing of events such as flash sync get screwed up. It may be that the shutter is indeed set to 1/200, and forced to that setting by the flash, but slow-moving blades are resulting in the curtain not being "out of the way" in time when the flash goes off.

The focal plane shutter is not blades, but is two very thin roll-up titanium curtains, both driven by the same motor. That is the beauty, shutter accuracy is controlled by a simple quartz delay starting the second curtain.

I think we need to hear from the OP first, to at least hear the actual shutter speed, and maybe the flash type and situation (off camera?) before we scare him telling him the shutter is bad. He said "problem only occurs with flash". Any thing is possible, but that probability seems pretty low at this point,

 

[Avishekh Gautam1]

I too have the similar problem but with Nikon d5500. That is, while taking photos on automode, and with flash on the photos have dark band on the upper third of every photos. I would like to mention only when the flash is turned on and the problem persists irrespective of whichever mode I use auto mode or program mode with flash on. With no flash everything is fine. The shutter speed is 1/60-1/200 usually in automode when i have observed the band, never outside that range.

 

[WayneF]

I too have the similar problem but with Nikon d5500. That is, while taking photos on automode, and with flash on the photos have dark band on the upper third of every photos. I would like to mention only when the flash is turned on and the problem persists irrespective of whichever mode I use auto mode or program mode with flash on. With no flash everything is fine. The shutter speed is 1/60-1/200 usually in automode when i have observed the band, never outside that range.

I cannot explain that. If the shutter speed is not faster than 1/200 second, then that should never happen. The camera will not even allow the shutter speed to go faster with a hot shoe flash it recognizes (can communicate with). So you say that part seems to be working, but somehow the sync timing is incorrect. Seems impossible, at least I've never seen it, and I cannot imagine any option that could cause it. I would be convinced that it is a defective malfunction, and you should contact Nikon about repair before warranty expires.

FWIW, what is the flash mode set on the flash?