5

I got a Nikon 3400 in May. While shooting some water droplets I noticed the droplets contain motion blur even at the highest shutter speed settings. So I tried to verify this by clicking a pic of the ceiling fan [In India]. I might not know the exact speed of the fan, but I am sure as hell the fan does not do more than 4000 rotations per second. And yet, I get motion blur on the fan. Images attached.

Is my camera broken? Should I get it repaired? Or am I not understanding something?

Image Settings:
Mode: Manual
Aperture: f/5.6
Shutter Speed: 1/4000 sec
ISO : 25600 (It was indoor)
Focus: Manual
Metering: Spot


Update: The problem does not occur when I click the image in Live View Mode. The image appears darker and the motion is frozen. But When I switch to View Finder mode, something's not right.

Added Settings Images.

Shooting Menu Page 1 Shooting Menu Page 2 Shooting Menu Page 3

Shot in Live View with 1/4000, f/5.6, ISO-25600 Above Image: Shot in Live View with 1/4000, f/5.6, ISO-25600

Shot in View finder mode with 1/4000, f/5.6, ISO-25600 Shot in View finder mode with 1/4000, f/5.6, ISO-25600

PS: I checked the image properties and all the settings and details are exactly same. The camera does not show any change in the shutter speed value it displays.

Update: Check out the experiment at http://atulbhats.com/cam-demo

  • 2
    Please upload the original picture somewhere so we can inspect it more thoroughly. You could also make a series of photos with decreasing shutter speeds, to see if the exposure stays the same at some point. – ths Aug 11 '17 at 16:19
  • 3
    I cannot duplicate your blur situation with Nikon 1/4000 second shutter. Check your actual shutter speed reported in your Exif. – WayneF Aug 11 '17 at 16:56
  • 2
    There's something funny going on here. I downloaded these and looked at the EXIF and it shows they are all at ISO 6400, all the same F4.2, but wildly different shutters. If that were true, they would have wildly different exposures, and they do not. You are either manipulating something, or the camera is not working as expected. Turn off auto-ISO, turn set to manual, and try again. Adjust ISO as needed MANUALLY to get adequate light. Look at the distance travelled by the fan from about 400th to 4000th -- almost the same. Data is bad somewhere. – Linwood Aug 11 '17 at 19:10
  • 1
    I looked closer at the EXIF. There's an "Exposure Difference" in the Nikon section, which is interesting. In order from slowest to fastest it reads 2.1, 1.1, 0, -1.5, -1.8, -2.4, -3.2. I think those are stops, and indeed the shutter difference is (-2.7, -.7, 0, 1.3, 2.3, 3, 3.3) if 1/400th is taken as the zero point. I cannot find a good definition of this value, but clearly the camera knows the exposure was off. Did you post-process these shots? Would you consider doing it again and providing plain, RAW images? – Linwood Aug 11 '17 at 22:55
  • 3
    Note Active D-lighting: auto in the EXIF. This may explain the brightness puzzle, if not the blur. – mattdm Aug 13 '17 at 20:09
10

You're thinking about the speed wrong. At 1/4000th of a second, if it rotated nearly 4000 times per second, it would go all the way around in one frame. Talk about motion blur. :-)

It looks like the tip moved about three inches, so if we assume this fan is somewhere around four feet from center to tip (almost as large as they get), the circumference of the circle at the tip is about 300 inches. So it moved about 1/100th of a revolution in 1/4000th of a second. That should mean it spins at about 40 revolutions per second, or 2400 RPM.

That said, something is still off here. That's an entire zero higher than a real-world fan should ever spin, and then some.

According to UL code, the tip of an overhead fan can move at up to 3200 feet per minute, or 53 feet per second. At that speed, it should have moved 5/32nds of an inch in a single frame. This looks like it moved about three or four inches. To be fair, I realize that fans in India might not comply with UL rules for fan speed, but I kind of doubt they spin twenty times faster.

If you look at the metadata, I suspect that your camera overrode your settings. Many cameras have a feature called safety shift (or, I think the Nikon term is "Cybernetic override") that, in some modes, can change the settings from what you're expecting if the camera cannot achieve the provided setting in the available light. So if it could not shoot at 1/4000th because it couldn't crank the ISO up high enough and open the lens far enough, it might have shot somewhere more in the neighborhood of 1/200th (my best guess based on typical fan speeds).

Either that or that is one seriously fast fan.

  • Given how exposed it is, that is certainly not an ISO 6400 image at 1/4000 shutter and f/5.6 indoors. 1/200 or 1/250 seems about right for what the lighting looks like and what the exposure level is, given the ISO and aperture settings. – AJ Henderson Aug 11 '17 at 15:11
  • @AJHenderson exactly my concern. The settings i've written are what I've set, and the same is stored in the image data. But It doesn't look like the camera is able to do 1/4000. The problem is it's not just this case because of ISO. I have tried clicking water droplets in a really well lit setting. Yet the droplets have motion blur. I see youtube videos with droplet shots at 1/250s and I wonder why my camera can't achieve this with 1/4000 ! – AtulBhatS Aug 11 '17 at 16:14
  • 2
    @AtulBhatS I'd be concerned if the metadata still shows the faster shutter speed. I'd expect that if the camera was adjusting the settings to prevent a drastic under-exposure, it would write the values used in the meta-data as the selecting settings would be inaccurate and irrelevant. – AJ Henderson Aug 11 '17 at 16:32
  • @AJHenderson It does show 1/4000 sec in image data and in Exif, ExposureTime => 10/40000 – AtulBhatS Aug 11 '17 at 18:26
  • 1
    2400rpm does seem awfully high. – mattdm Aug 11 '17 at 23:23
3

I don't care what the EXIF might say, the camera appears to be firing no faster than about 1/250 sec.

Picture 1

It is worth noting that The original picture [Picture 1] is under exposed. Picture 1 is reported to be taken at:

  • f5.6
  • 1/4000
  • iso 6400

This is equivalent to +3.67 stops from full sun at Ev_100 = 15. So the exposure should produce neutral gray in when Ev_100 = ~11. A bright ordinary interior is about four stops lower at Ev_100 = 7 [a]. Making up those four stops in shutter speed is (1/2000 -> 1/1000 -> 1/500 -> 1/250). That seems consistent with the motion blur in picture.

Second Sequence

The second set of pictures (picture 2 through picture 8) were taken about a stop brighter at f4.2. Picture 2 is reported to be taken at:

  • f4.2
  • 1/60s
  • iso 6400

This is +10.75 stops from Ev_100 = 15. It appears to be 'more or less properly exposed'. This is not inconsistent with an indoor lighting condition of around Ev_100 = 5.

Picture 3 is slightly underexposed relative to Picture 2. Picture 4 through Picture 8 are all similarly exposed despite the report that they were taken across a delta of 6 stops of shutter speed. The motion blur, like the exposure is similar in all of them. Both these suggest that the shutter speed did not change much from Picture 3 to the end of the sequence despite what is reported.

Reference Image

For reference, this image was taken of a 42" ceiling fan at turning at its highest speed. 1/4000s froze the blades.

enter image description here

The settings were f3.5, 1/4000s, and iso 40000. This is +7.33 stops from Ev_100=15. I used Auto ISO. Because the fan was white 'proper automatic exposure' across the entire image could occur at a lower ISO than if the fan were black. The fan is less than 10 feet from a large window and the photo was taken daylight hours.

[a]: An ordinary residential interior might be Ev_100 = 5-6.

  • So you suggest my camera needs to be fixed? – AtulBhatS Aug 15 '17 at 16:43
  • 2
    @AtulBhatS My guess is that: 1. it's either a setting, perhaps related to flash given that 1/250 is the maximum syc speed. 2. It is a built in feature of the camera because the D3400 is positioned as an entry level camera and the number of people who will say their D3400 isn't working when the fan blades are blurred like here is probably much much lower than the number of people who will think their camera is broken when they get a black image from shooting 4-7 stops below ambient. But I lean toward a setting overriding the shutter speed. – user50888 Aug 15 '17 at 17:42
  • @AtulBhatS: You are not mentioning flash, i assume it is off? – Orbit Jan 22 '18 at 16:47
2

The maximum speed of a ceiling fan is about 250 RPM. That works out to about 4 RPS. Given a 30 inch diameter fan, the circumference is about 94 inches. That means the tips of blades travel 94 X 4 = 276 inches in one second. In the time span of 1/4000 of a second the tips will travel 276 ÷ 4000 = 0.7 inches or 17.8mm

To me, the degree of blur seems greater than that caused by a 17.8mm movement. My conclusion is, the shutter did not actually operate at 1/4000 of a second. You next step, figure out how to read the imbedded EXIF file that your camera writes when it takes a picture. You can do this will most any good image editing software or download an app that will read this data.

Some additional gobbledygook: A focal plane shutter works by moving a curtain with a slit across the span of the imaging chip. The travel time of the shutter is far greater than the shutter speed setting. That’s unlikely the source of this blur because the actual shutter speed is just the time it takes for the slit to travel its own width. What I want you to know is; a focal plane shutter can encourage weird things. As an example, at a baseball game, a high shutter speed is used and this is sufficient to freeze all the players in place. However, by coincidence, the batted baseball is blurred in the picture. This happen because coincidently, the speed the image of the ball exactly corresponded to the shutter slit movement speed across the frame. Not saying this is what happened here, just pointing out photography has some built-in gibberish.

  • I did see the slit mechanism in some slow motion videos and I checked my camera the same day for the same. I did not find any slit moving over the sensor. The mirror just lifts up and exposes the sensor directly. There is nothing that covers the sensor after the mirror. Does it mean anything about the current issue? Do you suggest I contact the service center and ask them about the issue? Or is it something to do with my camera generally? – AtulBhatS Aug 11 '17 at 16:23
  • 1
    @ AtuBhatS -- Nevertheless this camera sports a vertical travel focal plane shutter. – Alan Marcus Aug 11 '17 at 16:42
  • @AtulBhatS There actually is a shutter mechanism there. If you look by using the camera's "raise mirror for sensor cleaning" function, it will also open the shutter all the way. – mattdm Aug 11 '17 at 17:31
  • A stuck shutter would almost certainly stick shut, not open. And if one of the shutters stuck open, either the top or bottom of the image would be massively overexposed in a continuous gradient. If both of the shutters stuck open, the entire image would be massively overexposed. This is almost certainly a configuration issue (e.g. safety shift). – dgatwood Aug 11 '17 at 18:06
  • @dgatwood any idea how to disable the configuration and get manual exposure as set? – AtulBhatS Aug 11 '17 at 18:29
1

Finally!

The Nikon Service center was able to solve it after trial and error replacing each and every part. The replacement that worked was with Aperture Control which even the service guy was not sure was wrong.

But it did the trick!

0

It looks like you used the built in flash on the camera. If you did then the Shutter Sync would drop to 1/200th. That is what is preventing you from taking higher shutter speed images. Don't use the flash and instead increase the lighting in the house somehow and take the same image at the higher shutter speeds and it should look frozen.

  • I don't see any evidence the camera's flash was used. The fan would be bright, and it would have definitely appeared frozen as opposed to blurred. – scottbb Aug 12 '17 at 18:58
  • Exif says the flash wasn't used. Good theory though. (At full power, many flashes have duration as lon as 1/2000th) – mattdm Aug 12 '17 at 19:04
  • Camera flash would also produce shadows on the ceiling, which is not seen in original image. – Grimaldi Sep 3 '17 at 16:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.