2nd curtain and freeze motion

[RobWatson]

Color me befuddled. Flash set to second curtain. Seems not to matter TTL or manual ...

Set the widget swinging from left to right - snap at the bottom of the arc for maximum speed.

Want to see a slight motion blur trailing. So what the heck is the leading blur from? Only thing I can come up with is a delay between the flash firing (long decay tail?) and the 2nd curtain actually closes.

Tried three flashes all get the same effect (FL-300R, YN560 & Vivitar 3700).

Subscribe to see EXIF info for this image (if available)

 

[Cruzan80]

Flash firing at dead center (start of motion of second curtain), leading blur from the curtain having more to travel before reaching the leading blur section of the photo? Not sure exactly what camera you are using, or what direction the curtains move (L->R, T->B).

 

[RobWatson]

Camera E-P1 sitting in the "normal" position so I expect the curtains move up/dn. I always though the 2nd curtain moved dn but I don't know why I think that ...

 

[davidzvi]

I've noticed this not working as I would expect either. 2nd curtain should fire at the end of the exposure. And maybe it's just because I'm use to doing this on my DSLRs. Like this:

 

[RobWatson]

I've noticed this not working as I would expect either. 2nd curtain should fire at the end of the exposure. And maybe it's just because I'm use to doing this on my DSLRs. Like this:

That looks just like I would expect. A bit of trailing motion blur and a freeze at the end.

Just a bit too much ambient so quite a bit of blur.

 

[Cruzan80]

If you are correct, and the curtains start at the top of the photo, and move down the frame, 2nd curtain sync would mean it would fire either at the start of the second curtain moving (allowing residual light to cause the streak in front of the pendulum), or it is firing at some arbitrary middle point, again allowing the streak, but less of it due to the remaining time the shutter has. Try flipping the camera 180 deg. and see if the top shows any movement (even at 100% crop)? If so, there is your cause. The only remaining thing to try and test is where it fires on the second curtain. You could put smaller fishing weights at 1/4 or 1/3 stop sections of the string, and see where it starts to "leading trail". I ave a feeling this is going to be manufacturer dependant, but may be the same for the Pen and OMD (or may not due to different shutters).

 

[Rudy]

Color me befuddled. Flash set to second curtain. Seems not to matter TTL or manual ...

Set the widget swinging from left to right - snap at the bottom of the arc for maximum speed.

Want to see a slight motion blur trailing. So what the heck is the leading blur from? Only thing I can come up with is a delay between the flash firing (long decay tail?) and the 2nd curtain actually closes.

Tried three flashes all get the same effect (FL-300R, YN560 & Vivitar 3700).

Subscribe to see EXIF info for this image (if available)

What shutter speed were you using?
To get a decent effect, you would need to be pretty far away from the max sync speed.
Rudy

 

[davidzvi]

That looks just like I would expect. A bit of trailing motion blur and a freeze at the end.

Just a bit too much ambient so quite a bit of blur.

Well for that one I was also panning a little.

If you are correct, and the curtains start at the top of the photo, and move down the frame, 2nd curtain sync would mean it would fire either at the start of the second curtain moving (allowing residual light to cause the streak in front of the pendulum), or it is firing at some arbitrary middle point, again allowing the streak, but less of it due to the remaining time the shutter has.

I would think it would not be arbitrary but instead be linked to shutter speed and flash duration. But not sure.

 

[RobWatson]

I'm thinking a 180 rotation won't do anything ...

If I can confirm exactly which direction the 2nd curtain moves to close then a portrait mode orientation with the 2nd curtain moving in the opposite direction to the motion would stop/reduce the leading blur.

Suppose I gotta lower ambient so at the speeds of interest there is not enough light to capture the blur - kind of defeats the whole purpose/effect I'm going for.

Funny thing is in my water drop sets (particularly the image in post #2) there is no trailing blur but only a leading blur ... as if the flash has a very long and modestly bright decay.

 

[RobWatson]

What shutter speed were you using?
To get a decent effect, you would need to be pretty far away from the max sync speed.
Rudy

That is probably the case - pretty much was working close to synch speed. That is the shutter speed that gave me the level of blur for the motion I wanted. A longer shutter speed (and lower ambient) kind of means no trailing blur ... that is what I'm trying to capture is a bit of blur to encourage the perception of motion.

 

[Cruzan80]

In this case, I meant arbitrary in the sense of "wherever the engineer decided it to be programmed". Not arbitrary in changing depending on circumstances after the camera shipped. Sorry about the confusion. For shutter speed, anything that has the flash firing after the first curtain has cleared wouldn't matter. If it were on 2nd sync with the first still in the frame, there is the possibility of seeing it start a leading trail in front of the object with a break in between (time it took first curtain to clear). Flash duration for any hotshoe flash would be over 1/250 to get to t.1, at least, so I am not sure this would matter unless using flash and 2nd curtain sync at a much higher shutter speed.

 

[davidzvi]

That is probably the case - pretty much was working close to synch speed. That is the shutter speed that gave me the level of blur for the motion I wanted. A longer shutter speed (and lower ambient) kind of means no trailing blur ... that is what I'm trying to capture is a bit of blur to encourage the perception of motion.

When I shot for the effect at events I'm in the 1/8 to almost a second range. That image was 1/3 at f14, ISO 2000.

 

[Cruzan80]

I'm thinking a 180 rotation won't do anything ...

If I can confirm exactly which direction the 2nd curtain moves to close then a portrait mode orientation with the 2nd curtain moving in the opposite direction to the motion would stop/reduce the leading blur.

This is what I was meaning to go for, when suggesting the 180 flip. Also, it looks like the trail is longer in the back than the front, so if flipping shows front longer than back, it is reproducable, and you know which way to orient it in portrait for which effect.

 

[kwalsh]

This can be a real pain to get right and as already pointed out when you get close to the sync speed it just doesn't work hardly at all. When I've actually needed a motion blur on a single object I've just cheated:

Subscribe to see EXIF info for this image (if available)

In this shot the Mentos was frozen by flash with no ambient light. Added a motion blurred copy of it on a layer and also adjusted the WB of that layer to be warmer like you'd expect from an ambient blur. I tried repeatedly to get a 2nd curtain sync to look decent and it was just not happening no matter how much I adjusted ambient light and flash. Adding a blur layer was easy as the image was already layered anyway (obviously I wasn't about to drop Mentos into Diet Coke right in front of my camera, the drops were without the Coke in the scene).

Getting any sort of significant effect requires shutter speeds like 1/15 or longer and the right blend of ambient light and relative motion.

 

[Cruzan80]

But I want to see the shot about 10 sec. later!!! With no layers...:evilsmile:

 

[RobWatson]

Getting any sort of significant effect requires shutter speeds like 1/15 or longer and the right blend of ambient light and relative motion.

I was afraid of this ... that means for some motion there is no hope except as you suggest. Otherwise a more complicated lighting setup like a light that can be turned just as the 2nd curtain flash fires.

Maybe a switch connected to a light sensor so the light can be thought of as a slave to the flash except it turns off instead of on when the 'master' fires. Hmmm ....