ghosting from preflash?

[aparis99]

I've noticed lately that some of my images, when using my 580exII, seem to have both the preflash, and the full flash, producing a bit of a ghost around the subjects.

Anyone know if this is normal or how i combat this?

[Bearmann]

I don't think you can have the pre-flash in the photo because I don't think the shutter is open when the pre-flash occurs. Probably what you are seeing is a mix of flash plus ambient light.

[Joelene]

Are you on second curtain sync?

[Titus213]

Can you post and example with exif?

I would suspect you are using Av and/or the shutter speed is too low.

[Wilt]

I don't think you can have the pre-flash in the photo because I don't think the shutter is open when the pre-flash occurs. Probably what you are seeing is a mix of flash plus ambient light.


^^ Right, the shutter is not open at the time of the preflash

[CliveyBoy]

Not during the main part of the preflash - the exposure testing and the slave flash setup - but does not the "Fire" instruction require that it is emitted by the master at the moment the shutter is open? I understand from Radiopopper's reading of the Canon patents that it does just that.

[Mark Vuleta]

Aren't the ETTL control signals IR & shouldn't be visible (or have so little effect in relation to the main flash output)?

How are things Clive?

[DC Fan]

Often, this comes from relatively bright ambient light and camera or subject movement when the standard 1/60 sync speed is used, not from pre-flash.

[CliveyBoy]

Hi, Mark, and how are you? I miss meeting up with y'all.

An ST-E2 emits a "near-IR" signal, which can be a reason for using that device. However the 550EX, 580EX and 580EXII use the main flash tube, producing visible light. The power is low, but can have an effect in low-light conditions.

The effect is reproducable, if anyone wants to test it.

See post 9 in:
http://photography-on-the.net/forum/showthread.php?t=715733

[mmahoney]

I used to think that the pre-flash never showed up in exposures until someone suggested that it did, and said to prove it by setting up two 580's in a master / slave config with the master flash turned off, then take a pic of both in a mirror.

I've not tried it but this poster said they did, and that both flashes showed up in the exposure.

[apersson850]

That "someone" was me. And yes, they both (master and slave) are on in such a picture. See the link above.

Should I post a sample picture?

On the other hand, the OP never stated that he uses Canon's wireless system. If his gear list in the signature is true to the letter, he has only one 580 EX II.

[aparis99]

Hmm... I think its b/c i was at 200mm, using 1/60th of a second, and that's too slow for 200mm...

that's what it is im sure... thanks alot guys!

[apersson850]

If it matters or not depends on the strength of the ambient light, relative to the flash light.

[vincent_su]

If it matters or not depends on the strength of the ambient light, relative to the flash light.

I agree and I have done the same. At slow shutter speed, with enough ambien light, the ghosting is caused by the camera shake or subject moving.

Dispite 1st or 2nd curtain sync, it will happen under the description above.

With a 200mm FL, a safe bet is to set TV mode at 1/200 or 1/250 shutter speed.

[apersson850]

The curtain you sync with is totally irrelevant in that case, for sure.

[CliveyBoy]

the OP never stated that he uses Canon's wireless system. If his gear list in the signature is true to the letter, he has only one 580 EX II.
AFAIK, if the 580EXII is set to master, it will perform a Master's job without knowing whether there are any slaves within communication range.

For example, a windowlight shot may need the background lifting. Put a 580EXII on-camera as Master, with the main flash OFF. In the background, set a 550EX on the ministand, as Group C. Set the Master A:B C ratios to anything.

In this case, the preflash exposure tests from Group A and Group B will tell the camera that there are no such flashes; it does not test-fire Group C. As far as the camera knows, there are no flashes (if indeed it bothers to deduce that). However, when the shutter has opened, the master will still emit a "Fire!" coding and the Group C flash fires.

So, I conclude that if you have set an on-camera flash as master, it will behave as such, regardless.

[apersson850]

In E-TTL mode, the master will order group C to pre-flash as well. Otherwise, it can't know what power ratio it's supposed to tell group C to fire with, in order to achieve the FEC you've set for group C.

But you are right in that the master will fire regardless of whether there are slaves responding or not. However, I don't really see the benefit of using it in master mode without slaves?

[CliveyBoy]

In E-TTL mode, the master will order group C to pre-flash as well. Otherwise, it can't know what power ratio it's supposed to tell group C to fire with, in order to achieve the FEC you've set for group C.

But you are right in that the master will fire regardless of whether there are slaves responding or not. However, I don't really see the benefit of using it in master mode without slaves?

My recollection was faulty - thanks for the correction.

There is no benefit, but someone learning to use Canon Flashes could have it set as Master because they fiddled. Even the learned might forget previous settings.

[apersson850]

That's true, of course. I've done it myself. Took two pictures before I looked at the flash's display and saw what was wrong.

At least the new 580 EX II immediately leave slave mode, if you put them on a camera and do the half-press.

[CliveyBoy]

At least the new 580 EX II immediately leave slave mode, if you put them on a camera and do the half-press.
Surprised at this, so tested a 580EXII on my 50D. The flash stayed exactly as set, and the 50D's menu control adopted the settings of the flash. Even when I set the camera to Wireless Mode Disabled, then attached flash to camera, turned both on, and fired a shot, the Master mode remained displayed on the Flash. The same when using FEL button.