Flash and small sb question

Hope this isn't a dumb question but here goes.

I'm using a 7D with a 24-105 lens, a 580ex2 flash and a small 4" by 4" softbox. At 24mm everything works fine but at 105mm I have to increase the flash output by +3 or everything is black (almost). I was shooting the same wall with only a slight natural backlight. With the softbox removed both photos are as expected. The flash was set to ettl with the zoom on. I have not noticed this happening when using the 28-135mm lens. I thought the ettl would compensate for the softbox. What am I doing wrong.

Lens blocking the flash maybe? Never used the 24-105mm. Some pictures would be nice.

ETTL and evaluative metering could be the issue as when you zoom it's getting a different exposure reading. Are you maintaining the same distance from the wall or changing distance to get equivalent framing? Like bobbyz said it would help if you could post pics.

Not sure what your specific issue is. I will state that PaceAce and I recently did testing with different flash units on different Canon dSLR models, all using a 5x7" softbox. We found that ETTL seems in general to incorrectly deal with ETTL flash when placed in a softbox (vs. simply aiming head up), and the result can be OVER- or UNDERexposure...we think which one will occur depends upon the lens which is mounted! One thing we did not test was whether taking the flash head off 0 degree (normal) position would correctly alter the exposure results with the softbox.

Wilt wrote in post #12056752
Not sure what your specific issue is. I will state that PaceAce and I recently did testing with different flash units on different Canon dSLR models, all using a 5x7" softbox. We found that ETTL seems in general to incorrectly deal with ETTL flash when placed in a softbox (vs. simply aiming head up), and the result can be OVER- or UNDERexposure...we think which one will occur depends upon the lens which is mounted! One thing we did not test was whether taking the flash head off 0 degree (normal) position would correctly alter the exposure results with the softbox.

That's interesting, Wilt. Did you publish any results or have numbers to illustrate the loss or change in accuracy? Or have you not pinned it down enough to draw any definitive conclusions yet?

There is some information around that says that some lens provide focus distance information that the ETTL-II algorithms use as an input. It may also depend somewhat on whether one has zoom set to manual or auto. Also, it would pay to make sure that the head has not inadvertently drooped to the macro position because of the weight of the mini-softbox.

I use TTL a lot but always with the flash pointed away from the subject and bouncing. I have very little problem with over or under exposure. I have camera set for ETTL Evaluative and use spot metering. I also use the center focus point.

I've purchased an tried a couple of the small softboxes, they are just too small and eat too much light to be of any value to me.

I just did this test series. Ambient was insignficant, 1/6 f/4 ISO100. Camera (40D) in Manual 1/80 f/4. Flash Metz 54MZ in ETTL mode, ETTL Averge mode set in CFn. Focus zone was on gray card (not center). All shots shown with identical LR parameters (match shot 1 of series for Exposure and Brightness and other parameters). Sequence:
1. Flash head direct, autozoom
2. Flash head ceiling bounce
3. Flash head direct with Wescott Micro Apollo softbox (5" x7")
4. Flash head up 10 degrees (to lose distance info) with Micro Apollo softbox

Lens: Canon 17-55mm at 35mm

Now the second half of the test series.
All shots shown with identical LR parameters (match shot 1 of series for Exposure and Brightness and other parameters). Sequence:
1. Flash head direct, autozoom
2. Flash head ceiling bounce
3. Flash head direct with Wescott Micro Apollo softbox
4. Flash head up 10 degrees (to lose distance info) with Micro Apollo softbox

Lens: Tamron 28-75mm at 35mm

Here are two photos of what I was asking about. The first at 105mm the second at 24mm. Everything else is the same. The softbox is not blocking the lens or the flash sensor.

HOSTED PHOTO
please log in to view hosted photos in full size.

HOSTED PHOTO
please log in to view hosted photos in full size.

That result is truly bewildering, Dan! Please try to post photos with embedded EXIF intact (do not use 'Save for Web' option when exporting)

These two are with the flash tilted about 45 deg. This angle would not allow the flash to bounce off the ceiling.

HOSTED PHOTO
please log in to view hosted photos in full size.

HOSTED PHOTO
please log in to view hosted photos in full size.

Sorry about the exif data. I process the raw files with the exif data intact. It seem to be stripped when I upload the photos.

TMR Design wrote in post #12056783
That's interesting, Wilt. Did you publish any results or have numbers to illustrate the loss or change in accuracy? Or have you not pinned it down enough to draw any definitive conclusions yet?

Still attempting to isolate the variable! Leo got mostly underexposure with softbox, but he did also see some overexposure in some cases...same camera, same flash, overexposure with 16-35mm and 35mm lenses, underexposure with 24-70mm lens and 24-105mm lens. I used to get underexposure with same flash and same modifier with my 20D, now I get overexposure with same flash and same modifier with my 40D.