I'm looking at some of my negs from an indoor wedding on 12-30-00 and some are kind of thin. I shoot a 50mm f/1.8 in manual mode at 1/45 and f/5.6 with NPH at 250 ASA. I use a Quantum 1+ battery, so no problem there. The dress just looks a little dingy and the blacks aren't quite black. The weirdest thing is that some look really nice. At any distance I might add. 6 feet or 15 feet. That shouldn't be a concern, though. The 550 is good for more than 60 feet with 250 ASA. Some negs at longer distances look great; others up close don't. I told the lab to redo some that were grainy and they gave me a note stating the negs were thin. Could this be an equipment problem? I just can't wrap my mind around this at all. I shouldn't have any trouble. I don't recompose and move the AF sensor off the subject anymore. That used to cause me some problems. If I have to recompose, I use FEL. Rarely though. I don't have a scanner, so just imagine the dingy dress and not quite black tuxes. :)

-- Colin Miller (ckmiller@pond.net), January 13, 2001

Answers

Update: I got out my L-508 and took some readings. Flash on camera, E- TTL, all modes (even that shitty green box mode), all different kinds of FEC and here's my results.

The flash does not work!!!! With up close shots and at like f/11, the meter said f/5.6. Every time. It seems like the flash is locked into a sort of manual output mode, except the output stays the same no matter where I stand. 5 ft, 10 ft, even 15 ft. The meter always read right at f/5.6 and 1/10. What is going on?!?!?!?!?!?! The FEC did not affect the meter reading ever. I loaded the camera with Gold 100, set the camera to 100 ASA, set the meter to 100 ASA, and proceded with the shooting. No change at all. When shooting in Tv mode an 0 FEC, the camera read 1/10 and f/2.8. and the meter came back at f/2.8 & 3/10. It was the closest out of any of the tests. I can't shoot a wedding at f/2.8. What do I tell the poeple at the end of the semi-circle group shot? Sorry, my camera can't keep you in focus. I am really pissed and I think the 550 is going back to Canon. I tried my 380EX also and the same thing. Incorrect readings for what the camera settings were. Please help!!!

-- Colin Miller (ckmiller@pond.net), January 13, 2001.

You have film in the camera right?

I know that measuring TTL or A-TTL metering needs film (not the pressure plate) to reflect light onto the TTL flash sensor(s). Not sure about E-TTL though. The pre-flash shouldn't need film. I don't know if it also monitors light reflected off the film for any reason.

-- Bob Atkins (bobatkins@hotmail.com), January 13, 2001.

I wish I had a solution, but I don't. Bob seems to be the expert on Canon flash, so I guess the next step would be to contact Canon. I don't even want to speculate on what the problem could be, but I guess since the 380EX did the same thing, it could be a camera problem. I know with my Elan IIE and 380EX I used to dial in 1/2 stop of FEC all the time with print film. Other than times when I was using Portra 400 rated at 320 or Portra 160 rated 100 the negs always seemed thin if I didn't. I haven't noticed the same problem with the 550EX, except when I tried it out recently with Portra 400 rated 400, and even then, I was using it at the extreme limits of flash range, so I expected slightly thin negs. Obviously something is wrong somewhere. Sorry I can't be of more help.

-- Brad Hutcheson (bhutcheson@iname.com), January 14, 2001.

I am FAR from any sort of expert on the Canon flash systems. To make them simple to use they are quite complex and Canon have never published any detailed specs on the newer system. The old "speedlite Reference Guide" was great, but only covered up to the 10s and 430EZ. The new EX series flashes and bodies like the ElanII, 7, 3 and 1v are still something of a mystery sometimes.

I've found flashmeter readings pretty much agree with in camera flash settings (ElanII and 380EX or 540EZ) as long as I remember to test with film in the camera(!) and I'm shooting a neutral (18%) gray equivalent target. Obviously if I'm shooting a white (or black) wall, the readings differ because the camera has no way of knowing what color the wall really is. It assumes 18% gray.

-- Bob Atkins (bobatkins@hotmail.com), January 14, 2001.

It PROBABLY doesn't matter if you had film in your camera when doing this test since E-TTL metering uses the regular (ambient) sensor to determine the flash exposure from the reflection of the preflash before the shutter opens anyway.

But it seems to me that if both the 550EX and the 380EX - "I tried my 380EX also and the same thing." - then I suspect that the faulty flash metering is a problem with your camera and not the flashes. After all the metering is done in-camera not in the flash.

-- Jim Strutz (jimstrutz@juno.com), January 14, 2001.

You didn't mention which metering mode you were using and that's probably the key. It's my experience that the only way to get absolutely consistent results with E-TTL is to use spot metering, FEL, and FEC. This would be particularly true for weddings where subjects will be wearing blacks and whites and the b/g could be anything.

For AVERAGE subjects with an average background, it is possible to use other modes and get great results but when your scene is constantly changing due even to changing focal lengths, spot metering is the only way that works consistently.

I'm using two 550's for portraiture and finally gave up and went to manual for the flashes (FEL was too much trouble) but I guess you're not going to be able to do that.

Here's a link to some shots taken with my setup. I most likely used spot metering for the individual subject and partial for the groups.

www.columbusoft.com/pf/photos.htm

Paul

-- Paul Ferrara (paul@columbusoft.com), January 15, 2001.

Ok. Another day has passed and I've calmed down a bit. More tests tonight indicate that FEL on a 18% gray card works very well. But seems to work only at 10 ft. I'll retets for that. I can't have my assisitant holding a gray card in front of a b&g for every shot. So if I FEL on her dress, I should add +1.5 or 2 stops of FEC. If it's him alone, I should activate -1.5 or -2 stops of FEC. I am appalled I never noticed this before. Please tell me I'm on the right track. If I have to use FEL for every shot I will. Otherwise, setting up the flash stationary, setting to manual output, and taking a handheld meter reading will be my best bet.

-- Colin Miller (ckmiller@pond.net), January 15, 2001.

Colin... When you were initially testing your ex flashes against you handheld meter... two things. 1) Already mentioned by others (you need film in camera for OTF reading) 2) The fact that you were getting the same 5.6 reading makes me assume that the flash meter was metering the preflash, (which probably is constant) and not the exposure flash. Regards... db

-- David Bindle (david.bindle@usask.ca), January 15, 2001.

Colin, you're correct. My approach was to take a non-flash reading from the subject's face, then from a gray card, then dial in the difference as FEC. From that point on, I'd FEL on the subject's face for each shot.

I think it would have been great if there was a way to lock in flash exposure for an entire session. That's a little off topic for your situation since you're probably moving around a lot but for portrait sessions it would have been perfect.

Paul

-- Paul Ferrara (paul@columbusoft.com), January 15, 2001.

You might also want to try placing the EX flash into TTL (rather than E-TTL) mode, since there may be situations in which E-TTL is more trouble than it's worth.

-- kurt heintzelman (heintzelman.1@osu.edu), January 15, 2001.

You might also want to try placing the EX flash into TTL (rather than E-TTL) mode, since there may be situations in which E-TTL is more trouble than it's worth. Kurt, if I'm not mistaken, E-TTL was a big help for outdoor fill and not so much for indoor work. If that's true, TTL would be fine. I ran some more tests in TTL, and the results are better obviously, but still not consistent. I got -.5 stops of FEC when I dialed in -2 FEC. *sigh* I will solve this dilemna one day, I am sure of it. Plus I'm switching labs right now, so that's a true headache.

Again, if I FEL a black tux I should dial in -2 FEC and the +2 FEC is for FEL off of the dress? If I can do this right, my troubles may be over. I will always have a black or white "blob" in the frame at weddings.

Thanks again for helping me out. These are always the toughest problems to handle via q&a forums. Everyone has done their best to help. Even if it meant asking me 4 times if I loaded the camear with film.... ;)

-- Colin Miller (ckmiller@pond.net), January 15, 2001.

er, camera with film......

-- Colin Miller (ckmiller@pond.net), January 15, 2001.

I opened this thread because I've been experiencing the same phenomenon with my 550EX on a D30 (and yes, the CMOS sensor is in :- ) It's been driving me crazy. I just got the EX so I haven't done nearly the scientific testing that Colin has. I sure hope one of us figures this out because the current results are very unacceptable.

-- Doug Cornell (ubudoggy@yahoo.com), September 27, 2001.