7D wireless flash trigger can't expose

[apersson850]

Although the built-in wireless flash trigger in the 7D is a very handy feature, it seems it can't expose properly.
Take a look at these two images. They are both shot with the master flash turned off and with two 580 EX II slaves, A and B, each illuminating the subject indirectly from the left and right.
In these two images, one is shot with the internal flash as a master, the other with yet another 580 EX II on the camera, acting as a non-contributing master.
Flash power is in both cases set manually (1/8 ) for both A and B.

The camera is in M mode, so shutter speed and aperture doesn't change. One could expect that the two images to be virtually identical, right? And indeed they are. There is hardly any point in telling which uses internal master and which uses external, since they are both OK.

[apersson850]

Now let's switch to E-TTL. A subject like this shouldn't pose any insurmountable problems for the camera to expose properly by itself.
Just like above, the master is set not to illuminate, but just the two slaves, which remain in the same positions as they were above.
All settings are of course the same for the internal and external master.

But what's up now? One image is handled by the 580 EX II being the master, and it does the job very well. But look what happens when the internal flash is a master. Completely overexposed.

Why? Is it more than me seeing this?

I could understand if E-TTL couldn't do it, being fooled by the dark fabric behind Kajsa (the cat), but that can't be it, since then it wouldn't work with the 580 EX II as master either. It's the camera metering, not the flash.

[PacAce]

Is this happening to you consistently? I've had this overexposure issue happen but only when using the Metz 58AF flash in Group B or Group C (Metz seems to work fine in Group A). I've tested the 7D with the 550EX, 580EX and 580EX II and all seemed normal with no overexposure issues.

[apersson850]

For this particular setup you see in the images above, yes, it's consistent. Even in E-TTL, I have to set the proper flash exposure by adjusting ISO and aperture, just like if the slaves were in manual mode.

I'll have to test this more to see if it's consistent, of course, but I was interested in hearing if somebody else can see the same thing. If it's only my camera I'll have to claim warranty issue, but if it's all of them, there's no point in anything else but waiting for a firmware update. It does work with the 580 EX II as the master, so it's not that the camera is unable to meter the flash properly.

I just tried a series of shots, and it's at least consistent. Using the internal flash as an E-TTL master doesn't work well. Evaluative metering, average or Flash Exposure Lock is all the same. Horrible overexposure. Since when setting the slaves manually, 1/8 power gives a good exposure, it's not that the ISO is too high or aperture too large or anything. Besides, it does work with the 580 EX II as a master.

[PacAce]

Originally Posted by apersson850

For this particular setup you see in the images above, yes, it's consistent. Even in E-TTL, I have to set the proper flash exposure by adjusting ISO and aperture, just like if the slaves were in manual mode.

I'll have to test this more to see if it's consistent, of course, but I was interested in hearing if somebody else can see the same thing. If it's only my camera I'll have to claim warranty issue, but if it's all of them, there's no point in anything else but waiting for a firmware update. It does work with the 580 EX II as the master, so it's not that the camera is unable to meter the flash properly.

If you give the wireless settings you used on the camera, I can try to see if I can replicate your exposure problem. It might just be a certain setting that causes this which I may not have tried.

[timbop]

I was just playing with mine with no trouble. Any chance you accidentally had + FEC dialed into the camera?

[apersson850]

I tried another subject, where instead of bouncing both slaves in the ceiling, they illuminate a sofa directly. Flash A from the left, B from the right.
In this case, where I had to set flash power down to 1/64 when testing it with manual flash power, both versions of E-TTL (internal and external master) renders the same exposure.

External flash:
Mode E-TTL II
Sync 1st curtain
FEB zero
Evaluative
Zoom 24 mm
Wireless on
Main flash off
Channel 1
Firing group A:B
Exp comp flash zero
Relation A:B 1:1

When used for the manual comparison, same as above, except:
Mode Manual
Firing group A+B+C
Group A power 1/8

Internal master:
Mode E-TTL II
Sync 1st curtain
FEB zero
Evaluative
Wireless External only
Channel 1
Firing group External (A:B)
Relation A:B 1:1
External flash exposure comp zero

When used for the manual comparison, same as above, except:
Mode Manual
External flash power 1/8

But as it seems setup dependent, maybe photos of how I've located my flashes, and the cat, would be more appropriate?

[PacAce]

Originally Posted by apersson850

I tried another subject, where instead of bouncing both slaves in the ceiling, they illuminate a sofa directly. Flash A from the left, B from the right.
In this case, where I had to set flash power down to 1/64 when testing it with manual flash power, both versions of E-TTL (internal and external master) renders the same exposure.

External flash:
Mode E-TTL II
Sync 1st curtain
FEB zero
Evaluative
Zoom 24 mm
Wireless on
Main flash off
Channel 1
Firing group A:B
Exp comp flash zero
Relation A:B 1:1

When used for the manual comparison, same as above, except:
Mode Manual
Firing group A+B+C
Group A power 1/8

Internal master:
Mode E-TTL II
Sync 1st curtain
FEB zero
Evaluative
Wireless External only
Channel 1
Firing group External (A:B)
Relation A:B 1:1
External flash exposure comp zero

When used for the manual comparison, same as above, except:
Mode Manual
External flash power 1/8

But as it seems setup dependent, maybe photos of how I've located my flashes, and the cat, would be more appropriate?

I'll give your setup a try with bounce. I've tested a similar setup previously but using direct flash lighting. I haven't tried bounced lighting yet so I'll give that a shot this time.

And, yes, it be helpful to get the physical layout of the flash placement relative to the subject and camera, too, so that I can try to replicate that as well.

[OldA1]

A customer came into the store yesterday with a similar problem using the pop up and one 420EX. The pop up was blasting the photo causing overexposure.

I have used the ST-E2 in the past and as a new 7D owner, I too am playing with the built-in wireless features.

I reviewed his settings and did not see anything that jumped out as being a WTF is that set for. I did not check what firmware he had, all I did was change the 420EX to group B and the test shots came out fine.

Peace
Mark

Originally Posted by apersson850

Now let's switch to E-TTL. A subject like this shouldn't pose any insurmountable problems for the camera to expose properly by itself.
Just like above, the master is set not to illuminate, but just the two slaves, which remain in the same positions as they were above.
All settings are of course the same for the internal and external master.

But what's up now? One image is handled by the 580 EX II being the master, and it does the job very well. But look what happens when the internal flash is a master. Completely overexposed.

Why? Is it more than me seeing this?

I could understand if E-TTL couldn't do it, being fooled by the dark fabric behind Kajsa (the cat), but that can't be it, since then it wouldn't work with the 580 EX II as master either. It's the camera metering, not the flash.

[apersson850]

Here are the two setups. In the top one, you can see the A flash in the window to the left and he B flash on the TV to the right, bouncing into the ceiling both of them. This is the one that does not work.

In the second case, the two flashes are on two chairs, A to the left and B to the right, illuminating the sofa where the dark blue plaid is hanging down. This works well.

It's repeatable too. After testing directly onto the sofa, I put the flashes back in the first position to take the picture of the setup. Tried it again on the cat once again. Same result, horrible over-exposure.
Tried it with the 40D, where I had to have the 580 EX II as a master, of course. Works perfectly, just like that configuration does on the 7D.

[apersson850]

Originally Posted by OldA1

all I did was change the 420EX to group B and the test shots came out fine.

OK, but having both A and B works fine with the 580 EX II as a master, and should work with the internal as a master as well. Which it indeed does when aiming directly towards the target at close range, but not when bouncing the ceiling at also pretty close range.

I can see what happens if I let them all be one group.

[timbop]

I had just one 580 set up as slave on A, bounced off the ceiling. With the on-7d flash set to just trigger the 580 - worked fine, with a nice exposure. I'll try again with 2 flashes

[apersson850]

I just tried with the two flashes being A and B, or both B, and combined that with the firing group being A:B or A+B+C. Same result. Excessive overexposure.
Now I'll try going down to one flash only.

[timbop]

OK, just tried it with 2 different 580's (1 A and 1 B) bounced: no problems. A:B ratios work too. In the "internal flash control" menu I have it set for the flash icon only, no FEC, and no FEC disaled into the strobes. I think the key to your problem might be in the camera's settings

[apersson850]

But the settings are just as listed above.
When taking a photo of the cat, with just one slave flash illuminating her via bounce in the ceiling, I get the same overexposure. Switching to manual and 1/8 power, just like I tested before, gives decent exposure.

I also verified that the slave flashes do indeed switch between E-TTL and M mode, so they do get the command from the master to switch mode.

However, if I now in E-TTL mode change from ISO 500 and f/5.6 (which works with manual flash power) to ISO 100 and f/8, I get a reasonable flash exposure with E-TTL.
If I do the same ISO/aperture maneouvre with a 580 EX II as a master, I get two images with almost identical exposure (the ambient is slightly different, but that's as expected).

Now you having 7D cameras that work, are your's earlier or later than mine? My serial number ends with 100711, I think. You can see it in the Exif in the images above.