YN-468 II overexposes when using iTTL

I just received a replacement YN-468 II that has the same issue as the first one I purchased.

When using it in iTTL mode, a majority of the pictures are overexposed.

Sometimes the exposure is good, but 9 out of 10 times the picture is overexposed.

I am using the flash on a D5100, but I have had the opportunity to try it out on a D90, but it still did not work.

I have tried numerous settings - Auto, Shutter Priority, Aperture Priority, changing the ISO, etc...

I have tried new batteries, as well as eneloops.

I have tried an SB-600 on my camera and it seems to work well.

I have noticed that if I bounce the flash the exposure is much more consistent, and it looks fine (by viewing the histogram).

Any suggestions as to what I might be doing wrong?

Has anyone else had this problem? (I have posted the same question to the Yongnuo FLICKR group.)

I have seen numerous reviews on the internet, and they all seem to praise the YN-468, so I believe that the flash should work, but alas it doesn't.

Welcome to the forum

If you fill out your profile we can better answer any questions that you might have.
You can do that at http://nikonites.com/profile.php?do=editprofile

Some useful links
http://support.nikonusa.com/app/ans.../nikon-product-manuals-available-for-download
Nikon | Imaging Products | Digitutor


Thanks

We have members you use that flash so hopefully one of them can answer your question.

Maybe you should read the manual that came with it....Or get a Nikon SB700 to solve the problem...

Obviously from your equipment list, you are not budget constrained. But for those of us who are, it could be that third party equipment is good enough to take recreational photographs.

From my experience I would have to agree with you that the SB-700 would be a 'better' choice, but it might not be the obvious choice when $$ are involved (~3X difference in cost between the YN vs Nikon).

Thanks for not really answering my question.

Can you post one of the bad pictures? Metering depends on the scene the camera sees. Generally white/light/bright things are underexposed, and generally dark/black/dull things are overexposed. Maybe by a full stop sometimes. But it sort of depends on what the camera is pointed at.

Are all of your flash compensation menus in a state known to you? If you see the +/- icon visible in the top LCD, better go check them all.

I don't know much about a Yongnuo flash unit, but it is just a flash, and the camera TTL system meters TTL preflash, and sets power level in the flash. Just saying, it is the camera that controls TTL. The flash should respond correctly of course, consistently.

And it is only beginners that imagine that the camera ought to always get it right. Exposure metering depends on the scene being metered. No matter how good your gear is, this is still true. TTL is often pretty close, but it also often needs a little flash compensation help. Which is one of the first things to be learned.

See How light meters work

There, at least see the roses about mid page. That is how things work. Then things go better when we realize it, and learn how to do our part to help.

I will have to take some pictures tomorrow, (I deleted the test shots).

What I can say is that when the camera is looking at the same scene (an 'egg shell' white wall) it will occasionally produce the correct exposure, and then the next shot will be over exposed. There does not seem to be any rhyme or reason. It is almost like it is a loose connection, but I am pretty sure that it is not the case.

The camera seems to communicate the zoom setting of the lens just fine, i.e. the millimeter display will change with the zoom, so I believe the digital communications are working.

The exposure is more than a 'little' off, this can be seen in both the histogram (much the data above the 90% level), and in the highlights image much of the wall and any people are saturated.

Sequential images will have total different exposure values, even if nothing is changed on the camera or flash. I don't believe it is a camera setting, as both my D5100 and a D90 had the same overexposure.

I used the flash for some family photos yesterday. It seems that in real world shooting, with the flash aimed slightly up (to diffuse the light), the results were mixed, but looked suitable. (I still need to created some test shots to upload.)

The article on how metering works is very informative. I considered much of it intuitive, except for the fact that the camera's metering (is it spot? area?) may not be what I think it is. I have to figure out how to adjust it, maybe I changed a default setting at some point. Maybe change the flash compensation.

What I am wondering is if one of the issues that I had is that my testing was done by pointing the flash directly at a scene. While, when I angle the flash the results are more consistent. Could it be that with this flash, it is the pre-flash used to measure the iTTL that is 'bad'. If that pre-flash was not consistent or repeatable, then the measurement might be off. Maybe 'aiming' the system at a specular object is not the right test. Maybe it is just that the flash compensation needs to be adjusted.

None of this really explains why the flash is not behaving consistently, and that is what bothers me. I cannot figure out why the exposure is not always the same. Sometimes it is fine, other times it is overexposed. Just the two deltas, nothing has been significantly underexposed.

Maybe the digital iTTL communication is the issue. Could it be that Yongnou did not properly reverse engineer the protocol? (Somewhere I found a reference for the protocol on the internet, it seemed simple enough.)

My intuition is that most of what could go wrong would lead to 'underexposed' rather 'overexposed' pictures, but I can see how things could go awry. I have not had much experience with flashes, so my naiveté is that the iTTL should work. Of course in my case it is not working so...

The solution may be to just use this flash in manual mode - just like the 'professionals'.

PROBLEM SOLVED!

So I did a few more test shots this evening, this time varying the way that the camera metered (This was hinted to in the 'how light meters work' article - see above.

As far as I can tell, the issue was with the 'matrix metering'.

I changed it to both 'center' or 'center weighted' and now the exposure is much better!

This change may be obvious to experienced flash users (strobists?), but to a flash newbie it is not. Maybe it is buried somewhere in an instruction manual.

Now the question is why did the SB-600 work well, but the YN-468 did not (see first post)? Could it be that the Nikon flash communicates the metering mode to the camera? Could the 'splash' pattern be more uniform with the Nikon, and so the matrix metering works better?

In any case, by changing to the center metering, the system (flash & camera) works much better. I won't say 100%, but now I feel it is worth the money (~$80 street price).

I did have a communication issue, that was fixed with a power cycle to the flash. I would suggest turning on the camera first, and then the flash.

I hope that my documenting this experience is useful to someone else. Feel free to comment.

I'm hesitant to say anything, because 1) I don't know about your case, and 2) it's hard to argue with what works, and 3) you're clearly working hard at it, giving it some thought. But a couple of facts are these:

Matrix vs Center metering is more about ambient than flash. Their results are more alike than not, usually hard to detect much difference. But the flash system (in camera) has its own metering, not Matrix and not Center, and is more like a smaller center zone. Not Spot, but not large (Spot metering is something entirely different, but also about ambient).

I like Center metering more myself, just what I'm used to, but it really is more about the ambient.

Regardless, it is the camera that does the metering, and it comes up with an exposure requirement for the flash. It simply programs the TTL power level into the flash, and all the TTL flash has to do is to comply, and flash accurately and consistently. If you are getting different results for the same picture (a few seconds apart, waiting for the Ready LED each time), that would not sound consistent. It would seem a warning sign if the flash is acting erratically, different than the SB-700.

You said bounce once, so I assume these are indoor pictures? What is your ISO level? If is Auto ISO, then indoors, the poor flash system will discover it is always working into high ISO, and competing with the room lights, and could explain overexposure. You might try ISO 100 with direct flash (maybe ISO 400 for bounce). Then in relatively dim ambient indoors, the ambient will be far down at ISO 100 (out of consideration), and it really won't matter if Matrix or Center..

I purchased a 468 II ittl, for Nikon to use with my new D3200. I could not get it to fire the pre-flash for TTL. It over-exposed almost every shot, unless I put it in manual mode and even then was generally overexposed. TTL would work once every 10 or so shots. So, I tried it on myh d5000 - and VOILA - ttl heaven. I then ordered a pair of 622-n wireless TTL triggers. I put the transceiver on the D3200 and the receiver on the off-camera 468 ii... and VOILA.. off-camera TTL. I then put the 468ii on top of the transceiver (on-camera) and VOILA... instant TTL. Then I tried just the 468ii on-camera (no 622-n trigger transceiver) and DISAPPOINTMENT - NO -TTL. So the foot on the 468ii must be off by some small fraction as to not make the right contact with the foot on the camera. Again, the flash worked without a hitch using the 622-n wireless with the d3200 (I even exchanged the d3200 for another, and the flash still did not produce TTL results), and with the d5000. I returned the unit in hopes that the replacement unit will be fine. I even ordered another one with another 622-n transceiver. The two flashes with my umbrella soft boxes will be my quick travel light kit. I can't wait for the 622 with digital read out to come out to make managing the output easier.

I had similar issues with my D5100, but I managed to mitigate much of the issues by changing the metering mode. Have you tried this method with the D3200?

There seems to be a bunch of useful tidbits in a post of the following thread (I have not yet fully digested the differences between the TTL modes): http://nikonites.com/flashes/17535-yongnuo-flash-owners-request.html

I suspect that the replacement flash will have the same behavior (as my replacement did).

Here are some of the basic the settings I have for use with my Yongnuo 468-II. I've used this flash on both my D5100 and my current D7100 with great success. Maybe this will help some of you who are having issues with exposure.

The flash unit itself is in TTL mode.
Camera is in "A"perture Priority, Matrix metering.

In camera settings (Flash/Bracketing menu) are as follows:
Synch Speed: 1/250 (This will vary with your camera but 1/250 is the default for the D7100, I suggest you determine what the default setting is for your camera and use it here.)
Flash Shutter: 1/60
Flash Control: TTL
Exposure Comp: Entire Frame
Modeling Flash: ON

......