580EX beamwidth anomaly - who goes first?

Doug Kerr

Forum Pro
Messages
20,899
Reaction score
13
Location
Alamogordo, NM, US
We seem to have confirmed that early production Speedlite 580EX units exhibit differentn behavior with respect to beamwidth control than later production units - the behavior of the earlier ones ("Behavior X", in the terms of my reserach) being, to me, clearly anomalous, and I suspect the result of a faux pas in firmware development.

For those not familiar with this issue, you can read about it here:

http://forums.dpreview.com/forums/read.asp?forum=1019&message=17303868

So who will be the first of you with a "Behavior X" 580EX to go to Canon and say, "My 580EX doesn't work right - will you fix it?"

I can't - I don't even own one!

Hoppy - how about you? You probably have two of each flavor! You could clearly describe to Canon how they work differently.

If anybody wants to, I can suggest some language to be used in the approach.

Best regards,

Doug

Visit The Pumpkin, a library of my technical articles on photography, optics, and other topics:

http://doug.kerr.home.att.net/pumpkin

'Make everything as simple as possible, but no simpler.'
 
I think they actually quit extending at 38.4mm (24mm X 1.6)?

When I first got my 20d, 17-85, & 580ex a year ago I called canon to complain about the flash not extending to the full wide position. I had to educate thier Techs on many of the aspects of thier equipment. It was of the most frustrating experienses I've had with service and almost enough to make me switch brands (but I didn't figure it would get any better with nikon either.)

When I sent the 20d/580ex into canon repair they kept it for 14 days then sent a B&H spec sheet for the 580ex with "24mm" flash coverage highlighted in yellow and the qoute "It don't work with 17mm". When I informed them that 17mm in 1.6x is about 28mm in FF they said "OH, sorry, send it back in."

Custemer Service told me I had unrealistic expectations. I guess working as advertised is unrealistic?

Since I've change custom function #11 to FF only I've been happy. Now that I have both FF and APS I'll leave all my flashes set to FF coverage.

P.S. When I sent the 20d into repair for the cover slip over the senor I recieved the repaired camera BEFORE I recieved the letter of adknowlegdement that they would repair it!

Mike
 
Hi, Mike,

Yes, I'm afraid this is going to be problematical.

I think the key to pressing the case is that they changed the design.
I think they actually quit extending at 38.4mm (24mm X 1.6)?
I'm not sure exactly what you mean. Tell me more.

I would think that the "Behavior X" units do not adopt any "wider" beamwidth (as lens focal length decreases) after they go to the 35 mm head position, which they do at a reported lens focal length of 35 mm. But of course I have never done any detailed testing myself (not having a 580EX!).
Custemer Service told me I had unrealistic expectations. I guess
working as advertised is unrealistic?
Well, the problem there is just what does "as advertised" mean! (We know what the machine is "supposed to do", but that's a different thing altogether!

For example, there are those who take the manual language at face value and say that the machine is not expected to provide automatic beamwidth control for lens focal lengths less than 24 mm (at which point a "Behavior X" unit will have set a head positon of "35 mm", which theoretically will well serve the field of view of a 35-mm lens on a 20D).
Since I've change custom function #11 to FF only I've been happy.
Now that I have both FF and APS I'll leave all my flashes set to FF
coverage.
But of course the intent is that they will automatically accommodate the sensor size. (That is, one should not have to change the CF when changing cameras.)
P.S. When I sent the 20d into repair for the cover slip over the
senor I recieved the repaired camera BEFORE I recieved the letter
of adknowlegdement that they would repair it!
Ironic, wot?

Best regards,

Doug
 
So who will be the first of you with a "Behavior X" 580EX to go to
Canon and say, "My 580EX doesn't work right - will you fix it?"
I fought with them for two weeks and over 30 emails about this. They refused to acknowledge there was a problem. Since they changed the design, I doubt they could make that lame claim again but I'm not sure I care since I rarely use direct flash anyway.

--
Lee Jay
(see profile for equipment)
 
For anyone planning to approach Canon about correction of "Behavior X" of his 580EX, here is some language you might wish to use:

***********

The automatic head zoom function of my Speedlite 580EX flash unit does not work correctly wren the unit is used on an EOS 20D body. No matter how small the lens focal length (even down to 17 mm), the flash never sets its head to a position any wider than 35 mm. That head position, on a 20D, is not appropriate for a focal length less than about 22 mm.

I determined the head position by visual inspection. I learned what the various positions looked like by setting the zoom manually with the flash off the camera.

Incidentally, I do have flash C/Fn-11 set to 0 so that sensor size compensation should be in effect.

I borrowed another 580EX from a friend to see how it handled this. With it (also on my 20D), for a lens focal length of 20 mm, the flash set its head to the 28 mm position; with a lens focal length of 17 mm, the flash set its head to the 24 mm position. That would seem to be the proper behavior on a 20D.

Clearly there is a flaw in my 580EX in how it determines the proper head positiion when used on a 20D. How should I proceed to have you correct that?

***********

Best regards,

Doug

Visit The Pumpkin, a library of my technical articles on photography, optics, and other topics:

http://doug.kerr.home.att.net/pumpkin

'Make everything as simple as possible, but no simpler.'
 
Cheers Doug - I seem to have one of the early ones. If I can find my receipt, I might give this a go.

Alan
 
Doug,

Do you know, or anyone else for that matter, at which production number this anomaly occurs? I just picked one up, but have yet to test it. It might help narrow down the field if we could get a prod number, or serial number for this. Thanks.
Tim
--
It is all about the moment…
once passed…never retrieved…
once captured…never forgotten.

This IS the essence of photography
 
Hi, Tim,
Doug,
Do you know, or anyone else for that matter, at which production
number this anomaly occurs? I just picked one up, but have yet to
test it. It might help narrow down the field if we could get a
prod number, or serial number for this. Thanks.
I am in the process of receiving reports from a survey of this. There have been about 16 reports accepted.

So far, the highest number reported for Behavior X is 1689xx, and the lowest for Behavior Y is 2330xx.

It could well be that the number was bumped to 200000 when the design change occurred! The change seems to have showed up in purchases as of about May, 2005.

Best regards,

Doug

Visit The Pumpkin, a library of my technical articles on photography, optics, and other topics:

http://doug.kerr.home.att.net/pumpkin

'Make everything as simple as possible, but no simpler.'
 
Thanks for the reply. Mine has a number starting with 317XXX, and I just got it about 3 weeks ago, and so far, so good. I do not see the same behavior. I'll keep testing, and report back to the group if I see anything. Thanks for all this information, and keep up the great work.
Tim
--
It is all about the moment…
once passed…never retrieved…
once captured…never forgotten.

This IS the essence of photography
 
It could well be that the number was bumped to 200000 when the
design change occurred! The change seems to have showed up in
purchases as of about May, 2005.
If that was indeed the case, I think Canon owes us a system wide recall. See if you can talk Chuck Westfall into about doing this. ;)

--
Ray Chen
http://www.arrayphoto.com
 
Mine says serial number is 025280 -= a very early one I assume
 
I contacted them last November. At the time they would not acknowledge any problem, and stated that "Canon has not recently issued any repair notices or replacement programs for this flash."

John
So who will be the first of you with a "Behavior X" 580EX to go to
Canon and say, "My 580EX doesn't work right - will you fix it?"
I fought with them for two weeks and over 30 emails about this.
They refused to acknowledge there was a problem. Since they
changed the design, I doubt they could make that lame claim again
but I'm not sure I care since I rarely use direct flash anyway.

--
Lee Jay
(see profile for equipment)
 
Hi, Ray,
See if you can talk Chuck Westfall into about doing this.
After some initial dialog, during a time when I suspect Chuck did not know that there was a bona fide anomaly afoot here, he has not chosen to reply to any of several of my further inquiries on the matter.

Best regards,

Doug
 
Doug,

My first 20d would not work even with function 11 set. When I finally returned the camera and got a new one everything worked fine as a FF flash..

When I said that it stopped setting the flash a acoverage of 38.4 mm I was technically incorrect. When the lens reaches 38.4 mm I believe it equates this to 24mm FF, and confounds this with the APS sensor (setting it to the next widest frame coverage of 35mm...), and this is a wide as it goes.

I still like the flash well enough that I bought 2 more this month from B&H...So I am hearing you say that the problem may be fixed with the 20d? I would never have thought to reevaluate this. I just put batteries in them and switched function 11.
 
Hi, Mike,
Doug,

My first 20d would not work even with function 11 set. When I
finally returned the camera and got a new one everything worked
fine as a FF flash..

When I said that it stopped setting the flash a acoverage of 38.4
mm I was technically incorrect. When the lens reaches 38.4 mm I
believe it equates this to 24mm FF, and confounds this with the APS
sensor (setting it to the next widest frame coverage of 35mm...),
and this is a wide as it goes.
Based on my understadning (and of course there could be surprises still to be unearthed!):

With the lens at a focal length of 38.4 mm, a Behavior X flash would display the lens focal length as 35 mm ‡, would then multiply that (35) by 1.6 to get 56 mm (the "full-frame 35-mm equivalent" of of 35 mm, which has no signficance), and would then set the head to "50 mm" (the next lower available head position).

A behavior Y flash , with the lens at 38.5 mm, would display the focal length as 35 mm‡, would multiply 38.4 by 1.6 to get 61.4 mm (the "full-frame 35-mm equivalent" of the lens focal length), and would set the head to "50 mm" (the next lower available head position).

‡ With either behavior, the flash takes the reported focal length, rounds it "down" to the next lower number in the series 24, 28, 35, 50, 70, 80, 105 (to 24 if the reported focal length is lower than that.) That is of course the list of head positions, but those numbers have no signficance with respect to lens focal length in a 20D.

Can you tell me a little bit more about what you actually observe when the lens focal length is 38.4. I am still not clear what unique thing happens, or what you see, when the lens is at 38.4 mm. You haven't said directly.

And of course with the lens at a focal length of 38.5 mm, the full-frame 35-mm equivelent of that would be 61.6 mm, not 24 mm, as you suggested above. (Mutliplication, division, it's all so confusing!)

Best regards,

Doug
 
Hi Doug,

Thanks for all your hard work . One question where is the number on the 580 flash? My 580 is about a year old and don't know where to find the number stating the date.
Ken Schmidt
 

Keyboard shortcuts

Back
Top