Reading mode: Light Dark

AlienBees Exposure Consistency: Bad News and Good

Go to last post
Stan Prevost said:
Thanks for all your work in alerting us to this issue with ABs. I
don't think it seriously degrades their value to most of us, and I
plan to order mine today.
Click to expand...
I might point out that I personally am not having a problem--I very rarely go below 1/4 power and I always use a remote option. I'd actually never seen the problem manifest in my own work.

I investigated based on the complaints of others.

I did not run any tests to check for potentiometer contamination...I should have though of that, having built simple radios since my days of glomming Allied Electronics catalogs in the 60s.

--
RDKirk
'TANSTAAFL: The only unbreakable rule in photography.'
 
Sorry, I didn't mean to imply that you had done anything that might be characterized as sensationalizing or maligning the AB units. I think the kind of info that you and others brought to our attention is very useful and was fairly presented. It won't matter to most, but will to some. Those not using a remote, and using higher-power units throttled back might experience unacceptable variations in exposure. One user was having trouble getting his lighting ratios to work out consistently, and found that the flash output variation was at least part of the problem. That is useful to know for the rest of us, so it can be worked around.

I doubt the issue is limited to ABs, but that is all that has been reported so far to my knowledge.

As I said, I am ordering mine anyway, in full knowledge of the reported behavior. I am very glad to know it now rather than struggling with unknown and unexpected behavior later. I will check my lights perodically, and if one exhibits unacceptable behavior, I will send it back to AB for repair, as they have said that it is not the norm and that they will repair any unit showing excessive variation. I will not be pleased with one stop variation at 1/4 power and will return such a unit.

Stan
 
Dude, first WOW! Thanks, this is very helpful, but you do need to get a life! ;-) THANKS!
Carl Snider said:
Interesting test, kind of got me curious. Since you make no mention
of how much of a variance we are talking about, I wasn't sure if it
should be of concern. So I decided to try and set up a test of my
own. Parameters follow:

Meter - Sekonic L-508, Tripod mounted, Fixed Position, Lumisphere
5' from ground, Lumisphere lowered. Meter set 1/125, ISO200.

Lights - (4)AB1600, Standard Reflector, Fixed position, Bulb 8'
from Lumisphere same height as lumisphere, Cable triggered by L-508.

Lights were powered on and allowed to sit for 1 minute then dumped.
After change to each power setting light was dumped twice before
first reading. Lights were fired for 15 readings (too lazy to do a
bigger sample).

Unit #1
Full power - f32 9/10 - No variance
1/2 power - f22 9/10 - No variance
1/4 power - f22 0/10 - 1 shot at f22 1/10
1/8 power - f16 0/10 - 2 shots at f16 1/10
1/16 power - f8 9/10 - No variance
1/32 power - f5.6 8/10 - 3 shots at f5.6 7/10

Unit #2
Full power - f45 0/10 - No variance
1/2 power - f32 1/10 - No variance
1/4 power - f22 2/10 - No variance
1/8 power - f16 2/10 - 2 shots at f16 3/10
1/16 power - f11 0/10 - 1 shot at f8 9/10
1/32 power - f5.6 7/10 - 3 shots at f5.6 4/10, 4 shots at f5.6 5/10

Unit #3
Full power - f45 1/10 - No variance
1/2 power - f32 1/10 - No variance
1/4 power - f22 2/10 - No variance
1/8 power - f16 1/10 - No variance
1/16 power - f8 9/10 - No variance
1/32 power - f5.6 8/10 - No variance

Unit #4
Full power - f45 0/10 - No variance
1/2 power - f32 0/10 - No variance
1/4 power - f22 1/10 - No variance
1/8 power - f16 1/10 - No variance
1/16 power - f11 1/10 - No variance
1/32 power - f5.6 8/10 - 5 shots at f5.6 7/10, 2 shots at f5.6
6/10, 6 shots at f5.6 5/10

Looks like Unit #3 will be used as main, but I don't really think
most of the variances are large enough to cause concern, with
exception of unit #4 at 1/32 power. Although since the meter itself
has a repeat accuracy of + - .1 EV, who's to say. Since the meter
can only display changes at the .1 level, variances could be as
small or large as the resolution algorithm permits (i.e. if reads
to thousandth .001 or up to .199). I'm not sure if the variances
documented here should be of any concern, I will have to shoot 15
shots at each power of poorest performing unit, subject Gretag
Macbeth colorchecker, and examine results. Then again, since I
don't really see any variance in the images I am getting, maybe
I'll just not worry about it. YMMV!

Hope this data is of some use, no real conclusions, just information.

Be well,
Carl
Click to expand...
--
http://www.pbase.com/kluken
 
ROFL - Curiousity killed the cat, and as it turns out, 90 minutes of my morning today!
kluken said:
Carl Snider said:
Interesting test, kind of got me curious. Since you make no mention
of how much of a variance we are talking about, I wasn't sure if it
should be of concern. So I decided to try and set up a test of my
own. Parameters follow:

Meter - Sekonic L-508, Tripod mounted, Fixed Position, Lumisphere
5' from ground, Lumisphere lowered. Meter set 1/125, ISO200.

Lights - (4)AB1600, Standard Reflector, Fixed position, Bulb 8'
from Lumisphere same height as lumisphere, Cable triggered by L-508.

Lights were powered on and allowed to sit for 1 minute then dumped.
After change to each power setting light was dumped twice before
first reading. Lights were fired for 15 readings (too lazy to do a
bigger sample).

Unit #1
Full power - f32 9/10 - No variance
1/2 power - f22 9/10 - No variance
1/4 power - f22 0/10 - 1 shot at f22 1/10
1/8 power - f16 0/10 - 2 shots at f16 1/10
1/16 power - f8 9/10 - No variance
1/32 power - f5.6 8/10 - 3 shots at f5.6 7/10

Unit #2
Full power - f45 0/10 - No variance
1/2 power - f32 1/10 - No variance
1/4 power - f22 2/10 - No variance
1/8 power - f16 2/10 - 2 shots at f16 3/10
1/16 power - f11 0/10 - 1 shot at f8 9/10
1/32 power - f5.6 7/10 - 3 shots at f5.6 4/10, 4 shots at f5.6 5/10

Unit #3
Full power - f45 1/10 - No variance
1/2 power - f32 1/10 - No variance
1/4 power - f22 2/10 - No variance
1/8 power - f16 1/10 - No variance
1/16 power - f8 9/10 - No variance
1/32 power - f5.6 8/10 - No variance

Unit #4
Full power - f45 0/10 - No variance
1/2 power - f32 0/10 - No variance
1/4 power - f22 1/10 - No variance
1/8 power - f16 1/10 - No variance
1/16 power - f11 1/10 - No variance
1/32 power - f5.6 8/10 - 5 shots at f5.6 7/10, 2 shots at f5.6
6/10, 6 shots at f5.6 5/10

Looks like Unit #3 will be used as main, but I don't really think
most of the variances are large enough to cause concern, with
exception of unit #4 at 1/32 power. Although since the meter itself
has a repeat accuracy of + - .1 EV, who's to say. Since the meter
can only display changes at the .1 level, variances could be as
small or large as the resolution algorithm permits (i.e. if reads
to thousandth .001 or up to .199). I'm not sure if the variances
documented here should be of any concern, I will have to shoot 15
shots at each power of poorest performing unit, subject Gretag
Macbeth colorchecker, and examine results. Then again, since I
don't really see any variance in the images I am getting, maybe
I'll just not worry about it. YMMV!

Hope this data is of some use, no real conclusions, just information.

Be well,
Carl
Click to expand...
--
http://www.pbase.com/kluken
Click to expand...
 
Yeah I know the feeliing as one measurbator to another. It actually made me feel a bit easier about ordering my first set of Bees, I was ocnsidering the WL upgrade, but it woudl run me another $400, so I figured I would rather get the wireless setup instead!
Carl Snider said:
kluken said:
Carl Snider said:
Interesting test, kind of got me curious. Since you make no mention
of how much of a variance we are talking about, I wasn't sure if it
should be of concern. So I decided to try and set up a test of my
own. Parameters follow:

Meter - Sekonic L-508, Tripod mounted, Fixed Position, Lumisphere
5' from ground, Lumisphere lowered. Meter set 1/125, ISO200.

Lights - (4)AB1600, Standard Reflector, Fixed position, Bulb 8'
from Lumisphere same height as lumisphere, Cable triggered by L-508.

Lights were powered on and allowed to sit for 1 minute then dumped.
After change to each power setting light was dumped twice before
first reading. Lights were fired for 15 readings (too lazy to do a
bigger sample).

Unit #1
Full power - f32 9/10 - No variance
1/2 power - f22 9/10 - No variance
1/4 power - f22 0/10 - 1 shot at f22 1/10
1/8 power - f16 0/10 - 2 shots at f16 1/10
1/16 power - f8 9/10 - No variance
1/32 power - f5.6 8/10 - 3 shots at f5.6 7/10

Unit #2
Full power - f45 0/10 - No variance
1/2 power - f32 1/10 - No variance
1/4 power - f22 2/10 - No variance
1/8 power - f16 2/10 - 2 shots at f16 3/10
1/16 power - f11 0/10 - 1 shot at f8 9/10
1/32 power - f5.6 7/10 - 3 shots at f5.6 4/10, 4 shots at f5.6 5/10

Unit #3
Full power - f45 1/10 - No variance
1/2 power - f32 1/10 - No variance
1/4 power - f22 2/10 - No variance
1/8 power - f16 1/10 - No variance
1/16 power - f8 9/10 - No variance
1/32 power - f5.6 8/10 - No variance

Unit #4
Full power - f45 0/10 - No variance
1/2 power - f32 0/10 - No variance
1/4 power - f22 1/10 - No variance
1/8 power - f16 1/10 - No variance
1/16 power - f11 1/10 - No variance
1/32 power - f5.6 8/10 - 5 shots at f5.6 7/10, 2 shots at f5.6
6/10, 6 shots at f5.6 5/10

Looks like Unit #3 will be used as main, but I don't really think
most of the variances are large enough to cause concern, with
exception of unit #4 at 1/32 power. Although since the meter itself
has a repeat accuracy of + - .1 EV, who's to say. Since the meter
can only display changes at the .1 level, variances could be as
small or large as the resolution algorithm permits (i.e. if reads
to thousandth .001 or up to .199). I'm not sure if the variances
documented here should be of any concern, I will have to shoot 15
shots at each power of poorest performing unit, subject Gretag
Macbeth colorchecker, and examine results. Then again, since I
don't really see any variance in the images I am getting, maybe
I'll just not worry about it. YMMV!

Hope this data is of some use, no real conclusions, just information.

Be well,
Carl
Click to expand...
--
http://www.pbase.com/kluken
Click to expand...
Click to expand...
--
http://www.pbase.com/kluken
 
"Popular Photography" has a review of the AB400 this month.

Besides generally praising it, they reported that the color temperature drops from 6100 degrees to 5700 degrees over the full power range down to 1/32 power.

This 400 degree drop in their example compares to a 350-370 degree measured by a poster in the Rob Galbraith forum for Profoto, White Lightning, and Elinchrome units over a range just down to 1/4 power for each of those units.

This is the only measured report I've seen of color temperature variation for an Alienbees unit, and it compares favorably with the report on more expensive units.

Also, the only complaints "Popular Photography" had about the AB400 was that the built-in flash sensor seemed less sensitive than they'd like because it's recessed in the rear panel, and the flash doesn't have an audible recylce beep.
--
RDKirk
'TANSTAAFL: The only unbreakable rule in photography.'
 
RDKirk said:
Based on complaints in the forum from several people, I tested five
samples of Alienbees lights (three B800s, two B400s) for
shot-to-shot exposure consistency at various power levels. I
tested with a Sekonic L-358.
Click to expand...
[snip]

-----------------------------------------------------------------------------

When I spoke with AB last year about the color temp variation I was told that it was around 200K end-to-end. Someone (can't pin down who) recently mentioned a 'Kodak' standard talking about an acceptable range of
TMc
 
Test parameter wise. I have four new AB 800s. When I get my studio finished I'm going to test them a bit.

thanks

don
RDKirk said:
dwn said:
Now I have some questions: How careful were you with your "test
parameters"?
Did you mount the flash meter in a fixed position,
Click to expand...
Yes.
RDKirk said:
was there any
other source of light,
Click to expand...
Low level ambient. The Sekonic L-358 measures and displays the
percentage of flash content to the exposure (versus ambient light),
and the flash exposure content was always 100 percent. At any
rate, the ambient light was constant.
RDKirk said:
did you stand in the same position each
time?
Click to expand...
The position of the meter was fixed.
RDKirk said:
Reason I asked is IF it were the inaccuracy of the sliders,
for instance, it would still be consistent if you did not move the
slider,
Click to expand...
I don't think it was simply the position of the sliders. Alienbees
lights operate on a different principle from most other electronic
flash on the market, and although it's a different principle, it is
apparently not by any means an unknown or disreputable principle.
The science is out of my league (the only information I have on it
is here: http://davidweikel.com/E20_Page/alienbees/bees.shtml ),
but I suspect that the inconsistency at low power levels as well as
the RETURN to consistency when the power levels are electronically
controlled is explainable as a matter of how Alienbees particularly
work.
RDKirk said:
and I'm not sure a flash tube would be that inconsistant
and still be used by Paul. Inconsistency in testing would be more
likely to show up at the lower end of the scale.
Click to expand...
Which is exactly what I reported: The inconsistency shows up at
the lower end of the scale.
--
RDKirk
'TANSTAAFL: The only unbreakable rule in photography.'
Click to expand...
--
dwn
 
Conversations with others revealed a testing flaw. I did NOT ensure the rear slider potentiometers were clean of contaminants by "swiping" them a few times before the test.

I almost never use the rear sliders--I bought my lights with the remote control. That means it's highly probable the slider surfaces were covered with a contaminate film (from the factory or even a thin film of oxidation). That would be a reason why I had inconsistent results with the rear sliders and not with the remote controls.

I'll try to run the tests again this weekend.

--
RDKirk
'TANSTAAFL: The only unbreakable rule in photography.'
 
RDKirk said:
I'll try to run the tests again this weekend.
Click to expand...
My new ABs should arrive Monday. As part of getting to know them, I will also be testing them throughout their range. Unfortunately, I don't yet have an exposure meter with very good resolution. My little Digiflash only reads out to 1/3 stop. But with careful use, it should show up any serious inconsistency.

Maybe I can borrow a better meter. Anyway, I will try to post results on factory-fresh units.

I don't expect anything of concern. But, that's why we test, to verify. :-)

Stan
 
Consistent at the very lowest level unless I started flashing @ less than 1 second. Good enough for me.
 
You must log in or register to reply here.

Similar threads

I
Change my Mind (or not) switching OM-1II TO G9II
2
Replies
37
Views
67K
IDontSpeakMonkey
I
lokatz
Noise reduction: Is it worth it? How much improvement?
2 3
Replies
44
Views
40K
Breakfastographer
B
Satyaa
ML journey with m43... part 3 exposure, framing and editing
Replies
0
Views
6K
Satyaa
Satyaa
Satyaa
ML journey with m43... part 1 the cameras
Replies
1
Views
8K
jalywol
jalywol
joger
LAB 35 f/1.2 on A7R IV ~ you can literally feel the smooooothness - tested against GM 35 & Loxia 35
2 3
Replies
59
Views
33K
joger
joger

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top