Magenta colour cast when using strobes

Seajay

Suspended / Banned
Messages
1,493
Name
Cathy
Edit My Images
Yes
Morning everyone! I need a little advice please.
I am using my Lencarta strobe with a large 7ft Westcott brolley with diffuser to take some newborn shots on the days I can't quite get enough natural light.I also have a white reflector to bounce back light from the other side.
Each time I get a magenta colour cast even though there is no stray light coming in or anything in the room with colour. My settings only 60D are shutter 125- F2.8- Iso 100. I do have white balance on auto but always take a grey card shot before I start and a few times through the session. Anyone any idea on what i am doing wrong?
Cathy
 
Switch white balance to flash.
Your white balance should be set to the dominant light source. When you use Auto, the camera is registering the light source just before the shutter opens, which means it sees the ambient light in the room, because the flash only goes off once the shutter is open.
 
Yes, set WB to flash. Or do a custom WB, or use your grey card reference in post-processing.

I don't know how your auto-WB works when using flash. If you are using an auto-TTL trigger, the camera will probably sense that and switch to flash WB, but if it's a basic manual trigger, the camera will probably apply auto-WB from the ambient. And that could well change during the shoot.

It's not uncommon for flash to have a slight tint. It's just as likely that the magenta is coming from the Westcott. Or there could be a number of other causes - the environment is unlikely to be neutral, ambient light, running a longer shutter speed.
 
Thank you Michael and Hoppy. I tried my white balance on flash before and it was exactly the same.So that is why I set it to auto in the hope the camera may work it out lol! I use my triggers that came with my lighting kit to work my lights. I might try and take the diffuser off and see what happens. Its the only other thing I can try.
 
Thank you Michael and Hoppy. I tried my white balance on flash before and it was exactly the same.So that is why I set it to auto in the hope the camera may work it out lol! I use my triggers that came with my lighting kit to work my lights. I might try and take the diffuser off and see what happens. Its the only other thing I can try.
You may find that the diffuser isn't white, this is common. But the cause of the problem is as already explained by Michael and Richard, and all that you need to do is to either set the white balance correctly or to set an auto white balance.
Depending on your camera, you may be able to set the Kelvin value of your flash (5500) or you may just be able to set it to "flash" which isn't normally right but which will be in the ball park.
If you set a custom white balance (an easy job but the procedure varies with different cameras) then it will be right, and will also allow for environmental factors, such as a red carpet, blue walls or whatever, all of which reflect unwanted colour into the scene.
Or, assuming that you are shooting in raw, you can simply include a colour target into a test shot and use this to set the white balance via your computer.

But, whichever means you use, the problem is due to the white balance setting on your camera, which doesn't know that you're using flash.
 
Thank you Michael and Hoppy. I tried my white balance on flash before and it was exactly the same.So that is why I set it to auto in the hope the camera may work it out lol! I use my triggers that came with my lighting kit to work my lights. I might try and take the diffuser off and see what happens. Its the only other thing I can try.

Another variable I forgot to mention is the flash power setting. Conventional studio flash gets warmer as the power is reduced, IGBT flash (mainly speedlites) can get a bit blue at lowest settings. Not big changes, not usually, but certainly noticeable if the power setting is changed significantly.

I've yet to find two softboxes from different manufacturers that are exactly the same colour. As Garry says, the usual culprit is the front diffuser material - you can often see it if you put them side by side. All minor problems though, easily fixed with a colour card reference in post.
 
Thank you! All this information is appreciated. I have never used Kelvin do this will be my next setting for my canon60D . It drives me crazy but it is easily fixed in pbotoshop with the magents slider for now.
 
When you use Auto, the camera is registering the light source just before the shutter opens, which means it sees the ambient light in the room, because the flash only goes off once the shutter is open.

Really? I thought most auto WB worked off the captured image from the sensor by generating some statistics from it and comparing the results to those from a database of reference images? That's certainly what 'phones do.
 
Really? I thought most auto WB worked off the captured image from the sensor by generating some statistics from it and comparing the results to those from a database of reference images? That's certainly what 'phones do.

no, my research is the same as Michaels, only time something different happens is when the camera can detect the flash (dedicated flashes) it will normally set WB to flash

Mike
 
no, my research is the same as Michaels, only time something different happens is when the camera can detect the flash (dedicated flashes) it will normally set WB to flash

Mike

Interesting. I can't speak for all phone types but a large number of them definitely do it the other way. And I'm guessing Lightroom's Auto mode does it that way too.
 
Really? I thought most auto WB worked off the captured image from the sensor by generating some statistics from it and comparing the results to those from a database of reference images? That's certainly what 'phones do.

Good question. A quick google draws a blank - lots of talk about 'blackbody radiation' but nothing about the actual mechanism that cameras use, or when the setting is applied.

I'd always assumed that AWB was calculated by the metering sensors, which could mean in the pentaprism for DSLRs or off the sensor in live view, or either/both. And is the setting decided pre- or post-capture? Some cameras have separate white balance sensors, but I think that's very early digital and history now.
 
Just a quick pic taken with a doll and magenta slider full up to let you see. This background was white but it has the same effect on all backgrounds I use.




pinktone.jpg
 
I'd always assumed that AWB was calculated by the metering sensors,
I know modern Nikons use the RGB metering sensor, also uses the scene recognition database...

From Nikon's site.
"Nikon's Scene Recognition System (SRS) recognizes the position, color, tones and characteristics of a subject or overall scene prior to capture; then, using information from the 3D Color Matrix Meter II (420-pixel RGB sensor or 1,005-pixel RGB sensor)—depending upon the model of D-SLR—compares that information to the camera's built-in image database to achieve accurate autofocus, auto exposure and auto white balance control."
 
Not even 6 months yet but it only happens when using strobes. Canon 6D.
If you are certain it is *only* affecting strobe shots then it is almost certainly a lighting temp issue. A custom WB should fix it. Or a manual WB setting will at least allow you to adjust one image and apply that to all of the other images.

I would avoid auto WB for studio shots... well, I don't but I should... and so should you.
 
Last edited:
I know modern Nikons use the RGB metering sensor, also uses the scene recognition database...

From Nikon's site.
"Nikon's Scene Recognition System (SRS) recognizes the position, color, tones and characteristics of a subject or overall scene prior to capture; then, using information from the 3D Color Matrix Meter II (420-pixel RGB sensor or 1,005-pixel RGB sensor)—depending upon the model of D-SLR—compares that information to the camera's built-in image database to achieve accurate autofocus, auto exposure and auto white balance control."

Thanks Steven, interesting to know :thumbs:

On Cathy's pink baby, I'm not sure what to make of that. The magenta slider is on max, so of course it's pink...?

FWIW, I don't use AWB for anything. Not because it's unreliable (which it often is) but because it can be inconsistent and change during a session just like auto-exposure settings can change, which is totally unhelpful. So I use the pre-sets (daylight, shade, flash etc) that will usually get things pretty close as a working reference, and cannot vary. But as I'm shooting Raw anyway it doesn't really make much difference.
 
I will take some shots next time using the Kelvin setting and let you know how it goes.For now I would like to thank you all and here's to magenta free pics next time.
 
Did I miss it somewhere, or did someone already ask if you shoot RAW. If I was having your problem I would fix it in Capture One v.9.1 it has new colour editing features and a very simple kelvin adjustment.
 
Thanks Steven, interesting to know (y)

On Cathy's pink baby, I'm not sure what to make of that. The magenta slider is on max, so of course it's pink...?

FWIW, I don't use AWB for anything. Not because it's unreliable (which it often is) but because it can be inconsistent and change during a session just like auto-exposure settings can change, which is totally unhelpful. So I use the pre-sets (daylight, shade, flash etc) that will usually get things pretty close as a working reference, and cannot vary. But as I'm shooting Raw anyway it doesn't really make much difference.
The amazing thing to me is that it does all of that before the image is taken... in that split second focus is achieved. Heck, the 3D AF mode even uses the color meter for subject tracking during autofocus.

I use a lot of automation when photographing wildlife and situations that are unpredictable/change a lot.
My biggest problem is I often forget to turn that stuff off when things are much more controlled, until after it's caused me a problem. Who needs to use ISO 6400 in a studio with strobes?
 
Last edited:
You might be getting a bit of ambient light creeping in at f.28 turn up the power on the flash and shoot at f8 see what that looks like.
Are you able to try a different camera and a different flash with the same setup?
 
I will take some shots next time using the Kelvin setting and let you know how it goes.For now I would like to thank you all and here's to magenta free pics next time.

The only problem is colour with the Westcott? Then to isolate and identify the extent of the problem, shoot a comparison test scene - anything* really, including your grey card. Fit the flash with a known good modifier, positioned as close as you can reasonably get (reduces the influence of spill). Include your grey card. Flash white balance. Raw. Take a test shot with the flash off, to check there's no ambient showing through. If the image is not very dark/black, increase flash power and push shutter speed up to max x-sync speed (1/250sec on a 60D**). Then fit the Westcott and shoot with and without the diffuser panel/s.

Do not change anything except the modifier. Adjust exposure with aperture and/or ISO for this test, not flash power. In processing, use the WB dropper on the grey card, and use that value for all pics.

In a strict test like this, you'd expect to see some very slight differences, but nothing too serious and nothing that can't be corrected in post-processing. If there is a significant difference, that may be a problem if you have more than one light source (much harder to correct). Consistency is often more important, and more noticeable, than accuracy. A cure might be a tip from Garry - give the Wescott a very light spray inside with car aerosol paint. Lots of subtle colours available. It's not something I've done, but sounds quite easy and should be a good fix.

*Maybe include that white sheet. There's perhaps an outside chance there is some artificial whitener in that, or washing powder, that could be making a difference. It's at least worth ruling out.
**You may get some slight shading at the bottom of the frame at 1/250sec, or even a black band creeping in. If so, use 1/200sec.
 
As above - discussion about how cameras do their calculations are interesting, but irrelevant to your problem, and I'd be very surprised too if the problem was related to the camera sensor - the problem is clearly that you set auto white balance.
Richard's advice above is spot on, even though the problem here is clearly due to the auto white balance - whenever an unexpected result happens, do a test or a series of tests to identify the cause of the problem. Most people don't seem to realise that modifiers can make dramatic changes to colour temperature, especially cheap ones, and this can also apply to expensive ones that have changed colour due to age.
Doing test shots using just the flash head fitted with the standard reflector will always produce a benchmark result, unless the flash head itself is a rubbish one - but the ones you have have extremely consistent colour temperature, shot to shot, and although the shot to shot variation is sometimes measurable with a colour temperature meter, it is never visible to the naked eye.
 
Back
Top