Magenta/purple cast especially with hotshoe flashes

(I'm not sure if this is right area, but I think that this has more to do with lightning than color slides generally)

This is something that I have noticed first time perhaps around year 2003 and suffered since. The E-6 color slide films tend to have magenta color cast (varies from light to heavy) when the slide is exposed using hotshoe mounted flash. It's usually stronger with direct flash than with bounced flash.

This has happen since with other Nikons I have: FE2, F601, F100, ... And with different flashes (SB-25, SB-26, SB-28, Sunpak 120J, Metz 45 CL-4,...). With manual and TTL.

I haven't pay too much attention for this, as I have mostly used studio flashes and/or B/W films or colour negative films.

Until I recently begun to take more colour slides from normal family life which requires quite foten camera mounted hotshoe flash, even direct flash... Then I took this serious. I need better colors for my 'memory photos'.


My thoughts so far:

• It could be developing error. But it's visible in the slides regardless of lab - and even in those that I develop myself
• It could be the color of the flash. But why it occurs with all flashes (and when same flashes are used with digital, the result is neutral)
• It's mostly visible when wall's (and ceiling) are white, but that probably just because it's easier to spot.
• Can it be simply feature of the films - but if it is, why there's no CC filtration specified for flashes in the film's datasheets? At least Sensia, Provia, Astia, especially Velvia 50, Ektachrome E100G, Elitechrome 100 and 200.. they all suffer from this.
• Why I am not seeing this with studio flashes? Perhaps the all reflectors/softboxes and the careful exposure gives more neutral result?
• As the colour temperature of the flashes should be quite near to daylight (perhaps bit colder), then why photos taken outdoor does not suffer from this issue?

Is there anyone who has run to same problem?
If I cannot solve this otherwise, I guess the best solution is CC filter, but how to determine the correction?

Next I will take some photos with gray cards included, then I can at least get some readings with densitometer from something that I is known to be neutral.

Here's sample of Provia 400X, photographed on light table. These are taken with Nikon FE2, using TTL and giving one stop more exposure from the body to compensate light wall's that would otherwise fool the TTL metering.

This was discussed in 2005.

(there was a url link here which no longer exists)

I was going to suggest that maybe the hotshoe flashes aren't putting out as much light on the walls. The shirt on the boy looks much whiter and is closer to the flash. The studio flashes have a higher output so would make a larger area whiter. It's basically an underexposure issue, I think. Not your problem, but the flashes not being powerful enough to reach the walls, just the subjects. Since the walls are a large area of the photo, it makes our eye see it all as purple, but I think if you crop way down to parts of the shirt and the woman's hair that they look much more white.
I don't do as much color, so take this with the usual grain of salt.

Winger has it dead right.
If you look at the wall behind the boy it's white. If you get an inexpensive "slave" flash and put it on a table, obscure it with a plant, lamp or some other thing. it should have the power to lighten the wall for you.

This happens with most hot shoe automatic flashes. The sensor reads the foreground(roughly central) subject and quenches the flash when that area receives enough light, if it didn't the foreground would wash out.

Thank you. I haven't use slaves in these as there usually childrens playing around (which limits places where slave can placed) but perhaps it's good solution after all.

several of the inexpensive ones are smaller than a cigarette package. If you wanted to you could use tape or a rubber band to keep them out of the children's reach.
When I used these I wouldn't point them directly at the wall, but upward so the light would bounce off the ceiling.

I have now done enough testing, developing my own using Tetenal's E-6 kit and using two commercial lab's (sad that there are only two professional lab in my country).
It seems that big part was really on the lab side. So problem wasn't directly caused by flashes, but flashes bring it more visible due the underexposed backgrounds.


Next... I will use additional flash when ever possible and I have to calibrate my home E-6 kit's time...