Philips Hue for VC Enlarging and Safe Light

[athbr]

I've never heard of anyone doing this so I am assuming it would be a terrible idea but...

Could I use a Philips Hue lightbulb inside my enlarger (durst m601) for VC control in lieu of making an LED head myself?

Ditto with safelights.

 

[koraks]

On the largeformatphotography I've seen some people discuss it and in any case Pere Casals over there seems to have used a remote controlled RGB les to his satisfaction. Not the Philips bulb specifically, but that would probably work as well.

 

[L Gebhardt]

Be aware of startup time. I’ve played with the led bulb Pere Casals mentioned and it’s not instant on, which will make test strips hard if you build up exposures.

 

[jim10219]

The problem is you can't control the bulb with a standard timer. The bulbs need to be left with continuous power and then controlled by an app. If you disconnect that power and then reconnect it, like you would a normal bulb, they'll go back to their default settings, which means a warm white. So in order to use them for VC exposures, you'd have to leave them powered on all of the time, and then try to control them through an app. And that makes timing difficult, plus now you have your phone glowing brightly in the darkroom. You can probably create a routine that would control the light in an app to get a set exposure time and color. But that would be a bit of a headache.

 

[athbr]

The problem is you can't control the bulb with a standard timer. The bulbs need to be left with continuous power and then controlled by an app. If you disconnect that power and then reconnect it, like you would a normal bulb, they'll go back to their default settings, which means a warm white. So in order to use them for VC exposures, you'd have to leave them powered on all of the time, and then try to control them through an app. And that makes timing difficult, plus now you have your phone glowing brightly in the darkroom. You can probably create a routine that would control the light in an app to get a set exposure time and color. But that would be a bit of a headache.

That part isn't much of a concern for me as I mess with arduinos quite a bit. So creating a physical controller that can manipulate the bulbs via UDP wouldn't be too hard. It would make for quite a nice automation project.

My biggest fears are those surrounding evenness of light, precision of colour commands, fogging from using them as safelights, and of course what Gebhardt mentioned about startup times.

 

[ic-racer]

This is not a new problem to solve. This first-generation Omega diffusion heads had a similar 'slow start' issue that was solved with a below-the-lens shutter.

 

[ic-racer]

The latest generation of Durst diffusion heads also had slow start with the 2000W incandescent lamp, which was solved with an above-the-lens shutter:

 

[koraks]

@OP, if you're handy with arduinos, I'd consider skipping the built-in automation altogether and start with led modules or strips, some fets and pwm control them directly.

 

[athbr]

@OP, if you're handy with arduinos, I'd consider skipping the built-in automation altogether and start with led modules or strips, some fets and pwm control them directly.

Koraks that was my original idea. But I saw one of these at the store and it just gave me the what if bug.

It certainly would cut down on wiring the leds myself and to boot would also allow controlling all lighting (safelighting, general lighting) from one physical control panel.

Any particular reason why doing from scratch would be advisable beyond cost?

cheers

 

[1kgcoffee]

The Philips hue is behind in technology from lifx. I currently use lifx for color enlarging and it works quite well. Have not managed to find the Hue and saturation values that correspond to 00-5 contrast filters. Haven't really tried. But if you can figure it out and are good at programming it's possible to program split grade printing into the light by timing the exposure for transition from the two colours.

I'm convinced that if this were available before digital everyone would printing on low tech rgb lights.

 

[koraks]

Any particular reason why doing from scratch would be advisable beyond cost?

Not having to deal with partly reverse engineering a probably highly integrated solution. Also from a physical perspective it's likely quite cramped inside the housing and it may be a b*tch to connect your controls to the existing circuitry.

The argument that this would be an integrated solution I don't quite follow; if you start with three led arrays in R, G and B, you would achieve the same, couldn't you?

 

[athbr]

Not having to deal with partly reverse engineering a probably highly integrated solution. Also from a physical perspective it's likely quite cramped inside the housing and it may be a b*tch to connect your controls to the existing circuitry.

The argument that this would be an integrated solution I don't quite follow; if you start with three led arrays in R, G and B, you would achieve the same, couldn't you?

Just to clarify some confusion. I meant a physical controller in the sense of a box with buttons. The commands are all sent to the hub via HTTP and the bulb has all the controllers for it inbuilt. Luckily the API seems pretty well documented. There is ofc the issue that the commands dont register as fastly and accurately as physical connections but only one way to find out.

So it would be an integrated solution in two regards:

1. It would be fully wireless allowing to connect all lights (general, safe, enlarger) to one control unit regardless of their distance
2. All the components needed for this (rg&b leds, dimmers, wifi shields) are all contained within one off the shelf product that hooks into a standard bulb socket

 

[athbr]

The Philips hue is behind in technology from lifx. I currently use lifx for color enlarging and it works quite well. Have not managed to find the Hue and saturation values that correspond to 00-5 contrast filters. Haven't really tried. But if you can figure it out and are good at programming it's possible to program split grade printing into the light by timing the exposure for transition from the two colours.

I'm convinced that if this were available before digital everyone would printing on low tech rgb lights.

Good to know considering they are cheaper to get set up with

Any idea how testing hue/saturation to match 00-5 would work in principle?

 

[koraks]

Thanks for the clarification; yes, that makes sense. I can see why you'd consider this route and it's certainly worthwhile pursuing. I think a http connection wouldn't be problematically slow for this application.

 

[1kgcoffee]

Athbr,
Yes, I'll post some stuff tonight. It goes from an Amber colour to magenta. Basically there are two layers in vc paper, blue and green sensitive . Middle grade reacts with both equally. Split grade printing is essentially creating a tone curve

 

[AgX]

The problem is you can't control the bulb with a standard timer. The bulbs need to be left with continuous power and then controlled by an app. If you disconnect that power and then reconnect it, like you would a normal bulb, they'll go back to their default settings, which means a warm white. So in order to use them for VC exposures, you'd have to leave them powered on all of the time, and then try to control them through an app. And that makes timing difficult, plus now you have your phone glowing brightly in the darkroom. You can probably create a routine that would control the light in an app to get a set exposure time and color. But that would be a bit of a headache.

You could use a shutter.

Already in heydays of incandescanrt enlarger lighting a shutter was proposed and offered.
(Though I do not see the benefit of such hassle in this case.)

 

[athbr]

You could use a shutter.

Already in heydays of incandescanrt enlarger lighting a shutter was proposed and offered.
(Though I do not see the benefit of such hassle in this case.)

Out of curiosity, how would one hack a shutter for this. Could I put a Copal between the lensboard and the lens?

How would you do it?

cheers

 

[jim10219]

Out of curiosity, how would one hack a shutter for this. Could I put a Copal between the lensboard and the lens?

How would you do it?

cheers

That would be one way to do it. You'd probably have to set it to Bulb mode though. If you wanted something simple, you could also place something opaque in the below the lens filter holder, preferably matte black to cut down on reflections. Or, if you wanted to do something a little more fancy, use the outlet on a standard timer to operate a shutter of some kind powered by a solinoid, so that instead of the timer powering the light, it powers the solenoid that holds the shutter open. That might require an appropriate AC/DC converter.

 

[btaylor]

Packard shutter mounted above or below the lens? Operate it with your foot.

 

[AgX]

Once, in the 80s, there was a in front of lens shutter for enlarging. Offered in Germany.

 

[1kgcoffee]

Good to know considering they are cheaper to get set up with

Any idea how testing hue/saturation to match 00-5 would work in principle?

5 is magenta. It is high grade and filters out all the green light. So almost everything is exposing the blue sensitized layer in the paper. Notice how on the image below it is opposite green
00 is yellow or amber, I'm not sure. Notice how it is opposite blue. So you are transmitting all but blue light, exposing the low contrast green layer for midtones.
cmyk-HSB by Aaron, on Flickr

The numbers in pink are hue.
The four numbers next to it are C-M-Y-K.
What the image does not show clearly, is that as you get closer to the center of the wheel, the saturation decreases until you get pure white light. With the LIFX bulb, the first number is hue, second is saturation, third is brightness HSB eg (330,0.25,1.0) The setting I use to print kodak gold on kodak endura paper. Brightness is nothing more than fstop for controlling exposure.

Now imagine you have a negative. Split grade printing means dividing the exposure between 00 and 5, or yellow and magenta. Exposing for the low contrast layer for midtones and highlights first, then burning in the shadows for some nice crunch. This is a crude analog attempt at adjusting the tone curve of the printed image. A 5 second exposure at 00 (what I THINK is (60, 1.0, 1.0)) will give you flat contrast. 5 seconds at (300, 1.0, 1.0) should give maximum contrast with blown highlights and deep shadows. Depending on the negative you will want to print on both to give rich full tones.

60 and 360 are GUESSES. The test would be to print off 60,1.0,1.0 first, then use a 00 variable contrast filter at something like 4000k or 5000kelvin colour temp, using the lifx bulb also, comparing the final prints of similar exposure. Repeat for 360,1.0,1.0 and VC5. The prints should be very close for similar exposure. And the light projected should look pretty similar.

If you expose at 360, 1.0, 1.0 you should get nothing as this is safe red light.

Most likely the 5 grades correspond to five points straight across the wheel from hue of 60 to 300 with I guess grade 2.5 being equal to a hue of 0 with 50% or less saturation. But maybe cyan would work? (180, 1.0, 1.0) since it is halfway between green and blue on the wheel. You would have to test grade 2.5 vs these.

Now, the questions which I have not deeply explored are:
1. Does 00 and 5 correspond to hues of 60 and 300 or something else?
2. Can you program two seperate timed exposures. These bulbs are almost instantaneous.
3. Can the colour changing feature be utilized to do a timed exposure either straight across the wheel, through white, or around, accelerating or decelerating to give the contrast you want?
4. Another possibility is that it is not 60 and 360 at full saturation, but maybe 50% to correspond to grade 00? Have a look at (60, 0.5, 1.0) and then compare to how a filter of grade 00 looks under the lens with no negative.

But, I do very little B&W printing. I don't have the time or skill to program this. My experience is mostly RA-4

 

[1kgcoffee]

One caveat for the LIFX bulb. It is connected to the timer and sometimes has a nasty habit of resetting to default colour of room light after being turned on multiple times. Maybe mine is a dud or they have fixed the issue. Xiaomi also came out with a similar bulb which is much cheaper. I have one but haven't experimented with it yet.

 

[athbr]

5 is magenta. It is high grade and filters out all the green light. So almost everything is exposing the blue sensitized layer in the paper. Notice how on the image below it is opposite green
00 is yellow or amber, I'm not sure. Notice how it is opposite blue. So you are transmitting all but blue light, exposing the low contrast green layer for midtones.
cmyk-HSB by Aaron, on Flickr

The numbers in pink are hue.
The four numbers next to it are C-M-Y-K.
What the image does not show clearly, is that as you get closer to the center of the wheel, the saturation decreases until you get pure white light. With the LIFX bulb, the first number is hue, second is saturation, third is brightness HSB eg (330,0.25,1.0) The setting I use to print kodak gold on kodak endura paper. Brightness is nothing more than fstop for controlling exposure.

Now imagine you have a negative. Split grade printing means dividing the exposure between 00 and 5, or yellow and magenta. Exposing for the low contrast layer for midtones and highlights first, then burning in the shadows for some nice crunch. This is a crude analog attempt at adjusting the tone curve of the printed image. A 5 second exposure at 00 (what I THINK is (60, 1.0, 1.0)) will give you flat contrast. 5 seconds at (300, 1.0, 1.0) should give maximum contrast with blown highlights and deep shadows. Depending on the negative you will want to print on both to give rich full tones.

60 and 360 are GUESSES. The test would be to print off 60,1.0,1.0 first, then use a 00 variable contrast filter at something like 4000k or 5000kelvin colour temp, using the lifx bulb also, comparing the final prints of similar exposure. Repeat for 360,1.0,1.0 and VC5. The prints should be very close for similar exposure. And the light projected should look pretty similar.

If you expose at 360, 1.0, 1.0 you should get nothing as this is safe red light.

Most likely the 5 grades correspond to five points straight across the wheel from hue of 60 to 300 with I guess grade 2.5 being equal to a hue of 0 with 50% or less saturation. But maybe cyan would work? (180, 1.0, 1.0) since it is halfway between green and blue on the wheel. You would have to test grade 2.5 vs these.

Now, the questions which I have not deeply explored are:
1. Does 00 and 5 correspond to hues of 60 and 300 or something else?
2. Can you program two seperate timed exposures. These bulbs are almost instantaneous.
3. Can the colour changing feature be utilized to do a timed exposure either straight across the wheel, through white, or around, accelerating or decelerating to give the contrast you want?
4. Another possibility is that it is not 60 and 360 at full saturation, but maybe 50% to correspond to grade 00? Have a look at (60, 0.5, 1.0) and then compare to how a filter of grade 00 looks under the lens with no negative.

But, I do very little B&W printing. I don't have the time or skill to program this. My experience is mostly RA-4

Hi 1kgCoffee,

Thank you so much. This is very comprehensive and I'm definitely going to keep referring to it. To get to your questions:

1. I think I'll do as follows. I'll find a negative that will make a decent test subject (or maybe just take a picture of a USAF chart for these purposes). And cycle between hues (50-310 with 10 degree steps) in a test strip to see where max and min contrast are to be found. I'll do the same for 00-5 and see if I can line them up.
2. I think this definitely can be done in software.
3. I think this also can be done but it would probably introduce new variables and wouldn't speed up the process meaningfully.
4. My assumption would be that here contrast = contrast hue * saturation + contrast white light * (1 - saturation). I believe it would be less complicated to use fully saturated light if possible.

Now a couple more questions for you if I haven't bored you out yet

1. How does Colour Temperature fit into this?

If I recall correctly higher temperatures correspond to more contrast. And If I also understand this, the LIFX has a separate panel for that.

How does that fit in to the colour wheel you presented? Does a (Hue, Saturation, Brightness) triple correspond to a particular temperature? Or is this actually a (Hue, Saturation, Brightness, Temperature) tuple, leading to a fourth variable to control? My naive assumption is that its best to keep things close to the temperature in my current incandescent bulb.

2. If the paper is actually Green, Blue sensitive then would using Green 120 and Blue 240 be equivalent to Yellow 60 and Magenta 300, respectively?

Cheers!

 

[athbr]

One caveat for the LIFX bulb. It is connected to the timer and sometimes has a nasty habit of resetting to default colour of room light after being turned on multiple times. Maybe mine is a dud or they have fixed the issue. Xiaomi also came out with a similar bulb which is much cheaper. I have one but haven't experimented with it yet.

I'm going to try and control it via software so I don't need a physical timer. If brightness can be dimmed down from 1.0 to 0 fast enough (or hue rotated fast enough to Red) then I can use these commands as a switch.

 

[1kgcoffee]

athbr,
Glad to help and I'm not bored at all. Very interested actually, glad to have someone on photrio with a shared interest in this bulb and would love to hear about your progress and help in any way I can.

From my limited testing the hues are probably not at full saturation. This is just visual comparison.

Here is some technical information from Ilford that might aid you:
https://www.ilfordphoto.com/amfile/file/download/file/1824/product/746/
According to that link I'm a bit off the mark. There are three emulsions, Blue, Green, and Blue Green. Maybe you can make sense of the log charts. Further in the technical information is a list of dual colour settings which gives magenta and yellow settings that correspond to filter grades. Notice how they are different for different dichroic heads. Using this information, we can convert using colorizer.org to get some starting points. Colorizer uses percentages of cmyk so you'll have to convert from the maximum for that head. Grade 00 for Durst is 115Y and 0M out of a maximum of 170. So that is 68% yellow. Typing 68 into colorizers for yellow gives HSB value of (60,0.68,1.0). Sounds about right hehe. This may save you a lot of headache.

To answer your questions,

1. You're right. According to the ilford link that I posted above, cold temperatures like 5000k will work but the grades are not evenly spaced. They will be evenly spaced with an incandescent bulb

Here are the values for color temperature:
http://www.vendian.org/mncharity/dir3/blackbody/UnstableURLs/bbr_color.html
and coverting hex to hsb:
https://rgb.to/hex/ffbb78
which gives us an HSB of (30,0.53,1.0)

The lifx bulb does have color temperatures built into the app so you can just use those.

2. I don't think the green and blue values would match up but I'm not sure why exactly. I think what you get with yellow and magenta is subtractive, so includes everything but certain portions of green or blue whereas green or blue has some of either in it - I think but not sure. It would probably give you muddy prints like a weird contrast curve in photoshop.

Please keep us posted!