How do you select the color temperature of your strobe(s)?

[Chris Ross]

As I understand things strobes emit a full spectrum centred around their colour temperature and they do this by generating a plasma which glows giving a black-body type continuous spectrum centred on a particular wavelength.  Technically it produces grey body radiation which is transparent to its own radiation.  This wavelength is what defines the colour temperature produced.  The temperature achieved inside the tube  corresponds to plasma temperature achieved so 4500-6000 degrees kelvin (Kelvin = degrees C plus 273) 

 

The peak wavelength moves to shorter wavelengths as the current density increases due to the plasma getting hotter.  If the strobe is using small linear tubes the current density needs to be higher to match the output produced from a larger tube so light is bluer.  The light produced is proportional to the volume of gas that is excited into a plasma.  So physically large tubes can use lower current densities and their light tends to be warmer.

 

Similarly a shorter pulse will shift warmer as the shorter duration means less energy is added to tube so peak temperature reached is lower, color temperature lower and peak wavelength longer.

 

[Adventurer]

2 hours ago, Chris Ross said:

Similarly a shorter pulse will shift warmer as the shorter duration means less energy is added to tube so peak temperature reached is lower, color temperature lower and peak wavelength longer.

Not sure if I understood what you wrote here correctly. They way I read it it‘s a wrong statement.

 

According to my lab experiment and oscilloscope values measured back with Subtronic Pro 160, Ikelite DS160 and Inon Z240, SEA&SEA YS.. some years ago the warmer flash tubes (Ikelite, Subtronic) create their higher power by leaving the light on longer at lower level. This also applies to all other ring flash tubes such as Seacam and OneUW.

 

The peak lumen number you were able to measure from a Z240 was higher than on the much stronger D160 or Subtronic. This gave me and my friends headaches, so we set up time logged testing with an oscilloscope. We then found that the z240 delivered his peak value much faster and the large (warm painted) plasma tube strobes took more time to ignite and needed stay on longer. All that time stuff in milli and microseconds.

 

There are some underwater applications were fast light delivery comes in handy and will result in sharper images.

Edited June 5 by Adventurer

[Guest]

41 minutes ago, Adventurer said:

Not sure if I understood what you wrote here correctly. They way I read it it‘s a wrong statement.

 

According to my lab experiment and oscilloscope values measured back with Subtronic Pro 160, Ikelite DS160 and Inon Z240, SEA&SEA YS.. some years ago the warmer flash tubes (Ikelite, Subtronic) create their higher power by leaving the light on longer at lower level. This also applies to all other ring flash tubes such as Seacam and OneUW.

 

The peak lumen number you were able to measure from a Z240 was higher than on the much stronger D160 or Subtronic. This gave me and my friends headaches, so we set up time logged testing with an oscilloscope. We then found that the z240 delivered his peak value much faster and the large (warm painted) plasma tube strobes took more time to ignite and needed stay on longer. All that time stuff in milli and microseconds.

 

There are some underwater applications were fast light delivery comes in handy and will result in sharper images.

The way it works is that a specific tube as an energy rating

 

Imagine that one has 100 J and another 200 J am using numbers that are easy to work out

 

When you shoot the first at half power and the second at half power they produce say 5500K

 

When you shoot both strobes at 100 J the first goes to 6000K and the second stays 5500K

 

As the lamp reaches the maximum energy the temperature gets cooler it is not the absolute amount of energy but what is the level of saturation in the bulb

 

In terms of power the way you deliver less power is to emit for a shorter time which reduces the energy so what Chris says makes sense

Edited June 5 by Interceptor121

[Chris Ross]

1 hour ago, Adventurer said:

Not sure if I understood what you wrote here correctly. They way I read it it‘s a wrong statement.

 

 

this is relative to the same strobe, so if any given strobe is pulsed for less time it will not reach the same peak/average temperature so it will be warmer than its maximum power temperature.  It doesn't apply comparing big round tubes to short linear tubes.

 

[DreiFish]

8 hours ago, Interceptor121 said:

thats right currently it is the reason why am not considering them

I am not sure how you can insert 21700 into a hold designed for AA if they are 18650 this is not going to be a major improvement

 

6 hours ago, Adventurer said:

Well, don‘t be too harsh with the manufacturers.
 

First, we do know if @DreiFish Spectrometer has been regularly and properly calibrated. Second, has it been done in a completely dark room ( I mean pitch black room ) or outside or living room environment?

 

The ambient light might heavily distort the spectrometers readings. 
 

If you want to be precise you need to take it into an Ulbricht Sphere.

 

However I highly appreciate Dreifish‘s home probing. Maybe he can let us more about his test environment? Furthermore I suggest to do multiple measurements for each unit to get an idea about measurement variance and errors.

 

Sadly I've misplaced my Ulbricht Sphere. It's probably somewhere inside the bag of infinite holding 😄

 

(I did do the testing in darkness to try to minimize ambient light impact) 

 

8 hours ago, Interceptor121 said:

It is the physics of xenon tubes higher energy particles are cooler

this is also rather interesting because it means that you need more power so you have margin to get warmer color and when you shoot at full power the light will already be cooler which we all experience in big school scenes

 

 

Mmm... I didn't observe the same results with the Marelux Apollo 3, but I suppose that uses 3 shorter straight flash tubes, not one large circular tube.  It'd be interesting if someone could test this with another circular flash tube strobe like the Retra or Ikelite or Seacam. 

6 hours ago, Adventurer said:

Xenon Strobe tubes and LEDs have a manufacturing variance and tolerance in that range. More closely to the 500k instead of 1000k though. So that offset may come from the component source.

 

Furthermore ambient light might have cooled down your measurement.

 

And as Massimo pointed out correctly full power measurement is cooler than in mid range color temp measurement. This would support your marketing accusation a little bit, if strobe manufacturers did not measure color temp at full output. The latter one however is a very practical approach as most of the users don’t fire at FULL all the time.

 

@DreiFish the Marelux MTL reading freaks me out?!? I cannot really get my head around why a lower strobe power should be the coolest of all 🙏😇 Maybe you guys can up with ideas for this? 

 

Marelux in MTL just shifts the power scale so that the top setting is actually 1/4th of full power according to their manual and observations. The color temperature is the same as at full power (within measuring error). 

[Guest]

1 hour ago, DreiFish said:

 

 

Sadly I've misplaced my Ulbricht Sphere. It's probably somewhere inside the bag of infinite holding 😄

 

(I did do the testing in darkness to try to minimize ambient light impact) 

 

 

Mmm... I didn't observe the same results with the Marelux Apollo 3, but I suppose that uses 3 shorter straight flash tubes, not one large circular tube.  It'd be interesting if someone could test this with another circular flash tube strobe like the Retra or Ikelite or Seacam. 

 

Marelux in MTL just shifts the power scale so that the top setting is actually 1/4th of full power according to their manual and observations. The color temperature is the same as at full power (within measuring error). 

Entirely possible that the same energy is spread to three tubes each one not at full power which would give better temperature or at least no drop

[Adventurer]

15 hours ago, Interceptor121 said:

16 hours ago, DreiFish said:

Sadly I've misplaced my Ulbricht Sphere. It's probably somewhere inside the bag of infinite holding 😄

 

(I did do the testing in darkness to try to minimize ambient light impact) 

 

 

Mmm... I didn't observe the same results with the Marelux Apollo 3, but I suppose that uses 3 shorter straight flash tubes, not one large circular tube.  It'd be interesting if someone could test this with another circular flash tube strobe like the Retra or Ikelite or Seacam. 

 

Marelux in MTL just shifts the power scale so that the top setting is actually 1/4th of full power according to their manual and observations. The color temperature is the same as at full power (within measuring error). 

Expand  

Entirely possible that the same energy is spread to three tubes each one not at full power which would give better temperature or at least no drop

 

I do not understand that we are discussing in the wrong direction.

DreiFish gave these values:

Marelux Apollo III = 6808K (at maximum I suppose)
Marelux Apollo III MTL = 7092K 

Difference almost 300K cooler in MTL.

As MTL mode reduces the strobe output to recycle faster, I would assume at least the same 6800K or slightly warmer color temp, when following your previous arguments about single flash elements output emitting higher energy and more cooler photons.

Do we agree on this point or is there a misunderstanding we can work out?

 

My intuition was that Marelux will evenly reduce the power output on each flash element.

--- 

So if the above is true and agreed upon, DreiFish's measurement on Apollo III might give us a hint that the 3 flash elements do not fire at the same time in MTL mode. Marelux might eventually archive the high recycle times by rotating the 3 elements for light emmission and giving up to maximum output at each element, before the other two take turn again.

Edited June 6 by Adventurer

[DreiFish]

22 minutes ago, Adventurer said:

 

I do not understand that we are discussing in the wrong direction.

DreiFish gave these values:

Marelux Apollo III = 6808K (at maximum I suppose)
Marelux Apollo III MTL = 7092K 

Difference almost 300K cooler in MTL.

As MTL mode reduces the strobe output to recycle faster, I would assume at least the same 6800K or slightly warmer color temp, when following your previous arguments about single flash elements output emitting higher energy and more cooler photons.

Do we agree on this point or is there a misunderstanding we can work out?

 

My intuition was that Marelux will evenly reduce the power output on each flash element.

--- 

So if the above is true and agreed upon, DreiFish's measurement on Apollo III might give us a hint that the 3 flash elements do not fire at the same time in MTL mode. Marelux might eventually archive the high recycle times by rotating the 3 elements for light emmission and giving up to maximum output at each element, before the other two take turn again.

 

Interesting concept. In the first instance, I'd actually attribute the difference between 6808k and 7082k readings as within the bounds of the margin of error of the Sekonic, rather than reflecting an actual difference in color temperature between Manual and MTL modes. So.. without further testing, which I can do when I'm back home, I wouldn't read too much into it. 

 

As for the flash tubes being used in sequence for MTL rather than all firing at once.. I suppose that's one way to implement it, but it would lead to pretty discernable changes in lighting and hotspots from shot to shot. Again, I haven't tested this, but it's straightforward to test on a blank wall. Will perhaps do it at home.