RustyBug wrote:
ooops!!! ... forgot about the gas vs. filament. (Won't a change in voltage through the gas still render a different result?)
Wikipedia has a nice writeup on this topic, and it's trickier than I thought. The spectrum is indeed affected by amperage but, for flash tubes in photo applications, the spectrum is not simply black body. The effective color temperature is influenced by spectral lines from the particular gas (Xenon in our case).
rico wrote:
Wikipedia has a nice writeup on this topic, and it's trickier than I thought. The spectrum is indeed affected by amperage but, for flash tubes in photo applications, the spectrum is not simply black body. The effective color temperature is influenced by spectral lines from the particular gas (Xenon in our case).
The physics of ionized Xenon gas discharge tubes are as follows: The color spectrum and flash duration are determined by arc length, gas fill pressure and applied voltage. Short arc length tubes with high fill pressure and high applied voltage produce very fast durations and extremely high color temperature (ala Edgerton).
For a typical studio flash tube that is designed to produce "daylight" color spectrum in the 5600°K range with Full power voltages typically between 300VDC and 900VDC, lowering the applied voltage will reduce power approximately based on Ohms Law (Power = Voltage squared/tube resistance) and results in a lowering of color temperature of typically 80°K per halving of power, and a lengthening of flash duration of approximately 20% per halving of power.
The process is not completely linear. Base flash duration can be optimized within a tube design. Very short duration tubes can be inefficient due to the power lost in the impedance of the flash capacitors and result in greater current flow through all the components, and will also have a shorter life span. The extreme currents can decrease the overall reliability.
My 30 years of research on these subjects conclude that, for a flash unit in the 600WS range, the optimal set of conditions . . . efficiency and reliability at short flash durations intersect at a full power t.5 flash duration in the 1/1500 to 1/2200 second range . . . typical for most high quality short duration pro flash units on the market. This range provides typical flashtube life of 250,000 to 1,000,000 flashes.
For further information, "t.5" indicates the time it takes for 1/2 of the available power to be dissipated while "t.1" indicates the time it takes to dissipate 90% of the available power. For a given non-IGBT system, the t.1 time is typically 3 times as long as the t.5 time. The t.1 time fairly closely resembles the action stopping capability in comparison to an equivalent camera exposure time using continuous light.
BrianO wrote:
Yes, it will. That's why different types of output control have such different results.
Some strobes change the output by varying the voltage, others by using a constant voltage but varying the amount of time that the voltage is applied.
Only IGBT controlled units can control the time the voltage is applied.
Capacitor switching is usually used in power pack systems to reduce power, often in conjunction with varying the voltage for fine tuning. If one reduces the capacitance in half one can expect the flash duration to be twice as fast and the color to remains essentially the same. However, as capacitance is reduced, the internal impedance of the flash capacitors dissipates more and more of the available power, so efficiency and flash durations deviate from a linear curve.
BrianO wrote:
What about thyristors as used in some hotshoe flashes? How do they work?
They work like Einstein. IGBT switching is easy at the power levels and longer flash durations used in hotshoe flashes. Current and voltage is much smaller so a single low power IGBT can be used. We have to use multiple high power IGBTs. These were not readily available and very expensive until recently. Photogenic Studio Max uses IGBTs, but they are way long flash duration in order to avoid expensive multiple IGBTs. Long flash duration eases the current load on IGBTs. We are switching close to 2000 amps . . . takes big IGBTs.
Paul Buff wrote:
Yes, it has a shoulder strap and we are working on a nice optional carry bag . . . should have it about the same time as Mini.
Cool.. that's awesome!!
I'm sure whatever you come up will be cool. But please try to keep in mind, conservative dark colors for the optional carry bag. The florescent bright colors for example on the VBII just stream for attention or "steal me" I'm easy to find when your stashing your VBII in a corner w/ some extra gear at a wedding.
BrianO wrote:
...Some strobes change the output by varying the voltage, others by using a constant voltage but varying the amount of time that the voltage is applied.
Paul Buff wrote:
Only IGBT controlled units can control the time the voltage is applied.
BrianO wrote:
What about thyristors as used in some hotshoe flashes? How do they work?
Capacitor switching is also possible. Turning off and on banks of capacitors instead off lowering voltage. Profoto uses this in combined with voltageregulation in their more expensive packs that's why you see a huge drop in K till you reach a full stop. WL X1600, X3200 carry this function too with a manual switch while profoto I believe have this integraded into the power variator.
Capacitor switch will shorten durations but with two caveats:
1. It is pretty much limited to full f stop changes and not practical for fine adjustments like 1/10f or even 1/3f.
2. It retains the tapered "tail", So a t.1 time of, say 1/2000 still has 10% of the energy left following the 1/2000 second period. 10% of the energy equates to -3.2f stops below peak energy. It takes another 1/666 second for the tail to fall to -6f. (t,03) So if you are shooting sports for instance, with a white ball, at 1/2000 second the ball is still illuminated sufficiently to produce considerable motion blur when compared to a true exposure time of 1/2000.
IGBT switching absolutely shuts the tail off, so for an equal t.1 time, the IGBT unit will stop action considerably cleaner than a capacitor switched unit,
Okay, so capacitor switching, IGBTs, and thyristors can all curtail the flash based on time, rather than changing voltage, but each has a different way of handling the details.
BrianO wrote:
Okay, so capacitor switching, IGBTs, and thyristors can all curtail the flash based on time, rather than changing voltage, but each has a different way of handling the details.
Correct. Almost all monolights use voltage control, with some like WLX combining cap switching and voltage control. Most power packs primarily use cap switching but augment it with voltage control for fine tuning. All speedlights use low power IGBT switching, which provides the greatest control of color and duration. It's necessary for TTL and HSS functions. Top end packs like Grafit, Scoro, etc us high power IGBTs, as does Einstein and Photogrenic Solair.
As mentioned, the sharp light cutoff of IGBTs results in cleaner and effectively shorter action stopping than equivalent t.1 times obtained from cap switching or voltage control.
IGBT flashes require a processor algorithm to vary charge voltage as power is adjusted to achieve constant color. Without the voltage control algorithm the color temperature rises dramatically as power is lowered . . . the opposite of voltage controlled flashes.
Einstein and ultra high end IGBT packs allow changing the voltage vs power algorithm to achieve either constant color or absolute minimum durations (at the cost of color getting bluer as power is lowered.) Einstein "action" algorithm allows color to rise from 5600°K up to about 6300° at low power to achieve 1/13,500 t.1 durations at reasonable color. Durations could be made even shorter with different algorithm but color would rise into the 10,000°K range. Einstein Color Mode holds color constant at 5600°K but still achieves 1/8000 second t,1 times at lower power settings.
Paul, I'm extremely curious and somewhat worried about potential glitching or crashing of a Profoto Compact300/600 or D1-500 power supply.
While I know that testing a non-house brand light is never a priority, I would very much like to know the compatibility. If you do come across these type of heads, please share your findings here, thanks!
"In the mean time I too would like to see an independent test of the mini vagabond with the following lights:'
I would too when it exists and when it's delivered. I'll be one of the first to buy one if it's any good.
"Do it right and In six or seven years you too can be as wealthy as Fred Miranda, or maybe even Mr. Buff."
What makes you think I'm not already? Pretty condescending Ellis. I'm sure Mr. Miranda appreciates you telling his paying customers to leave and go somewhere else. Great for business.
... forgot about the gas vs. filament. (Won't a change in voltage through the gas still render a different result?)