William wrote:
>>
Cheap flashes use small, inexpensive transistors that can't tolerate abuse.
The oscillator is therefore designed so that the internal resistance of
ordinary alkaline cells is the limiting factor in how much current can flow.
Substitute nicad or NiMH cells, and the oscillator draws so much current it
overheats and fails.
This story has been repeated for many years. I don't know if it's true, but
it's _theoretically_ possible.
<<
I believe it is largely true. Almost all on camera and even most handle mount
flashes with the invertor (voltage convertor) in the handle (eg Metz45,Sunpak
522,T45) use a blocking oscillator invertor. A blocking oscillator, with an
almost short circuit load (discharged flash capacitor), tends to draw almost
as much current as the battery can supply (limited mainly by battery
resistance,winding resistance of transformer and base drive current) The
Olympus T32 invertor is a modified blocking oscillators with a load current
sensing scheme to increase base drive current proportionate to flash
capacitor charging current. This reduces power consumption while idling but
does draw large currents during charging the flash capacitor. Although this
adaptive drive can increase maximum current it can still reduce transistor
dissipation, compared with fixed drive, since it keeps the transistors
saturated. (heat depends on saturation voltage and current). The transistors
can fail from over current or overdissipation (heating) whichever comes
first. With a lower resistance battery the peak current can come first in a
typical older flash design, or the transistors may overdissipate if the
transistors are not driven hard enough so at the now higher current they come
out of saturation.
Joel W. wondered about playing "Russian roulette" with his flash.
3rd party rechargeable **battery** packs (not the HV Turbo packs) such as
those from Quantum which are designed for T32 (and other flash) use, have
enormous short circuit current capability. In the case of the original 6V
Quantum battery probably over 100Amps (instantaneous) (It has a slow cutout
to prevent user burns!). It is essentially limited more by stray lead
resistance,connector resistances etc than by battery resistance. (Battery
resistance alone limits to ~160A for the Gates cells used originally) . AA
size NiCds alone have ~35A short circuit current while NiMh's have about
~40A, not a huge difference, while Alkalines are more like ~10A. What all
this means is that the Quantum packs are theoretically tougher on your flash
than the NiMh's. Since Quantum has succesfully sold it's batteries for years
it seems likely that the NiMh's are ok, provided you don't recycle full
dump's continuously.
Other members have posted comments about the problem of inadequate transistor
heat sinking for many repetitive flashes at full power. This is somewhat
inherent in a small plastic (thermal insulating) plastic box flash. The flash
tubes also get hot, so melting of the plastic reflector is a theoretical
issue. Remember the flash tube visible light conversion efficiency is only
something like 50% and each flash ~70Joules. Handle mount flashes like the
Sunpak or Metz units, have an internal metal heatsink on the transistors but
also probably overheat with continued recycling full dumps, since the
heatsink has a finite thermal capacity and it is thermally insulated by the
case. Another reliability issue with continuous recycling at full dump, is
that the Mitsubishi series thyristor (SCR) used in many small flashes
(including T32,Sunpak 522) is only rated for a relatively modest number
(thousands) of full dumps and that only with a rest between each flash.
Check the Quantum battery website for some related discussion in the pdf
files for the various battery packs they make. (compatible with Olympus and
others).
Making your NiMh-unsafe flash work with niCd's,NiMh etc : You can modify your
Q310 or similar NiCd unsafe flash to use rechargeables by adding a small
series resistor to the wiring from the battery. You essentially simulate a
high resistance Alkaline. It will slow recycle time a bit with genuine
Alkalines!
You need to add a series resistance of about 120mOhm per cell or 0.48 Ohm
(0.5Ohm nearest value) for 4 cells. In theory you need ~2W resistor but the
duty cycle is low so you should be able to get away with a 1W or maybe less
resistor. Years ago I converted a non-TTL Cannon flash for Olympus TTL use
and added a series resistor to protect it from NiCd's at the same time. This
"Cheap" flash had a warning about NiCd's printed in the case.
Regards,
Tim Hughes
NoSpam_TimHughes@xxxxxxxx
(remove NoSpam_ to email me)
|