Dave_H
Flashlight Enthusiast
Primary CR123A heavily loaded spends most of its time down around 2.5v/cell. To mimic with
LiFePO4 would require one diode drop per cell, wasting about 20% of battery capacity, should not
be necessary using a 6v bulb.
Running a small bulb at 10% above its rated voltage, only about a half volt higher, may shorten life a bit
but not likely the primary reason for "popping". Just trying to understand the failure mechanism...
Failure of incandescents is largely due to repeated stress on filament as result of high inrush current to cold
filament at turn-on. Possibly it's more of a factor with smaller bulbs.
Measured cold resistance of a 120v 100W bulb was around 12 ohms. More representative is a 4.8v 0.4A
krypton bulb, less than 1 ohm.
I recall a product for household bulbs, a small slug inserted in the socket, which claimed to reduce
this problem; don't know what was inside, possibly a PTC resistor.
Series diode will use up about 10% of the battery in this case. Thinking about it though, it probably helps reduce the
cold inrush, so Katherine' s may not be a bad idea. If it were me I would have a spare bulb and try with/without.
This all assumes there is enough space for a diode "insert"; or that it is otherwise wired in. Perhaps some
incan. lights already have this?
Using two parallel diodes e.g. 1N4004 make sure they are identical parts preferrably from the same
batch, otherwise one may "hog" the current which makes two in parallel less effective. You may find
a single device, single or dual diode in DPAK with about same thickness as 1N4004. May be able to scrounge
such a part from old PCB. Another possibility is regular silicon rectifier (not Schottky) SMT device in SMC etc.
package, typically 3A or better.
Dave
LiFePO4 would require one diode drop per cell, wasting about 20% of battery capacity, should not
be necessary using a 6v bulb.
Running a small bulb at 10% above its rated voltage, only about a half volt higher, may shorten life a bit
but not likely the primary reason for "popping". Just trying to understand the failure mechanism...
Failure of incandescents is largely due to repeated stress on filament as result of high inrush current to cold
filament at turn-on. Possibly it's more of a factor with smaller bulbs.
Measured cold resistance of a 120v 100W bulb was around 12 ohms. More representative is a 4.8v 0.4A
krypton bulb, less than 1 ohm.
I recall a product for household bulbs, a small slug inserted in the socket, which claimed to reduce
this problem; don't know what was inside, possibly a PTC resistor.
Series diode will use up about 10% of the battery in this case. Thinking about it though, it probably helps reduce the
cold inrush, so Katherine' s may not be a bad idea. If it were me I would have a spare bulb and try with/without.
This all assumes there is enough space for a diode "insert"; or that it is otherwise wired in. Perhaps some
incan. lights already have this?
Using two parallel diodes e.g. 1N4004 make sure they are identical parts preferrably from the same
batch, otherwise one may "hog" the current which makes two in parallel less effective. You may find
a single device, single or dual diode in DPAK with about same thickness as 1N4004. May be able to scrounge
such a part from old PCB. Another possibility is regular silicon rectifier (not Schottky) SMT device in SMC etc.
package, typically 3A or better.
Dave
