LED Test light

[baintx]

I built a test light using A resistor and LED in series.
The total voltage is 74 DC volt, the resistor is 3900 ohm.
Current calculates at .0189.
LED's are rated at 30 ma and 3.2 to 3.8 volts.
My LED's work for a short time then burn out. Can anyone
explane to me why and what I need to do to correct this problem.

 

[Nigel Goodwin]

I built a test light using A resistor and LED in series.
The total voltage is 74 DC volt, the resistor is 3900 ohm.
Current calculates at .0189.
LED's are rated at 30 ma and 3.2 to 3.8 volts.
My LED's work for a short time then burn out. Can anyone
explane to me why and what I need to do to correct this problem.

Under those conditions they shouldn't burn out, so I suspect the conditions are probably other than you thought?.

74V is a very strange value, where is it coming from?.

 

[TheOne]

Make sure that the resistance is 3.9K by checking with a meter. I have seen 390R at 2% (or 392R with more colored bands) being mistaken for 3.9K by people in the past.

 

[baintx]

Voltage source is a locomotive with regulated generator output.
Measured my resistor at 3.88 k.

 

[tansis]

That explains it , your getting spikes in the supply volage from inductive loads.

 

[TheOne]

Maybe it will be a good idea to put a 3.9~4.7V-1W zener diode in parallel with the led to protect it from any spikes and shunt the excess current.

 

[baintx]

Thanks for the help, I will try the zenor idea.
I appreciate the information and help.

 

[Phasor]

put a 3.9~4.7V-1W zener diode in parallel with the led

To clarify TheOne, it is implied as anti-parallel, not direct parallel... (zener normally reverse biased)