I'm trying to make a low voltage detector similar to this one with whatever parts I have on hand. Instead of 47K resistors, I tried 56K resistors, and I used 2N3904 in place of C945 NPN's and for the diode I used 1N5231 zener. I begin to think I burnt some 2N3904's because with a 9V battery the lights seem to work properly at first then when I disconnected the battery and tried again, the lights did not work correctly. Sometimes, they both went on when only one should. Then I was reading up datasheets on 2N2222 and 2N3904 that the max voltage for the BE junction is 6V, so I wondered if the transistors are faulty due to 9V being fed into the base. If so, then what transistors that are commonly available should I use in place of then 2N3904 and 2N2222 that can handle higher voltages?

 

9V forward bias on the base will kill every transistor since saturation VBE is about 1V.
6V is the max reverse voltage.
You need to limit the current not the voltage.

 

I have also tried a resistor and diode in series with my setup as well. What would you say the highest values for the resistors I should use if I want to draw minimal current and still have the 2-transistor detector function well using a battery supply from 1.5 to 9V? I do have ultra bright lights (in the 10,000mcd range) I can use as output and they can still shine with 10K current limiting resistors.

 

Can you post your schematic here (file size <300k)?

 

Your schematic attachment is X'd out in your first post, so you need to post it here.
Can't help you without that.

 

ok I re-uploaded it. and I uploaded my working circuit.

It makes me wonder if the 2N3904's work differently between manufacturers. I replaced them with PN2222's and the thing works every time.

 

...
It makes me wonder if the 2N3904's work differently between manufacturers. I replaced them with PN2222's and the thing works every time.

I would say no after a career spanning 5 decades.
Maybe if your 2N3904's were counterfeits

 

Agree with PB.
There should be no significant difference in the operation of the 2N3904 versus the PN2222 in your circuit.

The Zener is upside-down in the bottom circuit.

Will you be using the circuit with a momentary push-button connection or continuously connected?

 

Be aware the Zener won't have a sharp turn-on characteristic if the current through it is only a few uA, as limited by the 100k resistor R5.

 

Which (above) means you need to make R5 and R3 much lower resistances. I would set it up to that the Zener can get at least 50 ma when the battery is fresh. The test current of 20 milliamps would be much better. I would also re-arrange the resitors:

To "test" button.
|
|
Zener
|
|-------------Resistor (2.2k) -------------Base of transistor
|
Resistor (1k)
|
|
Ground

This way the transistor won't get any base current until there is a significant amount of current running through the Zener.

 

oops I showed the circuit wrong. I have the zener the correct way. and I'm happy with the circuit.