[SOLVED] Need a simple (discrete) buffer

[d123]

Hi,

How can I buffer the following circuit, without using an op amp? Do I need to buffer it, were it to be fed to e.g. an ADC reference pin?

I tried simulating a BJT NPN one, the common collector version (biased base, two capacitors, and without the capacitors), but I'm not sure why I've been unable to get the 1V in > 1V out.

Tried with a differential pair and common collector output stage and pretty pathetic to see the failed attempts to get anywhere near 1V out, I don't want to design an op amp, nor do I have that level of ability, I would just appreciate hints or suggestions on how to buffer this circuit, thanks.

Is there an issue with using a 1.5V supply and a ~680mV Vbe NPN transistor as a buffer for the below circuit?

 

[BradtheRad]

If you add 0.6V to your input, then you get 1V output from the classic NPN emitter follower. It automatically subtracts .6V.

Notice the bias sinks as much as 160 uA. If this is not too much for your bias to sink or source, then it may also be all right if the supply goes as low as 1.1 V.

 

[d123]

Hi,

Thanks Brad, I think I may go with a common base PNP as that way the buffer is only 1 BJT and 1 resistor (every additional component in that circuit subtracts from linearity over a temperature range), but want to check connecting it to another device won't affect desired 1V; and as with the emitter follower, I had to boost the input signal a little to cover losses across the PNP. I'll read a bit more today about discrete buffers anyway, and look into the emitter follower, possibly your solution is actually preferable, thanks.