I am sorry but I have a very newbie question. I have some circuit from which I get the two types of outputs, depending on how I look at it with an oscilloscope. In the first case it is just directly connecting to the output through a very big resistor without any load. In the second case, I use 1k to ground. Obviously, the first case is problematic, because I am basically capacitively read the output without any load. However, this is the shape that I want. Could you please suggest a proper way of arranging the output stage to get the pulse shape as shown in the first case?
Can you put some numbers on these waveforms, especially time and maybe give other details. I'll try to answer with what's given...
The first case looks like a high impedance source, something like a 10k output impedance. Its not affected by 100k (likely 100k with 10 meg in series from the scope). You should be able to just use a 10x scope probe- the 100k isn't doing much. It might be isolating the direct cap load of the probe...
The second case is pretty confounding- it doesn't look possible. The first pulse is what you might get with a capacitively coupled output, the width of the pulse is a few RC's but the second pulse is somewhat impossible from a linear circuit, it rises before the output rises then follows it down. This is non causal and suggests some kind of triggered delay from the rising edge - not linear. A classic under cap coupled output would have the the first edge create a pos pulse and the second create a neg pulse.
Thank you. Here is an example of the problem I am dealing with.
I have a high impedance input signal on R1. I am trying to convert it to a low impedance output on R2. Obviously, if I further reduce the input current by increasing R3, my output voltage on R2 becomes zero. Naively, I thought that the proper way of dealing with it would be to have an op amp buffer placed in-between R1 and R2, but somehow it does not work. Does anyone have a suggestion? The input signal is about 10-20mV.
What is Purpose of R2?
If you want to drive coaxial cable with 50Ohms WaveImpedance with signals of high enough frequenzy or short enough rise / fall times
and you want to correctly source terminate?
Then R2 needs to be in series to the output, not to ground
I have a high impedance input signal on R1. I am trying to convert it to a low impedance output on R2.
It's impossible. No any passive network can convert high impedance input to low impedance without loosing signal power a much.
I thought that the proper way of dealing with it would be to have an op amp buffer placed in-between R1 and R2
Exactly. And may be even without R1 & R2