Now the questions at the other end of the probe.
The more I read, the more I get confused. I have huge gap in my knowledge. I don't get this termination business...
What I'm missing?
My scope has normal input stage: 1MΩ± 2%, the input capacity is 18pF± 3pF) and connector is normal BNC.
I know that the "standard" connection is 50Ω cables and terminals, tees and all that stuff that goes in to BNC.
I know that scope probes use special cabe and the whole shebang is not that standards 50Ω stuff after all, specaially when it gets into higher frequency.
I'm thinkking of constructing something like this:http://www.dgkelectronics.com/high-voltage-differential-probe/
(I would be perfectly fine with lower frequency/rise time stuff and really would like something that has lead trough and not SMD)
Most of the stuff I understand to to certain degree there. He choose not to use FET input stage to raise the input impedance and that fits me too.
But the output stage beats me....this 50Ω business again.
Another thing is that do I need the BNC connector there? If I just want coax out, could I terminate the cable right on the PCB and how much art is there? I imagine that the core is soldered and shield is clamped over the copper a little further back.
Could I make one quick and dirty probe and use pigtail + BNC from the old 1:1 or 1:10 probe? Or I'm better of with RG-174 + BNC?
If I'm making the some one-time fixtures, how much active probe amplifier output stage, cable, termination and such changes when -3dB measurement system frequency is at 200 KHz, 1MHz, 10 MHZ or 100MHz?
I feel really dumb at this point. I know classic small signal/small frequency electronics and I know that when frequency grows, new players will show up.
I used to read something about transmission lines, but I haven't learn rule of thumb when one approach is chosen and then next and when something will do and when not on this measurement system. Must have fallen asleep when that was being hammered in our brains.