I am designing simple 200MHz scope for use with FPGA board and came up to idea of software adjustable offset and gain of input preamplifier.
It is using TI ADC08200 and since I am not much experienced in hi frequency analog stuff I have only slightly altered their original preamplifier with help of LTSpice.
This is original design:
This is my altered version:
I am using quad DAC (I have two identical channels) and quad OPAMP (LMV324M) one of DAC/Opamp channels is making reference voltage for ADC making me able to adjust "gain" but this is limited to reference voltage ranges so it might be better to adjust it somehow on OPAMP, but since I didn't want to use digital pots for their high values (starting on 1kOhm as far as I know) and TI's datasheet clearly states that preamp should incorporate at least 2x or 3x gain which would be not possible with 1kOhm on inverting input (instead of 10+12 ohm).
The second channel is used for adjusting offset and I am not certain wether this is ideal way of doing it. As far as my LTSpice testing it seemed to work but I don't know if it works in real circuits. With one slight problem that it seems that it also adjusts gain a bit. I don't know if it's the cause of misadjusting gain since only difference is that I am not using pot to adjust offset but I adjust voltage direclty.
Next thing which bothers me is how to correctly make protection from overvoltage?
Now I am using pair of silicon diodes on preamp input and pair of diodes on ADC input with help of 5.6V zener diode to get 0.6V by going 5.6V above -5V rail which allows silicon diode to sink overvoltage below -100mV. I think it should work this way and LTSpice simulation seems to confirm it, but I think there is better and simpler way of doing it.
I though about using shottky diodes, but I think that since shottky diodes have high reverese leakage they introduce high distortion and when I have tested it in spice it was noticeable.