I want to make a measurement preamp. I want to use it for distortion measurements. I also would like see and measure the ripple from regulated supplies on my scope or pc.
I plan to use my Isotope preamp pcb for this purpose. I am ok with doing some hacking of the pcb for alterations. If it doesnt work out, I will try something else, this is just the first cut. Preamp and soundcard measurements are in the pcb thread.
I will have a lot of questions about how to keep the noise low, the bandwidth, best opamp for this, how to test etc. Any recommendations for the opamp?
I will try to keep the resistor values low to keep the noise low. Layout is also quite tight, will use small twisted pairs for wiring. What other precautions do I need to take to avoid noise and oscillations etc.
I will need some coupling caps in between the stages or the dc will saturate the output. Each channel has an rf filter at the input. I think a cap parallel to the feedback resistor to limit the bandwidth might be a better way to handle HF stuff. Does it make sense to have both or should I take out the rf filters.
I read a bit on the net about the notch filters. I checked the pcb layout and I can easily fit Fliege notch in one channel of the pcb.
If I need to use the twin-t notch filter, then its more difficult to fit it on the pcb, I will proably need to take the filter off the pcb or make the whole thing on a general purpose pcb, or maybe even a quick and dirty home etched pcb. I probably will be able to use some trimpots to fine tune too. So far it looks like this is a better choice due to deeper notch.
Is matched np0/c0g smd caps ok for this. I can use polyester too if its a must, just harder to use with an smd board.
I think I should not be too worried about tuning the filter to 1khz, if its at 1005hz, I will make the generator output a 1005hz frequency. There shouldnt be much of a difference in measurements. I should probably focus on getting a deep notch and low impact on harmonics. Whats the Q that I should be aiming for.
I simmed one in ltspice. The Q and notch depth seem to be at cross purposes, so it seems like a tradeoff. Is there a sweet spot for this. Whats an acceptable loss on the second harmonic. I should be able to measure it in arta and subtract from the 2nd harmonic numbers of DUT. In sims, with a 55db 1khz notch, I loose about 1db on 2khz; on 64db its 4.3db.
If I use my pc soundcard (asus essence st) for measurements, on its own it does some 110-115db, so I guess I can't rely on it much for anything better than 100db. If I am not wrong, the gain should be the same as the notch filter's notch depth, give or take a few db.
I also looked at the Bainters notch filter, math looked more complicated, anyone tried it?
I plan to use my Isotope preamp pcb for this purpose. I am ok with doing some hacking of the pcb for alterations. If it doesnt work out, I will try something else, this is just the first cut. Preamp and soundcard measurements are in the pcb thread.
I will have a lot of questions about how to keep the noise low, the bandwidth, best opamp for this, how to test etc. Any recommendations for the opamp?
I will try to keep the resistor values low to keep the noise low. Layout is also quite tight, will use small twisted pairs for wiring. What other precautions do I need to take to avoid noise and oscillations etc.
I will need some coupling caps in between the stages or the dc will saturate the output. Each channel has an rf filter at the input. I think a cap parallel to the feedback resistor to limit the bandwidth might be a better way to handle HF stuff. Does it make sense to have both or should I take out the rf filters.
I read a bit on the net about the notch filters. I checked the pcb layout and I can easily fit Fliege notch in one channel of the pcb.
If I need to use the twin-t notch filter, then its more difficult to fit it on the pcb, I will proably need to take the filter off the pcb or make the whole thing on a general purpose pcb, or maybe even a quick and dirty home etched pcb. I probably will be able to use some trimpots to fine tune too. So far it looks like this is a better choice due to deeper notch.
Is matched np0/c0g smd caps ok for this. I can use polyester too if its a must, just harder to use with an smd board.
I think I should not be too worried about tuning the filter to 1khz, if its at 1005hz, I will make the generator output a 1005hz frequency. There shouldnt be much of a difference in measurements. I should probably focus on getting a deep notch and low impact on harmonics. Whats the Q that I should be aiming for.
I simmed one in ltspice. The Q and notch depth seem to be at cross purposes, so it seems like a tradeoff. Is there a sweet spot for this. Whats an acceptable loss on the second harmonic. I should be able to measure it in arta and subtract from the 2nd harmonic numbers of DUT. In sims, with a 55db 1khz notch, I loose about 1db on 2khz; on 64db its 4.3db.
If I use my pc soundcard (asus essence st) for measurements, on its own it does some 110-115db, so I guess I can't rely on it much for anything better than 100db. If I am not wrong, the gain should be the same as the notch filter's notch depth, give or take a few db.
I also looked at the Bainters notch filter, math looked more complicated, anyone tried it?