I'm using a USB-6001 NI Card to send an AC signal to a 20x amplifier then across a home built capacitor device (sample). It's for a research project to measure a change in electric field for my PhD. I want to output up to 120 volts potentially (post amplifier) and then have the voltage drop across the sample measured by the same NI card. The input of this card only goes to +/- 10v and so I've been trying to use an in series passive attenuator with the circuit. The attenuator I have currently is 20dB / 50 Ohm / 4Ghz / 2W. The problem I'm having is that running the circuit with the attenuator in place causes large clipping at about 1V Vp input, much lower than the 10V originally, when running through an oscilloscope instead of the NI card I have a larger range, I didn't find the limit of the oscilloscope through the attenuator. This makes me believe it's something to do with the NI card. Any input would be appreciated.
1. The attenuator expect a 50 ohm load, but the input impedance of the 6001 is much higher, so you measure the clipping at 10V 😉
(Two 100 Ohm SMD forming a 50 Ohm load at the input migth help, or a 50 Ohm feedthrough or a BNC T with a 50 Ohm termination)
Or 2. maybe the termination itself limit the output... A lot of RF gear is sensitiv to high voltage , so that feature is your bug 😄
Or 3. the input impedance of the attenuator also is designed for 50 Ohm , but your source isn't ...
or all of them 🙂 (I vote for 3)
So make a schematic with all the impedances (reading the specs and/or measure by yourself) and do some math 😉
I would go with a 100:1 Scope Probe ( usually 10M Ohm input impedance with some pF ), match the 6001 input to 1M Ohm and do a probe calibration ! (also for the capacitive part of your impedances! )
Thanks a lot for your reply. haha no my Phd is not in EE, as you might have figured my undergrad work (Physics/Nanotech majors) only covered very brief classes on circuitry. My PhD is based on nanodiamonds and using them as biological sensors. I was suspicious of the impedance and thought it might be the cause of my issues but I have a lot of reading to do in order to understand fully what's going on. Thankfully today I found that by using an old attenuator, I could apply an offset and avoid the negative sine clipping and also avoid the positive (positive has a larger range before clipping for some reason), meaning I had about 10v in the positive to work with post attenuator input. This means I have a 100V drop (amp is actually 10x not 20x) across the capacitor and a massive electric field per nanometer (to apply across the diamonds). For now, I will stick with what I've got but take your advice to do some reading and better understand what's going on, in future I may need a higher voltage. I was also about to ask how I can match a large G-ohm impedance of the NI card but I think you've partly answered that in your last line, am I correct in assuming that as I increase the impedance to magnitudes approaching that of the NI card I should have a larger voltage range before I see clipping? OR is it more valuable to match to output impedance (only 0.2ohm) of the NI card with an attenuator of low input impedance?
