Multifunction DAQ
What is correct wiring for oscilloscope function with DAQCard-6024E and SCB-68 hardware?
Hello Rovalo!
The main difference I see is that your 6024E is an I/O module sampling up to 200kS/s compared to a scope which is often in the range of several to many MHz. So at least a factor 100 difference. On the other hand the 6024E has 16 AI channels (and many more AO, DI, and DO). Therefore don't expect the bandwidth or performance of a scope inside an I/O card!
Regarding your probes: Your PWM signal is a square wave with low/high changes at less than 1kHz. Don't know what the smallest change in your signal's timing is, maybe you need 100kHz sampling rate in the end?
Please consider that your 6024E's input is not a scope input, you are not connecting (potentially compensated) probes to it, instead just a bare metal wire. Therefore the voltage and current actually reaching your input is close to what the tip of your wire "sees". As there is e.g. no 9MOhm resistor in the tip of your "probe", there is no longish charge time of the 100pF capacitor inside the "scope's" input.
Have you tried connecting your sender? What signal do you see? Have you measured a know signal as comparison?
Yes, the Field Wiring Guide is amazing!
Ok, so we are moving in a different direction now. Bias resistors can indeed make a huge difference in many cases. The signals you are measuring, are they coming from the RC transmitter PCB or are you using an actual RF sender and receiver nearby? If so, differential connections, shielding etc. seem indeed very important to me.
Regarding figure 17 of the guide: Please note that the signal source shown there is grounded, therefore it makes sense to have the shielding connected to its GND. If I understand you correctly, your source is floating. Connecting the shielding to it might work as an antenna. It's impossible to predict if the shielding on the same ground plane or the "antenna" part has a bigger influence. Just test the signals you get when connecting the shielding to the device or an external GND (e.g. on the DAQ module). One might look better. Keep in mind that moving your devised around could change the whole set-up, so you might re-test this setup in the future.
Suggestions for further improvements are more than welcome, thank you!
I still dare to write this: In case you want to test if additional shielding is needed, kitchen aluminium foil is good temporary makeshift shielding.
Hello Rovalo,
My impression is that neither the wire thickness (is generally small) nor twisted/non-twisted would make a big difference.
If you have really long wires then twisted pairs on a differential voltage measurement help preventing noise from induction in the resulting measurement. As your wires don't transport any reasonable current (diameter) or high-frequency signal (surface area -> litz wire), it can be as small as you want in your case. The capacity of your cable grows with the volume of metal. However, I assume that your PCB can source enough current to overcome the little capacity of the wire. Especially as we are talking about a 1kHz signal, not 100 MHz.
Shielding is always good. If twisted pair with shielded pairs or single shielded wires with a different shielding gives better results is something only a test can show.
Edit: You mentioned UTP. Sounds like you have Ethernet cables. Please note that there are many different names around but two cables with the same name don't necessarily have the same shielding inside. If you e.g. have two SF/FTP cables, have a look inside to check which one is better...
