unexpected square wave noise

[KevinLilian]

Hi, I just run a simple test: Connecting PXI4461 with the PXI 4498 through a BNC cable. Using the LabVIEW provided examples: Voltage-continuous output.vi for PXI4461 and Voltage-continuous input.vi for PXI4498. I want PXI4461 to output a square wave, however, on the detection side, I found that there are unacceptable peaks on the rising edge and falling edge. There are also some oscillating signals during the stable stage. I am just wondering how that happens and is there any method to resolve that?

Thank you
Kevin

[johnsold · Accepted solution]

Kevin,

FRom the NI449x Specifications: 

Part of what you are seeing is the effect of the AC coupling. The -3 dB cutoff frequency is 0.5 Hz. In effect the device sees the horizontal segments of the input waveform as a DC signal and the coupling capacitor begins to charge. This accounts for the slopes of the acquired signal during constant portions of the input.

The ringing and overshoot are due to mismatched impedances. The output impedance of the 4461 is 22 ohms while the input impedance of the 4498 is 10 Mohms||35 pF. Mismatched signals cause reflections on the transmission line (the BNC cable).

The ringing can be eliminated by having matched impedances at the source and load. However, the minimum acceptable load of the 4461 is 600 ohms which is much higher than the 4498 output impedance.

A network like this will match impedances at both ends of the cable but will reduce the signal by about 1/2.  Your real load may not find the increased source impedance acceptable.

Because the 4498 does not have a DC coupling option, there is really nothing you can do about the sloping signals.

Lynn

[KevinLilian]

Hi, Johnsold,

Thank you very much for your reply.

Kevin