PXI-5421 fails to drive 50 ohm load

[ebalci]

Hi,

We have a PXI-5421 signal generator.

When I connect the output to a benchtop Agilent DSO-6034 oscilloscope I see a distorted the waveform if the output impedence of the signal generator and input impedence of the scope are both set to 50 ohm. I am using the NI-SCOPE soft panel to generate a 1MHz, 1Vp-p square wave and I use the Edit>Device Configuration menu option to set output and load impedence attributes of the 5421.

The waveforms are attached as PNG files:

print_00.png : 5421 settings are 50 ohm output impedence and 50 ohm load impedence, but the scope input is at 1Meg,  so the signal amplitude is doubled as expected.

print_01.png : 5421 settings are 50 ohm output impedence and 50 ohm load impedence, and scope input is set to 50 ohm. 5421 cannot drive the load. The edges of the square wave can be seen, but the signal drops to zero as if there is an over-current condition or so.

print_02.png : 5421 settings are 50 ohm output impedence and 1Meg load impedence, and the scope input is also set to 1Meg. The soft panel setting and the measured values match as expected.

What can you comment on this issue?

Is my signal PXI-5421 damaged?

Message Edited by ebalci on 08-16-2007 05:15 PM

[EBL]

Hi, ebalci.

The waveform has become AC-coupled somehow. Looking the plot you supplied, I would judge the time constant to be about 16 ns (corresponding to about a 10 MHz -3 dB frequency). If the source and load are both 50 ohms, that indicates that the size of the AC-coupling capacitance is around 160 pF.

The user guide for the scope indicates that AC-coupling is not allowed in 50-ohm input impedance mode. Further, the 5421 is always DC-coupled at the output. Hence the principal possibilities are:

1) The 5421 is defective.

2) The 6034 is defective.

3) The cable is defective.

I can suggest a few possibilities to determine the source of the problem.

a) If you have another substitute for any of items 1, 2, or 3, try that in its stead. For instance, if you have another 1 MHz square-wave generator, connect it to the scope and see what the waveform looks like. (Of course, the scientific method dictates that you should change only one of these elements at a time.)

b) If you have no substitutes, try changing the amplitude of the square-wave (at 1 MHz, or any frequency where you see the distortion) to see if the waveform improves. The idea is to switch the gain-setting attenuators on the board. If that has an effect, the problem is probably with the 5421.

Hope this helps,
Ed L.