- This topic has 1 reply, 2 voices, and was last updated 3 years, 10 months ago by
.
- Posts
I am generating a 700MHz CW signal with a VSG60A at -20. When comparing this signal between the Spike software, and an Agilent E4407, the E4407 shows peak power at 700MHz, the Spike software shows the peak power at 699.997331MHz. Why is there such a difference? And is this difference in the BB60C or the Spike software? The BB60C has been on and running for several days.
Sweep settings: Center = 700MHz, span = 2MHz, Ref level = 0, Auto gain, Auto Atten, Auto preamp, RBW = 10KHz, VBW = 10KHz, Video units = power, Detector = average.mcline,
The biggest source of error is going to be from the RBW. With a 10kHz RBW, the frequency readout on a marker in Spike is going to be +/- several kHz, which is what you see. Lowering the RBW will improve the frequency accuracy of the marker.
Another source of error is going to be from the frequency difference between the two devices. If they do not share a reference, they may be some fraction of a ppm offset from each other. At 700M center frequency, this could be on the order of a 100Hz.
You could use a high resolution frequency measurement in the software such as the frequency difference meter utility. This will give you a much higher resolution than the marker approach.
For a given RBW, the points in the sweep might only be spaced RBW/4 Hz apart. For a 10khz RBW, that might mean 2.5kHz resolution on the points. Similar limitations apply to the E4407 as well, but the frequency of the point the peak fell on might have simply been closer to the true peak of the VSG60. There are certain configurations with the BB60C on which this might happen as well.
I hope this helps. Let me know if you have additional questions.
Regards,
Andrew
You must be logged in to reply to this topic.