Please critique this SA usage

[Chris Wilson]

Someone who lives very close to me has said he is seeing a strong spurii at 137.400kHz on my 137.550 LF WSPR signal. I changed the amps power supply in case it was that, as it was an unregulated supply. It's now running off a pair of series connected 48V HP computer server supplies. Bear in mind this amp gives a good 1kW out, but antenna efficiency at such a low frequency is pitiful. I looked at a full power transmission via a small active RX antenna connected directly to my SA. Am I right in thinking given the RX antenna is within 15 feet of the TX antenna my signal at 137.400kHz is well down and should not be a problem? I am not very good with an SA... Thanks.

[dmills]

You need to turn the input attenuator up until the carrier is not shooting off the top of the screen.

Then look at the number of divisions between the carrier peak and the spur, the analyser is in 10dB per division mode so you can easily calculate the difference.

It is worth looking the same distance the other side of the carrier as well as AM spurs (Typical of power supply noise in a class E amp) will tend to show up as both sidebands.

You may well be seeing a lot of mixer spurs internally generated by the analyser given that you have the input power being well excessive, these will likely drop anway far faster then you would expect as you attenuate (It is one of the ways you know they are mixer intermod).

73 Dan M0HCN.

[ConKbot]

Yep, any power coming in over-driving you main mixer will result in a bunch of spurs, even if it's off screen. Most HP analysers I've seen like the mixer to run at  -10 dBm, so set your attenuation to bring your input signal below -10 and try again.

We've had engineers come in "guys the transmitter has a ton of noise all over the GPS band!"   when they were feeding the SA 15dBm from a coupler but had the SA at the minimum 6dB attenuation when they were looking at the GPS band.  A filter to kill the carrier convinced them that it was just poor measurement technique.

[tggzzz]

Yep, any power coming in over-driving you main mixer will result in a bunch of spurs, even if it's off screen. Most HP analysers I've seen like the mixer to run at  -10 dBm, so set your attenuation to bring your input signal below -10 and try again.

To make it clear to a beginner, that is -10dBm at the mixer's input, not the instrument's input.

We've had engineers come in "guys the transmitter has a ton of noise all over the GPS band!"   when they were feeding the SA 15dBm from a coupler but had the SA at the minimum 6dB attenuation when they were looking at the GPS band.  A filter to kill the carrier convinced them that it was just poor measurement technique.

As with any measurement, but especially with RF instruments, it is worthwhile playing around with the controls to see what happens. If you expect what you see, great; if not then also great - because you will learn something.

The most excitign sound in science isn't "Eureka!", but "That's odd..."