Spectrum analyser

[Banus_radio]

Hello all

I've purchased a sa5006 spectrum analyser from China which is good enough for what I want.
I'm learning about rf and am beginner
It has a tracking generator which outputs the frequency at 50ohm, I feed this into a band pass filter and connect output of filter into analyser allowing me to tune it up looking at attunation.

But is there anything I can buy in replace of the tracking generator that will output noise across the whole band so I can connect and tune a low pass filter and see where it drops off on the display? Instead of just looking at a single frequency

Thanks in advance

[Banus_radio]

Actually ignore my post, I think I've just confused myself lol. It already is producing noise. I'm a plonka

[thewisepranker]

A tracking generator shouldn't produce one frequency or noise - it should track the frequency at each sampling point across the span you have assigned, at the same level you set across the whole span, which is exactly what I think you are looking for.
The result should be a straight, horizontal line across the screen for most of the spectrum that your analyser is capable of looking at if you connect the tracking output to the RF input with a good piece of coax. I'm not sure how good this will be on an Atten (or similar) but that's the idea.

In other words, if you set the centre frequency to 100 MHz and the span to 20 MHz, the tracking generator output will be 90 MHz when the cathode ray tube dot is at the far left of the screen. Imagine you can slow down the scan of the CRT so that you can watch it scan left to right. The frequency will gradually increase proportionally as the dot slowly scans to the right, until it reaches 110 MHz (the far right of the screen) and will then snap back to 90 MHz the instant the CRT dot returns to the left hand side of the screen. Sort of.

[Banus_radio]

you are correct

I confused myself earlier as ive only had the unit several hours.

Its an awesome bit of kit. If I set it up to a larger span (say 100mhz wide) then its not a horizontal line, its several curves that look like half circles approx 20mhz wide each when attenuation is on, ifI tun attenuation off completly then i get a straight line about half way up. This will be great for looking a filter cut off's etc.

Thanks for your help

[thewisepranker]

Interesting.

[thewisepranker]

This has been bugging me a bit. Please can you post a picture of the semicircles you describe, with the rest of the analyser front panel in the picture?

[Banus_radio]

Yeah sure, gimme an hour and ill take a photo.

[Banus_radio]

Hers some pics to show you what i mean. It only does the semi circles if full atten is on (40db on output and 40db on input)

If I only have 40db on input and use full output I get a straight'ish line

[Banus_radio]

For some reason, I select the picture files and click submit but they dont show grrrrr.

I click add files, add my jpeg photos then click submit above it, nothing appears... HELP?

[thewisepranker]

Upload them to Tinypic?

[Banus_radio]

Nice one, done that

[thewisepranker]

Since it gets worse at higher frequency I would suggest that it's a problem with the stepped attenuator switches being not properly matched.

[Banus_radio]

I connected a lowpass filter, 3 lots of 4 turns coils at 8mm internal diamater with 33pf either side of each. (total of 3 coils and 6 caps) Is that 6pole?

I connected the tracking output to it. Do these look ok or is cutoff a bit high? Im trying my best to learn

[Banus_radio]

Since it gets worse at higher frequency I would suggest that it's a problem with the stepped attenuator switches being not properly matched.

Thanks for the info. I wont try to fix it as I have no clue what im doing lol..

[Banus_radio]

Next i connected to an old G**G band1 receiver which I had to remove from one of my boxes years ago as the link was to weak.My link back then was on 61mhz.

This bandpass filter tells a different story?? saying its around 55mhz.. defo nowhere near 61mhz


its two lots of 6 turn coils with a 2 turn tap on the output,47pf caps to ground..

Ive just retuned using the analyser to 61mhz its now working fine. I know this is at 50ohms and my antenna is prob 75ohm, not sure if that makes a difference?

[Banus_radio]

BUMP

Can anyone confirm the low pass filter looks correct?

[Albert H]

The lowpass does look right. You need to experiment with different coil layouts, screening between coils, height of the coils above the groundplane, and so on.

My usual Band II LPF is three four turn coils at 8mm diameter, with 39pF caps to ground - and importantly, an 18pF capacitor in parallel with the second coil, to resonate at 121MHz - you really don't want anything emitted around there!

I also remove the groundplane beneath the coils (and stand them off by about 5mm) to keep their "Q" up and prevent losses. The capacitors are important too - if you have any kind of mismatch, the voltages developed across those capacitors can be huge, so the capacitors have to be suitably rated.

[Analyser]

I connected a lowpass filter, 3 lots of 4 turns coils at 8mm internal diamater with 33pf either side of each. (total of 3 coils and 6 caps) Is that 6pole?

I connected the tracking output to it. Do these look ok or is cutoff a bit high? Im trying my best to learn

No, it looks fine. You need some space above the top of the band (108MHz) before the filter starts to roll off. Depending on the filter type the filter may start to roll off early or some ripple will appear just before it cuts off, and if that's too close to 108MHz it could affect your amplifier.
In your case the filter starts to roll off at about 120MHz and you can see some ripple (a small knee shape) just before this. If the filter was tuned too low the ripple would appear at, say, 105MHz and possibly cause you problems.

Another thing you can see with your tracking generator is how much the filter attenuates at the second harmonic frequencies. For 88MHz which is usually the worst case, this is 176MHz. See how many dB down this frequency is compared to 88MHz and that's how much the filter should attenuate the harmonics generated by your amplifier.

[Banus_radio]

Excellant, I appreciate the replies. I will have an experiment and update my findings.