Buying 0~10MHz Spectrum analyzer, solutions and advice

[Frex]

Hello,

Thanks for all comments.

1audio ,
i really don't understand why i will see images below fmax/2 ?
The LO sweep between min and max input frequency (100Hz to 10MHz) ,and "zero IF" at output mixer occur only when Finput is equal to LO frequency.
But may be you say that when n-th harmonics of LO are mixed in the analyzer passband, it can create 0 IF ?
I understand that. This is why generally IF is higher than Fmax/ to push LO harmonics always beyond the the frequency range of the analyzer.
Perhaps it will be better to take a 10MHz IF, using quartz bandpass filtering, with LO sweeping then between 10 to 20MHz...
I'm not very familiar with RF design, but if think to evaluate different solutions.

jackinnj,
I had take many time to search interesting info to do a SA in 0-10Mhz range, and of course i had see many about SDR, but i'd find nothing really interesting.
If you know some link which you think valuable for my design i would be very pleased.

simmconn,
I already have a HP8568B (100Hz1.5GHz SA), and i don't think the 3585 will work much better.
The idea is to learn, in same time to do a valuable instruments which help me.


Frex.

 

[1audio]

Harmonics of both the LO and the input signals will all mix. A lot depends on the desired spurious free dynamic range etc. The RF stuff gets quite involved (like audio).

 

[EUVL]

Frex,

I presume you know about this :

http://techdoc.kvindesland.no/radio/complex_measurements/20051028154529677.pdf

I wonder what you think of it ?


Patrick

 

[Frex]

Hello EUVL,

Your technical document is interesting and talking about build dual-conversion spectum analyzer. It seem to be very simple.
However, it use old diode-based mixer at it's input, and the dynamic performance appear very limited. !i don't think it could be a very good stuff for audio testing, but probably a good way to start in the spectrums analysing.
Thank you for the document !

Frex.

 

[1audio]

There is nothing wrong with diode mixers. They really are the best choice in high performance spectrum analyzers. But the needs for audio are different. You need input conditioning, attenuators etc. all of which will degrade the performance of the mixer. For audio you can use an active mixer but it won't necessarily be better in IM or dynamic range. And probably won't have the frequency range.

 

[EUVL]

Demain,

Could you explain what you mean by input conditioning explicitly ?

Also I found two more with much more detailed descriptions and easier to get components :

http://www.holmea.demon.co.uk/SpecAnHtml/SpecAn.htm

http://www.qsl.net/n9zia/wireless/pdf/9808035.pdf
http://www.qsl.net/n9zia/wireless/pdf/9809037.pdf


Patrick

 

[1audio]

Input conditioning-
Input attenuator
Impedance matching, buffer amp
Anti-image filter/bandpass filtering

All the stuff to make it more useful. On the microwave spectrum analyzers you go directly into the diode bridge unless you have a specific image/spurious signal issue. it give the best dynamic range.

 

[jackinnj]

Don't forget input protection -- the little beep which goes off when the analyzer starts to perspire under too high a voltage.

I think ruined diode mixers are what causes microwave engineers to lose their hair.

 

[EUVL]

Would you be so kind to post some example schematics ?

Thx,
Patrick