Physically testing mixers/frequency multipliers

[MobiNaz]

hello,

I want to test my frequency multiplier using network analyser and signal generator. How can I check the conversion loss?

Thanks for your co operation

 

[tony_lth]

Hi, you can test it with signal generator and spectrum analysis. 1st test signal generator output with SA, read it's power such as Pin(dBm), then test the desired harmonic power with SA, read it's power Pout(dBm), so the conversion loss is Pout-Pin.

 

[G0HZU]

Some network analysers have an offset function on them allowing the testing of frequency converters etc. I don't know if this works with multipliers but I assume it will.

Personally, I would use the method outlined by tony above. This is because you can also look at rejection of unwanted multiplier terms and this can help you define any filtering requirements etc.

 

[flyhigh]

Hi

Perhaps the only motivation to use signal generator and network analyser will be the possibility to measure output reflection coefficient in the real operating conditions. Not sure how this is important to you though.

flyhigh

 

[MobiNaz]

Thanks for replying guys.

I have got the spectrum analyser and signal generator, now when I tested my signal generator, for input power level from 30 to 10 dBm, Spectrum Analyser shows 1.67dBm. For 0, -5dBm, for -10 dBm -15dBm and so no.. Any idea what is wrong inside?

For power levels 0 to -30 dBm my output power at the frequency F1 (The frequency of input signal) I am getting 5 to 6 dBm difference in my simulated and actual results. However, I can not get any output for power level -10 to -30 dBm at 2x F1(comparable to noise) and for 10dBm and 0dBm input power the output at 2XF1 make no sense at all. What could be possibly wrong?

Very worried. Thanks for your help in advance.

 

[davenn]

I presume its a passive freq multiplier... ie. it is not powered, other than the RF going into it.
As such there will be large losses through the multiplier. for example, my 1.296GHz to 10.368GHz (x9) step recovery diode multiplier .... 1Watt (30dBm) in at 1296MHz gives ~ 15mW (~12dBm) out at 10.368GHz

5-6 dBm difference between sim and actual .... not really suprised.... simulated resulted NEVER
(well extremely rarely) = the real world

Dave
VK2TDN

 

[tony_lth]

. for example, my 1.296GHz to 10.368GHz (x9) step recovery diode multiplier ....
VK2TDN

Hi, davenn, that's X8, not X9.

 

[davenn]

Hi, davenn, that's X8, not X9.

Doh... thats what ya get when I answer messages late at nite hahhaha maths skills not fully functional

rest of the comments are still valid tho

D