I have used the part LTC5583IUF#PBF for Dual RMS power detector for the operating frequency of 4.4GHz to 5GHz operation.
I have used Test Circuit Differential Input Configuration (page no. 16 in Data sheet) frequency 5800MHz for my applications.
RF Input Power level is -20dBm
I got the VOA voltage level 0.9V @ 4.4GHz and 1.01V @ 5GHz
VOB voltage level 1.21V @ 4.4GHz and 1.1V @ 5GHz.
Why these 110mV variation over the frequency range and also not getting same variation on both outputs?
Are you sure the circuitry on each channel is nearly identical?
It's possible that 4.4 and 5 GHz is too far removed from the center frequency of the balun, which would cause a large differential amplitude and phase error driving the LTC5583 inputs. That would explain why the 5GHz voltage readings are closer together than the 4.4 GHz readings.
Both channels were not identical but we simulated, got the insertion loss <0.2dB (4.4GHz to 5GHz) and amplitude imbalance of differential line is <0.1dB and return loss is <14dB. The PCB layout file and sch file are attached.
Sent in separate email.
I probed at input of 0.3pF capacitor and each arm is having 5 to 6dB return loss over the frequency range. How to match over the frequency range?
You might try slight variation of the 0.3pF capacitor value. Also, a new PCB layout with shorter balanced length PCB traces to the IC would help. See proposed layout, above.
Which balun manufacturer and part number is on your board?
I'll ask our apps engr for this particular device to comment. Watch for a reply from andy_m.
I used balun TDK_HHM1733B. I will go with new design But before that I need to do 8 sets with existing layout. I changed the capacitor 0.3pF but even though I didnt reduce the variation. Could you suggest some other ?
the variation over frequency may be due to the balun and matching. it's at the edge of the frequency range for your application. consider another balun that is more center, hhm1583 or hhm1595. below is the list of other balun you can pick from. it's 2:1 ratio so it maybe necessary to rematch it.
the layout does not look symmetrical for the two input channels, as stated earlier. the de-couple caps need to be close to the balun, on the same layer to ground. you may need to add another de-coupling cap at the LTC5583 IC. pin 6and pin 1 may help.