The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> RF Design >> Up conversion mixer ...
https://designers-guide.org/forum/YaBB.pl?num=1204755484

Message started by sanredrose on Mar 5^th, 2008, 2:18pm

Title: Up conversion mixer ...
Post by sanredrose on Mar 5^th, 2008, 2:18pm

I am designing a Gilbert Cell upconversion mixer whose specifications are as follows :

f(LO) = 4.5 GHz
f(IF) = 3.5 GHz

I am interested in pulling out the sum output of 8 GHz. While performing measurements we observed anomaly in the spectrum at 4.5 Ghz. This was due the fact that there was a 5 degree in mismatch in the Oscillator source.

Any ideas on reducing the LO leakage ???

Saravanan T K

Title: Re: Up conversion mixer ...
Post by ACWWong on Mar 13^th, 2008, 3:59am

LO feedthru can be due to dc offset in your mixer. check your layout for systematic offsets and run some monte-carlo simulations. Another major source will be crosstalk, be it in the silicon or in the package. Good layout planning (i.e having tx output bodnwires perpendicular to the lo carrying (the LO itself or LO supply) bondwires should help

If you have access to dc bias points, triming dc offset can reduce the leakage to crosstalk induced leakage.

Also when you say 5 degree mismatch, what do you mean ? duty cycle ? this can be improved by improving oscillator design and layout. it usually higher level harmonics or crosstalk non-symmetry which impacts fundamental duty cycle.

cheers
aw

Title: Re: Up conversion mixer ...
Post by Saravanan T K on Mar 17^th, 2008, 7:39am

I am trying to simulate this using Cadence - Spectre RF. I set the offset in the psin applied to the Local Oscillator inputs. At the output i observed the spectrum by selecting a voltage net and performing dft and db20 conversion.

Do you need any more details about offset let me know .. I will catch a snapshot and put that on the forum ...

Title: Re: Up conversion mixer ...
Post by khouly on Mar 24^th, 2008, 7:06am

u can reject this LO feedthrough by high pass filter at the output of the Mixer , other way u can use subharmonic mixer , so the LO will be half the 4.5 Ghz and u use the 2nd harmonic

khouly