LMX2594: Harmonics of output

Thomas Fabricius

Part Number: LMX2594

Dear TI

PLLatinium Sim, provides fine information on spurs relating to the phase detector frequency. But what can be expected about (uneven)-harmonics of the output frequency ? If used as LO In a wideband receiver these are of substantial importance.

/tfabric

over 6 years ago

0 Dean Banerjee over 6 years ago

TI__Mastermind 28021 points

Thomas,

PLLatinum sim does not predict harmonics and we have some data in the datasheet.

In addition to this, I have this document I have written to address harmonics.

Regards,
Dean

LMX2594 Harmonics.pdf

0 Thomas Fabricius over 6 years ago in reply to Dean Banerjee

Prodigy 100 points

Hi Dean,
Thanks for the response. Can you maybe explain what you have measured, is it 4 different frequencies with two different measurements methods slow and fast ? I would expect harmonics of the output frequency at say 8Ghz to be n x 8GHz, how can I see these ? Or is it spurs not related to output frequency but merely the phase detector frequency ?!?!?
/tfabric

0 Thomas Fabricius over 6 years ago in reply to Thomas Fabricius

Prodigy 100 points

I am very interested in harmonics (and spurs, which I can predict via PLLatinium). If you could show a spectrum from 2GHz to 26GHz at for instance a frequency synthesis of Fout=3.2GHz (Fvco=12.8GHz, Fpd=200MHz), this will show harmonics at 6.4GHz, 12.8GHz, 19.2GHz, 22.4GHz (and spurs). This is interesting because the first harmonics 6.4GHz and 12.8GHz lies within the operational band of up to 15GHz.

0 Shawn Han over 6 years ago in reply to Thomas Fabricius

TI__Genius 9920 points

in your system, an external RF filter could help reduce harmonics.

0 Thomas Fabricius over 6 years ago in reply to Shawn Han

Prodigy 100 points

Yes band selectivity will help. But in order to calculate how much out of band attenuation to design for, we need to know the level of all spurs including harmonics. Another alternative is to filter the LO. But the versatility of a wide tuning range will influence the choices.

0 Noel Fung over 6 years ago in reply to Thomas Fabricius

TI__Guru** 110680 points

Hi Thomas,

in the graph, we were measuring some devices with extreme behaviors, we call them Fast and Slow devices. In other words, we were trying to find out the worst case harmonic behaviors.
2,3,..,5 mean 2nd, 3rd and 5th harmonics.
Since our test equipment can only go up to 26.5GHz, the higher harmonics of some VCO frequencies are not measurable.
For example, the purple traces represent the measurement of 5th harmonics, the highest VCO frequency we can set is 5GHz.

0 Thomas Fabricius over 6 years ago in reply to Noel Fung

Prodigy 100 points

Thanks I understand! This was all I needed...

Clock & timing

Clock & timing forum

LMX2594: Harmonics of output