Part Number: LMX2594EVM
I am using LMX2594 EVM and Reference Pro module with TICSPRO software. While doing phase noise measurement of frequencies in the range 3.5GHz to 5.5GHz it is noticed that there is a reduction in value of 6dB from what is expected. I tried with the following settings ;
fOSC = 100MHz
FPD = 200MHz
N div = 40
fVCO = 8 GHz
Ch Div =2
RF out = 4 GHz
phase noise @ 10KHz offset from the carrier of 4GHz is -110dBc/Hz.
In LMX2594 datasheet page no:14 , it is given that the phase noise @10 KHz offset is -110.1dBc/Hz for fVCO of 8GHz. So the expected value of Phase noise at 10KHz offset is -116dBc/Hz for fVCO of 8GHz with channel divider of 2. But the measured value is -110dBc/Hz.
The phase noise measurement is done with R& S Spectrum Analyzer FSU 20Hz-8GHz.
Other observations are of the following;
fOSC= 100MHz : FPD= 200MHz : N Div =55 : RF out = 5.5 GHz : fVCO = 11 GHz : Ch Div = 2 : measured value of phase noise @ 10KHz offset = -107dBc/Hz but expected value is -113dBc/Hz.
fOSC= 100MHz : FPD= 200MHz : N Div =50 : RF out = 5 GHz : fVCO = 10 GHz : Ch Div = 2 : measured value of phase noise @ 10KHz offset = -109dBc/Hz but expected value is -115dBc/Hz.
fOSC= 100MHz : FPD= 200MHz : N Div =70 : RF out = 3.5 GHz : fVCO = 14 GHz : Ch Div = 4 : measured value of phase noise @ 10KHz offset = -111dBc/Hz but expected value is -117dBc/Hz.
Please note we are looking into this and will respond shortly.
For more information on TI Clocking and Timing tools and training please see below links:
We are glad that we were able to resolve this issue, and will now proceed to close this thread.
If you have further questions related to this thread, you may click "Ask a related question" below. The newly created question will be automatically linked to this question.
In reply to Aaron Black:
The data we took in the datasheet was using a very clean reference clock from Wenzel. The 10kHz offset phase noise is 10dB better than the Reference Pro.
In the datasheet Figure 9, we have a plot at 3.5GHz, the 10kHz phase noise is -117dBc/Hz, which matches very well with the 6dB improvement when the frequency gets divided by 2.
In reply to Noel Fung:
Please share the Part number of the Wenzel oscillator.
The phase noise requirement of Reference Oscillator given in the "LMX2594 EVM Instructions" application note is -159dBc/Hz at 10kHz offset. This maybe arrived at the worst case requirement. In our application, the maximum frequency is 5.5GHz with divide by 2 operation at 11GHz VCO. The 11GHz synthesized phase noise at 10kHz offset is -107dBc/Hz. Reference oscillator phase noise requirement will be -107+20log(55)+10=-152dBc/Hz, for 0.4dB degradation. We tested using CVPD-922 oscillator, which is claiming a phase noise of -154dBc/Hz at 10kHz offset. But still we are not achieving the phase noise.
The LMK61 Ref Pro oscillator is having only -142dBc/Hz at 10kHz offset.
In reply to Vishnu R:
The Wenzel clock is 501-4623G, unfortunately, it is discontinued. Here is the phase noise.
3487.1157.100 MHz Wenzel 501-4623G.txt
The best phase noise XO from Crystek I have ever seen is CVSS-945, it is a sine wave VCXO.
The Wenzel oscillator is having a phase noise of -160dBc/Hz at 10kHz offset.
But based on the calculation the required phase noise is -152dBc/Hz ( -107+20log(55)+10=-152dBc/Hz) for the 11GHz VCO setting.
I've given a 10dB margin considering a degradation of 0.4dB. But still the phase noise is off the requirement.
I did a simulation, if I use the datasheet phase noise specification of CVPD922, then the synthesizer output phase noise (at 5.5GHz) will be identical with a Wenzel reference clock.
Did you measure your CVPD922 XO phase noise, isn't it same as the datasheet claimed?
All content and materials on this site are provided "as is". TI and its respective suppliers and providers of content make no representations about the suitability of these materials for any purpose and disclaim all warranties and conditions with regard to these materials, including but not limited to all implied warranties and conditions of merchantability, fitness for a particular purpose, title and non-infringement of any third party intellectual property right. No license, either express or implied, by estoppel or otherwise, is granted by TI. Use of the information on this site may require a license from a third party, or a license from TI.
TI is a global semiconductor design and manufacturing company. Innovate with 100,000+ analog ICs andembedded processors, along with software, tools and the industry’s largest sales/support staff.