VCO operation

Thread Starter

gkini19

Joined Jun 15, 2021
17
Hi All,

I was looking at several papers of radar transceiver that operates at 77GHz to 88 GHz focusing on the VCO and Chirp PLL architecture.

So if we want the output of the VCO to be 77GHz to 88 GHz, all the papers for radar transceivers use VCO with a multiplier to generate frequencies in the range of 77GHz to 88 GHz. (Say 20 * 4 or 38*2 )

What would be the technical reason to use multipliers? why can't we just have an architecture of the VCO that operates at 77 GHz?

Please let me know
 

Papabravo

Joined Feb 24, 2006
21,157
It is either a mater of cost, or the lack of a fundamental source at that frequency. It is often cheaper to use a multiplier and a fundamental source rather than to try to create a fundamental source.
 

Thread Starter

gkini19

Joined Jun 15, 2021
17
It is either a mater of cost, or the lack of a fundamental source at that frequency. It is often cheaper to use a multiplier and a fundamental source rather than to try to create a fundamental source.
Thanks for the reply.

I wanted a comparison with the technical specifications such as Phase Noise, matching, and all such stuff. COuld you please let me know
 

Papabravo

Joined Feb 24, 2006
21,157
You asked a fairly simple question and I gave you the primary reason for using them, A forum post or two will hardly provide the deeper answers you seek. I have not studied this presentation in detail but it appears to have a solid overview of some problems and solutions. The references at the end may also be useful

NTTI-Paris-Multipliers-Jul-2007.ppt (obspm.fr)
 

Wolframore

Joined Jan 21, 2019
2,609
I agree with @Papabravo, design and costs gets higher as we go up in frequency. Things get more difficult to make as we need faster slew rates and all that stuff. Makes sense to multiply it near the end.
 

Thread Starter

gkini19

Joined Jun 15, 2021
17
I agree with @Papabravo, design and costs gets higher as we go up in frequency. Things get more difficult to make as we need faster slew rates and all that stuff. Makes sense to multiply it near the end.
You mentioned things will get more difficult. Could you please tell me what are those? It would be great if you tell me.
 

Papabravo

Joined Feb 24, 2006
21,157
You mentioned things will get more difficult. Could you please tell me what are those? It would be great if you tell me.
Some difficulties you will encounter include, but are not limited to:
  1. Inability to use standard lumped components, with the possible exception of very small SMT resistors
  2. The need for CAD tools that can compute results based on microstriplines and other controlled impedances
  3. Obtaining suitable power levels for your requirements
  4. Costs of specialized components and fabrication techniques, like building structures on a silicon chip instead of a circuit board.
  5. Cost of test equipment and instrumentation. A typical VNA might still fetch over $100K
  6. Inability to find people with the requisite skills and knowledge
 

Thread Starter

gkini19

Joined Jun 15, 2021
17
Some difficulties you will encounter include, but are not limited to:
  1. Inability to use standard lumped components, with the possible exception of very small SMT resistors
  2. The need for CAD tools that can compute results based on microstriplines and other controlled impedances
  3. Obtaining suitable power levels for your requirements
  4. Costs of specialized components and fabrication techniques, like building structures on a silicon chip instead of a circuit board.
  5. Cost of test equipment and instrumentation. A typical VNA might still fetch over $100K
  6. Inability to find people with the requisite skills and knowledge
Thanks, what about the technical aspects like phase noise, Noise figures, and such stuff. Doesn't it degrade when we use multipliers?
 
Top