65 nm Process and VCO - VCO suffer from high phase noise

EDAboard.com Forum Index -> Analog IC Design & Layout -> 65 nm Process and VCO - VCO suffer from high phase noise

Author

Message

AdvaRes

Joined: 14 Feb 2008
Posts: 1038
Helped: 46
Location: At home

14 Apr 2009 17:40 65 nm Process and VCO

Hi all,

Does 65nm technology permits to have VCOs with low phase noise (below -100dBc at 1MHz) ?
In fact my VCO suffer from high phase noise (-65 dBc) and I couldn't improve it.
Is that a limit or I could have made mistakes in transistors dimentionning ?

Please help.

Thanks.

Last edited by AdvaRes on 16 Apr 2009 11:28; edited 1 time in total

psmon

Joined: 04 Jun 2007
Posts: 173
Helped: 22

14 Apr 2009 18:46 65 nm Process and VCO

What is your oscillation frequency?

AdvaRes

Joined: 14 Feb 2008
Posts: 1038
Helped: 46
Location: At home

15 Apr 2009 9:06 Re: 65 nm Process and VCO

psmon wrote:

What is your oscillation frequency?

2GHz.

playoff

Joined: 24 Aug 2007
Posts: 7

16 Apr 2009 10:12 Re: 65 nm Process and VCO

Whats your offset frequency?
-100dbc(at)10khz is not difficult.

AdvaRes

Joined: 14 Feb 2008
Posts: 1038
Helped: 46
Location: At home

16 Apr 2009 10:59 Re: 65 nm Process and VCO

No it is -65 dBc @1MHz offset from 2GHz.

Google
AdSense
Google Adsense

16 Apr 2009 10:59 Ads

psmon

Joined: 04 Jun 2007
Posts: 173
Helped: 22

16 Apr 2009 22:57 65 nm Process and VCO

You are using ring oscillator? LC VCO cannot be so bad.

AdvaRes

Joined: 14 Feb 2008
Posts: 1038
Helped: 46
Location: At home

17 Apr 2009 9:55 Re: 65 nm Process and VCO

psmon wrote:

You are using ring oscillator? LC VCO cannot be so bad.

The VCO is a Ring of 4 differential delay cells.

saro_k_82

Joined: 17 May 2007
Posts: 266
Helped: 55
Location: India

18 Apr 2009 5:02 Re: 65 nm Process and VCO

Your phase noise looks poor. Phase noise directly trades with power. How much is your power spend? The minimum length transistors are noisy. Increasing the lengths by 20nm or so normally shows huge difference in phase noise

fanatic

Joined: 10 Dec 2008
Posts: 10

18 Apr 2009 6:22 Re: 65 nm Process and VCO

AdvaRes wrote:

psmon wrote:

You are using ring oscillator? LC VCO cannot be so bad.

The VCO is a Ring of 4 differential delay cells.

what kind of ring cell do you use ?
and what is your process?
ah, pn below -100dbc/hz(at)1M is not difficult for 65 process
more power and select a good structure, and optimum pn

AdvaRes

Joined: 14 Feb 2008
Posts: 1038
Helped: 46
Location: At home

18 Apr 2009 15:31 Re: 65 nm Process and VCO

saro_k_82 wrote:

Increasing the lengths by 20nm or so normally shows huge difference in phase noise

Thanks saro_k_82,

Do you mean a 20um ? Coz 20 nm Channel length is not permitted by the tool. The minimum channel length permitted is 65nm.

I noted that when channel length increase the phase noise decrease considerably. However it has to be very far from the 65nm. With 180nm process VCO's transistor length is 200 nm for good phase noise.