The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> RF Design >> How does VCO affect the reference supr?
https://designers-guide.org/forum/YaBB.pl?num=1105960216

Message started by allen on Jan 17^th, 2005, 3:10am

Title: How does VCO affect the reference supr?
Post by allen on Jan 17^th, 2005, 3:10am

Hi,

I'm designing a frequency synthesizer. When an ideal VCO, which is described in veriloga, is placed in the loop, the simulated reference spur is below -90dBc. However, when a VCO circuit with nearly the same gain is used, the reference spur is only about -52dBc. Can anybody tell me what has happened? How does the VCO affect the reference spur?

Thanks

Allen

???

Title: Re: How does VCO affect the reference supr?
Post by boa on Jan 18^th, 2005, 6:34pm

In which frequencies do you have spurs? Are they multiples of VCO freq?

Also it would be good to learn more about your PLL, its spec and how you simulate it.

Title: Re: How does VCO affect the reference supr?
Post by allen on Jan 19^th, 2005, 5:07am

Thanks boa :)
The frequency of the VCO is 2.4GHz, and the reference frequency is 5MHz. The spur is at 5MHz, 10MHz ... offset from the carry frequency. The VCO's gain is about 150MHz/V, and the open loop bandwidth is 100KHz. The spur is to be designed below -70dBc.
To simulate the spur, I first run a transition simulation with spectre, then use DFT function in the calculator to get the specturm of the VCO's output. Is it right to simulate spur in this way?

Allen

Title: Re: How does VCO affect the reference supr?
Post by boa on Jan 20^th, 2005, 8:55pm

The spurs should appear because of reference freq coupling through PFD/CP to VCO control voltage. Another thing is that your filter should be able to suppress them.

Actually, I do not understand the reason why PLL's output has these spurs when you simulate with a real VCO (I suppose it has a proper phase noise characteristics) if it works fine with an ideal Verilog-A VCO..

Title: Re: How does VCO affect the reference supr?
Post by maserati on Jan 26^th, 2005, 4:26pm

Based on my experience, the DFT function from the calculater is not accurate. But it should NOT have > 10dB difference...
try another way to sim the spur and compare

Title: Re: How does VCO affect the reference supr?
Post by allen on Jan 27^th, 2005, 4:50am

Thanks boa and maserati.

The reference spur decreases to -77dBc after I shorten the maximum step in the simulation option. Therefor, I think the degeneration of spur may be caused by simulation error. However, the simulation time is extremly long when the simulation step is short, and I still don't know the simulation result is the correct one( it may continue to decrease if the step gets even shorter). I wonder if there are some more effecent ways to simulation reference spur.

Allen.

Title: Re: How does VCO affect the reference supr?
Post by rf-design on Feb 1^st, 2005, 11:38pm

If there is no coupling via substrate or supply to the VCO impacting the spur performance the model approach is acurate enough. Only the VCO voltage ripple impact the spur then.