|
The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl Design >> Analog Design >> Question on a VCO design with differential stage https://designers-guide.org/forum/YaBB.pl?num=1128088045 Message started by tumeda on Sep 30th, 2005, 6:47am |
|
Title: Question on a VCO design with differential stage Post by tumeda on Sep 30th, 2005, 6:47am I have a problem about a VCO circuit design with differential stage architecture, its schematic is attached here. I am wondering about why the output signal cannot oscillate ??? I have also done a PZ analysis, and found that there are positive poles in practice. Is anything wrong with my design? Any feedback is greatly appreciated. I will try to attach the schematic here later. |
|
Title: Re: Question on a VCO design with differential sta Post by Paul on Oct 1st, 2005, 1:16pm Hi, Maybe you should give some more information about your simulation set-up. What are your start-up conditions? In simulation you have to use initial conditions or an external stimulus to trigger the oscillation. Search this forum for more information. Paul |
|
Title: Re: Question on a VCO design with differential sta Post by tumeda on Oct 2nd, 2005, 1:27pm -Hi Paul Thanks! This is the schematic of VCO and the PZ anaylsis result. The underleft is a voltage input, where 1V is applied. In a transient simulation, two initial conditions are set on the two nets, 1.5 V for one and 0V for anohter, as showed in the picture. could you tell what the wrong is? 
|
|
Title: Re: Question on a VCO design with differential sta Post by ywguo on Oct 3rd, 2005, 8:55pm Tumeda, The image is not here. :) Yawei |
|
Title: Re: Question on a VCO design with differential sta Post by tumeda on Oct 4th, 2005, 1:06am I have updated these pictures. Sometimes, You need still click right mouse to select the "show picture", then you can see its! Finally. I have found a free space to save picture in internet. Thanks! |
|
Title: Re: Question on a VCO design with differential sta Post by Andrew Beckett on Oct 4th, 2005, 9:46pm Tumeda, there's still no picture. When you use the img tags, you have to give a URL between the open and close tag, where the URL is a location containing the picture. The image also needs to stay at that URL otherwise people won't be able to see it. Regards, Andrew. |
|
Title: Re: Question on a VCO design with differential sta Post by Ken Kundert on Oct 4th, 2005, 9:51pm I also recommend that you use the "Preview" button to assure everything on is correct before you post. -Ken |
|
Title: Re: Question on a VCO design with differential sta Post by Andrew Beckett on Oct 4th, 2005, 10:01pm Ken, Good point. I know that when I first did this, I made a mistake, and so had to have a second attempt at posting the picture. What was worse, I meant to hit the Preview button and my mouse slipped :-/ Regards, Andrew. |
|
Title: Re: Question on a VCO design with differential sta Post by tumeda on Oct 5th, 2005, 1:43am Sorry for my mistakes! I have really previewed these pictures before I posted them. But the URL address of these pictures seem to be not constant, they vary always. If you open this page, please try to click right mouse to select the "show picture", I hope you can see them. So far, I donot have a better idea of it. The direct URL address of my pictures page are given. You can direcly click the under URL to see it. Sorry again for putting you in trouble! http://cn.pg.photos.yahoo.com/ph/notting_zj/detail?.dir=1b00&.dnm=d692.jpg&.src=ph |
|
Title: Re: Question on a VCO design with differential sta Post by Forum Administrator on Oct 5th, 2005, 8:15am I apologize for the trouble you are having, but the situation will be improving soon. Sometime in the next few weeks we will be upgrading the Forum to the latest version, which supports the uploading of images and files. -Ken |
|
Title: Re: Question on a VCO design with differential sta Post by ywguo on Oct 5th, 2005, 6:39pm Hi, Tumeda, Have you tried not to set the initial voltage on those two nodes? Yawei |
|
Title: Re: Question on a VCO design with differential sta Post by tumeda on Oct 9th, 2005, 1:44pm As your suggestion, I made a simulation without any initial setup, the circuit can not oscillate yet. But I have tried to set the initial voltages at another levels, for instance, 1V and 0.5V for two nodes, the circuit can oscillate now. But, a new problem comes up. I have set the transistor design parameters (W/L) in another values, the circuit cannot oscillate again, while there are also poles in right side. Who can tell me, is there any critiera for VCO oscillation? |
|
Title: Re: Question on a VCO design with differential sta Post by ywguo on Oct 9th, 2005, 6:50pm Tumeda, Do you mean the simulator cannot analyze the oscillator again? Quote:
Would you please paste the errors and warnings on the forum? Best regards, Yawei |
|
Title: Re: Question on a VCO design with differential sta Post by Paul on Oct 10th, 2005, 12:56am Hi Tumeda, tumeda wrote on Oct 9th, 2005, 1:44pm:
There are the so-called Barkhausen criteria, but I'm sure you are aware of that. Frequently you need more gain in a ring oscillator to guarantee start-up than defined by the above criteria. You may have enough gain to obtain RHP poles, but not enough to start it up. When you change transistor ratios, are you sure your operating point is still correct? In your schematic, I'm also concerned that when you use active loads in saturation, you may need common-mode feed-back to guarantee a proper operating point. Frequently you use MOS transistors in triode region as resistive loads in such schemes, but it seems not to be your case. Paul |
|
Title: Re: Question on a VCO design with differential sta Post by tumeda on Oct 12th, 2005, 5:27am Hi ywguo Sorry for tap mistake! I wanted to mean that , the circuit cannot oscillate again. :P Hi Paul About Barkhausen criteria, as my personal opinion, it is not very suitful for VCO oscillator. Even we can use a AC simulation to find the gain and phase, the AC simulation depends on the DC points. The DC point of VCO oscillator varies always during the simulation. Which conditions can accutally trigger a VCO oscillation is still unclear for me? ??? About the PMOS as the load in differential VCO, Which region is best for VCO, triode or saturation region ? |
|
Title: Re: Question on a VCO design with differential sta Post by fehler on Oct 15th, 2005, 8:51pm Dear Tumeda, I am afraid that Barkhausen Criteria still holds for oscillator. Although osciallator has no stable o.p, it is instructive to check B.C. first. Please have a look at Razavi's paper. My question is, how about your gain? I have a look on you schematic, you have 4 stages, and one of them is cross- connected. And then it is sure that there will be a positive feedback, if the gain condition is met. |
|
Title: Re: Question on a VCO design with differential sta Post by Paul on Oct 17th, 2005, 12:18am Hi, as I already mentioned you should check the gain per stage. An oscillator (in the real world) starts up from noise, so you need enough gain... Paul |
|
The Designer's Guide Community Forum » Powered by YaBB 2.2.2! YaBB © 2000-2008. All Rights Reserved. |