Expert Advice Responses Page (April)

Published Tuesday, April 15, 2008 | Last Update 4.15.08
Here are your replies to the April Expert Advice column

Dan Nehring - VP of Engineering, Valpey Fisher Corporation. Oscillators are like cars. Some are powerful, sleek (expensive) performers, some are just economy clunkers (clockers), and most are in between.

Comments: Hi, In this particular editorial author keeps his view on different kinds of noises and disturbances that is affecting the circuit and system performances. Along the line, author keeps focus on high speed digital noises and High frequency noises disturbing the performance of LO and other components. Definitely, in oscillator Phase noise trade off is the biggest challenge to deal with, where OCXO stability aspect also important. Thus, PLL multiplied solution provides some advantages over noises, in contrary in digital design jitter plays major role. So, focus on jitter and noise vectors and SSB phase noises in this editorial issue provide some good concept. Therefore, Valpey fisher ref oscillator and frequency control module can show some resistances against noises, it may lead to good noise trade off design. finally, well focuses on the specification and their usages in the microwave systems and components articulated good ideas from the article. In brief, this article provides good amount of interest on noise types and it's behavior in a microwave systems. Thank you. swapnadip

swapnadip ghosh, SASKEN finland
4/21/2008 6:31:16 AM


Response from the Author: Hi Swapnadip, Thank you for reading my article and for your comments. As you summarized, my intent had been to give a general outlook on the sources of noise and how to avoid them in choosing an oscillator as well as interfacing to it. I wanted to do this without a bunch of math or computer analysis, and to focus on how to really think about your circuit and system in practical hands-on scenarios. I believe that if you can first have a good conceptual/practical feel for the issues, then you will design in the right stuff. So, this was the approach I took and hopefully I will have stimulated some thought. I am glad that you found it of interest. Thanks, Dan

Comments: Hi, I am interested in vibration which increasing significantly crystal oscillator phase noise. You described in article that when system profile specifications is available, these can be designed out or avoided. What we can do when we don't have this kind of information? How to design oscillator circuits when it works in environment when vibration is changing (different acceleration sensitivity, different direction of vibration vectors)? Marcin

Marcin Rybakowski, Nokia Siemens Networks Poland
5/7/2008 8:41:23 AM


Response from the Author: Hi Marcin, Thank you for your interest in this topic. The fact is every oscillator will have an inherent vibration sensitivity that you are pretty much going to have to live with. Some are much better than others. Some are specially designed for best performance. In your case, you are designing in an environment where you do not have much advance knowledge of the vibration you will encounter. If that is the case, you will likely have to assume some worst case assumption for your vibration. From the other end you will have to determine the allowable phase noise levels that you can tolerate under the vibration conditions and present this as your specification goals. This could result in overkill. If you are doing the oscillator design yourself, you need to remember that a big issue is that the oscillator mechanical structure itself can magnify the external vibrational effects. So, the internal oscillator mechanical design must be rigid and not have any mechanical resonances within the range of vibration frequencies. Your goal should be for the phase noise degradation observed when the part is vibrated to be no more that that predicted from a vibrational sensitivity calculated from a 2 G static tip over test. - Regards, Dan