About Group Delay of the LNA
时间：04-08
整理：3721RD
点击：
Hi all,
I designed an LNA in 2.2-4.4GHz band and calculated the group delay using:
Group Delay = -[d S21(p)]/[360*d f], where S21(p) is the phase of the S21 and f is the frequency. The plot is attached herewith. I found that, my grpup delay varies in between .045-.065ns in the bandwidth. Now, if my center frequency is 3.8GHz, then the period is .2632ns and the groupd delay variation is (.065-.045)/.2632 or 7.6 % .
Can anyone please write me how the group delay is graded? I mean, though lower the variation of group delay the better it is and It would be best if the delay variation were 0 %, but probably that is impractical. So, my question is- what is the satisfactory range? Is 7.6% too high or too low? Is it characterised in % as I did or there is any other metric?
I guess, the acceptable range would be different for the analog and digital applications. The analog system would be more sensitive. Someone please put some light on this topic.
Thank You ...
I designed an LNA in 2.2-4.4GHz band and calculated the group delay using:
Group Delay = -[d S21(p)]/[360*d f], where S21(p) is the phase of the S21 and f is the frequency. The plot is attached herewith. I found that, my grpup delay varies in between .045-.065ns in the bandwidth. Now, if my center frequency is 3.8GHz, then the period is .2632ns and the groupd delay variation is (.065-.045)/.2632 or 7.6 % .
Can anyone please write me how the group delay is graded? I mean, though lower the variation of group delay the better it is and It would be best if the delay variation were 0 %, but probably that is impractical. So, my question is- what is the satisfactory range? Is 7.6% too high or too low? Is it characterised in % as I did or there is any other metric?
I guess, the acceptable range would be different for the analog and digital applications. The analog system would be more sensitive. Someone please put some light on this topic.
Thank You ...