- 积分
- 2857
- 在线时间
- 1596 小时
- 最后登录
- 2009-4-9
- 阅读权限
- 100
- 精华
- 6
- UID
- 29222
- 帖子
- 2737
- 精华
- 6
- 经验
- 2857 点
- 金钱
- 2785 ¥
- 注册时间
- 2007-10-21
|
楼主 |
发表于 2007-12-17 23:18
|
显示全部楼层
here is another simulation to show that how the ancillary amplifier (the current dumper) can correct distortion in the main amp.
it is basically the same as the one I used earlier. However, I added a little bit of distortion signal in the main amp (lt1022). V4 is a 2vp 20khz sine wave, unreleated to the input signal. You can view the new "main amp" as the lt1022 + v4.
obiviously, the output signal from the new "main amp" is highly distorted, either the amptitude error of 2/8 = 25%, and significant phase margin error (20khz vs. 21khz).
the green trace shows the opamp output, in this case, it is pre-distorted to undo the distortion from its own distortion mechanism (V4) and the ancillary amp.
the red trace shows the overall distortion from the new "main amp". it is essentailly the opamp + the distortion we added via V4.
the blue trace shows a clean output on the output.
you can change the type / magnetude of the distortion and the analysis would be the same.
the essence of the AES article is that you can correct the errors in the main amp, whatever it happens to be, by picking up the error signal and run it through an ancillary amp and summing them up at the output.
The main amp does NOT have to have high current capability: you can choose your ancillary amp and other components to accomodate that.
That is NOT to say that you do NOT need current capabilities in all cases.
Hope this helps. |
-
|