Voltage-to-Current Converter

wwm101 · 发表于 2009-6-25 16:04:26

马上注册，结交更多好友，享用更多功能，让你轻松玩转社区。

您需要登录才可以下载或查看，没有账号？注册

×

In the paper 'A Wideband CMOS Variable Gain Amplifier With an Exponential Gain Control'，a Voltage-to-Current Converter is realized and the structure is shown in the following figure。

In the figure，VC** denotes the voltage copier and current generation。here i don't understand why VC** can copy voltage from node A to node B。

Could anybody explain it？ thank you。

[ 本帖最后由 wwm101 于 2009-6-25 16:16 编辑 ]

wind2000sp3 · 发表于 2009-6-25 16:27:57

hi, wwm101

I suspect that the gate of the other NMOS of the differential pair should be connected to node "B". So the negative feedback loop will be established, and all things will make sense. Please think about it, thanx.

wwm101 · 发表于 2009-6-25 20:55:25

原帖由 wind2000sp3 于 2009-6-25 16:27 发表

 登录/注册后可看大图

hi, wwm101

I suspect that the gate of the other NMOS of the differential pair should be connected to node "B". So the negative feedback loop will be established, and all things will make sense. Ple ...

Thank you for your reply.
But how to explain the function of the NMOS whose gate is connected to node"A" and source is connected to the gate of differential pair of the other NOMS at the right side? thank you.

wind2000sp3 · 发表于 2009-6-26 09:34:50

wwm101 & waveguides:
Sorry to Mr. Waveguides, you are not totally right.
This NMOS is a Slew-Rate-enhancement element. In the case of node A is suddenly higher than node B by Vthn, this NMOS will be turned on very soon. At this transient, the gates of the current mirrors are pulled down, and node B is increased as a consequence.
From this analysis, we can find out that the differential mode input voltage of the differential pair will never be larger than Vthn, which makes the Opamp works in linear region all the time. Also, from this point of view, the NMOS' channel length will be the minimum length allowed in a given process, and its width will be as large as 100um, smaller or bigger, depending on the slew rate requirement.

wwm101 · 发表于 2009-6-26 16:14:45

wind2000sp3 &waveguides:
Thank you very much for your reply.
Here and now ,i can understand it.

wind2000sp3 · 发表于 2009-6-26 16:32:53

wwm101,
Thank you too, for your interesting topics. There are so few topics discussing Linear Analog techniques in this forum, instead, almost all the members attach there reference in their topics. This job should be completed by a junior clerk, not an IC engineer.

wwm101 · 发表于 2009-6-27 19:18:14

原帖由 waveguides 于 2009-6-27 09:06 发表

 登录/注册后可看大图

I still think there are a few problems with this cir. Actually quite a few. I'm not sure this would be a working cir or not. hehe

It can work when i connect the gate of the other NOMS of differential pair at the right side to the node 'B' .
The simulation result was shown in the following.
But the result also shows that the NMOS doesn't enhance slew-rate significantly.

schematic

tran simulation result