Could you please help me with how to calculate the VGS of this current mirror? Thank you very much!

zzw_rst

Could you please help me with how to calculate the VGS of this current mirror? Thank you very much! [Copy link]

 

maychang

Is there such a current mirror? Can MOS tubes form a current mirror?

Never seen before.

zzw_rst

maychang posted on 2020-10-9 09:46 Is there such a current mirror? Can MOS tubes form a current mirror? Unprecedented.

This circuit can work. For example, I want to design ID=5.5mA. How do I conceive it?

maychang

zzw_rst published on 2020-10-9 12:09 This circuit can work. For example, I want to design ID=5.5mA. How do I conceive it?

This circuit ensures that the drain currents of the two MOS tubes are equal.

To make the current of the right tube I, the drain resistance of the left tube should be 5V minus the MOS tube turn-on voltage (strictly speaking, it should be minus the drain current divided by the transconductance) divided by the required drain current I.

maychang

zzw_rst posted on 2020-10-9 12:09 This circuit can work. For example, I want to design ID=5.5mA, how do I conceive it.

Of course, the premise is that the turn-on voltage and transconductance and other parameters of the left and right MOS tubes are completely equal.

maychang

zzw_rst posted on 2020-10-9 12:09 This circuit can work. For example, I want to design ID=5.5mA, how do I conceive it.

The title is "How to calculate the VGS of this current mirror?" VGS should be the MOS tube turn-on voltage plus the drain current divided by the transconductance.

zzw_rst

maychang posted on 2020-10-9 12:43 The title is "How to obtain the VGS of this current mirror?" VGS should be the MOS tube turn-on voltage plus the drain current...

The current and voltage of the circuit components should be able to be calculated before actual measurement (of course, the premise is to know other parameters besides current and voltage), so ID is unknown. If ID is known, VGS is directly = 5V-ID*R

maychang

zzw_rst posted on 2020-10-9 14:02 The current and voltage of circuit components should be able to be calculated before actual measurement (of course, the premise is to know other parameters besides current and voltage), so ID ...

What you said on the third floor was "For example, if I want to design ID=5.5mA, how do I come up with the idea?" Then the ID should be known.

zzw_rst

maychang posted on 2020-10-9 15:20 What you said on the third floor was "For example, if I want to design ID=5.5mA, how do I conceive it." Then the ID should be known.

I made this assumption, which made it complicated. I don't know the ID.

maychang

zzw_rst posted on 2020-10-9 17:33 This is my assumption, it's complicated. I don't know the ID

Parameters such as the MOS tube turn-on voltage and transconductance are determined by the manufacturing process. Once the tube is manufactured, these parameters are determined and cannot be modified from the outside. They can only be measured using certain methods.

gmchen

I don't quite understand the meaning of the original poster's question.

The current mirror composed of MOS tube usually works in the saturation region, and its current expression is:

Where mu is the carrier mobility, C is the capacitance per unit area of the gate, W/L is the width-to-length ratio of the gate, VGS is the gate-source voltage, and VTH is the threshold voltage. If the above parameters of the field effect tube are known, and VGS=5V-I*R is known from the figure, I should be able to be obtained

zzw_rst

gmchen posted on 2020-10-10 14:00 I don't quite understand what the original poster meant. The current mirror composed of MOS tubes usually works in the saturation region, and its current expression is: Where mu is...

Just manually calculate VGS and ID

zzw_rst

gmchen posted on 2020-10-10 14:00 I don't quite understand what the original poster meant. The current mirror composed of MOS tubes usually works in the saturation region, and its current expression is: Where mu is...

How do you calculate the VGS of the above formula?

gmchen

zzw_rst posted on 2020-10-10 17:50 How do you calculate the VGS of the formula above?

I mean if we know the parameters of the MOS tube, we can get the I

Specifically, the electron mobility, gate capacitance, and gate width-to-length ratio are known when designing the MOS tube. After the MOS tube is made, its turn-on voltage is also known. At this time, only the two parameters ID and VGS remain in the two equations, and the equations can be solved.

zzw_rst

gmchen published on 2020-10-10 19:19 I mean if you know the parameters of the MOS tube, you can get I. Specifically, the electron mobility, gate capacitance, gate width-to-length ratio, and then when designing the MOS tube...

Happy National Day, teacher! I have a question that has stuck for many days. It is not a question about this floor, but a question about frequency response. Since I cannot send pictures in private messages, I ask you here. I hope to get your advice. The transfer function derived from the following circuit is different from the standard form of zeros and poles (of course, printing errors are not ruled out), but two time constants τs and τp are also obtained. I want to ask about the process and reason of this derivation. The book does not give it, but just says that the accurate transfer function is difficult to derive. If I encounter more complex and more capacitors, I don’t know what form of transfer function to convert it into.

gmchen

zzw_rst posted on 2021-10-4 22:53 Happy National Day, teacher! I have a question that has been stuck for many days. It is not a problem with this floor, but a problem with frequency response. Because I cannot send pictures in private messages, ...

I just got home from the National Day holiday, so I’m late in replying.

The transfer function of the above circuit can be written directly based on the circuit:

H(s)=[Rp||(1/sCp)] / [Rs+(1/sCs)+(Rp||(1/sCp))]

That is, the voltage division of the series connection part of the resistor and capacitor and the parallel connection part of the resistor and capacitor.

Simplifying the above equation will give you the result you see.

If we further simplify the complex fraction, we can get a function form that is consistent with the standard form.

zzw_rst

gmchen published on 2021-10-29 07:41 I just got home from the National Day holiday, so I am late in replying. The transfer function of the above circuit can be directly written according to the circuit: H(s)=[Rp||(1/sCp)] / [ ...

Thanks for your reply, I basically understand it, but now I have forgotten a lot, haha. What I want to ask is how the author knows how to convert it into this form with time constant. Now I roughly know why the author lists it this way, because the above form can be obtained by multiplying the two extreme standard denominators. The author does this as an empirical method.