MOSFET - Current Mirror Explained

Summary

This educational video from ALL ABOUT ELECTRONICS dives into current mirrors, crucial for ICs, using MOSFETs. It explains the significance of stable biasing voltage for amplifiers, the challenges of maintaining it in battery-operated devices, and temperature effects. The video illustrates how current mirrors generate stable reference currents, adjusting them as multiples or fractions via the W/L ratio. It also touches on the channel length modulation effect and its minimization, and concludes with applications in biasing amplifiers and the cascode current mirror for improved gain and stability.

Takeaways

• 🔬 Current mirrors are used in integrated circuits as stable current sources for biasing MOS amplifiers.
• 💡 The advantage of using a constant current source for biasing is high voltage gain and improved biasing stability.
• 🔋 In battery-operated devices, current mirrors are crucial as supply voltage decreases over time, affecting the bias current.
• 🌡️ The biasing current is temperature-dependent, and current mirrors help maintain stability despite temperature changes.
• 🔄 Current mirrors can generate copies of a reference current source, which can be the same or a multiple/fraction of the reference.
• 🚀 The design of a current mirror using MOSFET involves controlling the gate-to-source voltage to change the drain current.
• 📉 The drain current ID in a MOSFET is given by the expression involving the device parameter k, Vgs, and Vt.
• 🔗 The current mirror circuit uses a diode-connected transistor to generate a Vgs proportional to the reference current.
• 🔍 The channel length modulation effect is considered in the discussion, which affects the drain current with changes in Vds.
• 🔗 The W/L ratio of MOSFETs determines the relationship between the reference current and the drain current in the mirror.
• 🛠️ The Early voltage (Va) is introduced to quantify the change in drain current due to changes in Vds, considering the channel length modulation.

Q & A

What is the purpose of a current mirror circuit in electronics?

A current mirror circuit is used to generate a stable current source that can be a copy or multiple copies of a reference current source, ensuring stable biasing for amplifiers in integrated circuits.

Why is it important to bias amplifiers with a constant current source?

Biasing amplifiers with a constant current source provides high voltage gain and improves biasing stability, which is crucial for maintaining consistent performance in battery-operated devices where supply voltage can reduce over time.

How does a current mirror circuit help in ICs with multiple amplifiers?

In ICs containing hundreds or thousands of amplifiers operating at different points, a current mirror circuit helps generate a stable biasing voltage for each amplifier, overcoming the challenge of generating a stable voltage for each.

What is the role of temperature in the operation of a current mirror circuit?

Temperature affects the drain current of a current mirror circuit because the mobility (μn) and threshold voltage (Vt) are temperature-dependent. As a result, the biasing current can change with temperature.

How does a MOSFET operate as a voltage-controlled device in a current mirror circuit?

A MOSFET operates in the saturation region where the drain current (ID) is controlled by the gate-to-source voltage (Vgs). By controlling Vgs, the drain current can be adjusted to match the reference current.

What is the significance of the W/L ratio in MOSFET current mirror circuits?

The W/L (width-to-length) ratio of a MOSFET determines the drain current it can carry. By adjusting the W/L ratio, the current mirror circuit can generate currents that are multiples or fractions of the reference current.

How does the channel length modulation affect the drain current in a MOSFET?

Channel length modulation causes the drain current to change with the change in voltage Vds. This effect can be minimized by increasing the length of the MOSFETs, thus reducing the impact on the drain current.

What is the Early voltage (Va) and how is it related to channel length modulation?

The Early voltage (Va) is the reciprocal of the channel length modulation coefficient (λ). It represents the change in drain current due to a change in Vds, and it helps to minimize the effect of channel length modulation by increasing the length of the MOSFETs.

How can a pMOS current mirror be used in an IC?

A pMOS current mirror can be used similarly to an nMOS current mirror to generate a stable current source. It ensures that the source and drain voltages (Vsg) are equal, and the drain current (Id1) is determined by the W/L ratio and the reference current.

What is the advantage of using a cascode current mirror as an active load in an amplifier?

Using a cascode current mirror as an active load in an amplifier improves both voltage gain and biasing stability. It provides a higher output impedance and transconductance, enhancing the amplifier's performance.

Outlines

00:00

🔬 Introduction to Current Mirrors with MOSFETs

This paragraph introduces the concept of current mirrors in electronic circuits, specifically focusing on their implementation using MOSFETs. It discusses the importance of current mirrors in providing a stable current source for biasing amplifiers in integrated circuits (ICs), which enhances both voltage gain and biasing stability. The paragraph also addresses the challenges associated with maintaining a stable biasing voltage, especially in battery-operated devices where supply voltage can decrease over time. It highlights the need for a stable current source to overcome these issues and mentions the use of a diode-connected transistor to generate a voltage proportional to the reference current, ensuring that the MOSFET operates in saturation for stable current output.

05:06

🔍 Understanding the Design of Current Mirrors

This section delves deeper into the design principles of current mirrors using MOSFETs. It explains the relationship between the gate-to-source voltage (Vgs) and the drain current (ID) in a MOSFET, and how controlling Vgs allows for the manipulation of ID. The concept of generating a stable reference current and copying it to other MOSFETs is explored, with an emphasis on the importance of device parameters such as oxide capacitance and threshold voltage. The summary also covers the mathematical relationship between the drain current of the reference transistor and that of the MOSFET being used, including the effects of the W/L (width-to-length) ratio on the current output. Additionally, the paragraph introduces the impact of channel length modulation on the drain current and how it can be minimized by adjusting the MOSFET's dimensions.

10:10

🛠️ Exploring pMOS Current Mirrors and Multiple Current Generation

This paragraph extends the discussion to pMOS current mirrors, providing a comparative analysis with nMOS transistors. It describes the setup of a pMOS current mirror circuit and the conditions required for it to operate in saturation. The focus then shifts to generating multiple copies of a reference current using a single source, with examples given to illustrate how different W/L ratios can produce currents that are multiples or fractions of the reference current. The paragraph also touches on the practical applications of these current mirrors, such as biasing source followers and common source amplifiers, and suggests the use of cascode current mirrors for improved voltage gain and biasing stability.

15:12

📢 Closing Remarks and Invitation for Further Discussion

In the concluding paragraph, the presenter summarizes the key points of the video, reinforcing the concept of current mirrors and their design using MOSFETs. They invite viewers to engage with the content by asking questions or providing suggestions in the comment section and encourage them to like and subscribe for more educational content. This paragraph serves as a call to action, aiming to foster a community of learners interested in electronics.

Keywords

💡Current Mirror

A current mirror is an electronic circuit that copies a current from one part of the circuit to another while keeping the current constant. In the context of the video, current mirrors are used as a stable current source for biasing MOS amplifiers in integrated circuits. They are crucial for maintaining high voltage gain and improving biasing stability, as they provide a constant current regardless of changes in the biasing voltage.

💡MOSFET

MOSFET stands for Metal-Oxide-Semiconductor Field-Effect Transistor. It is a type of transistor used for amplifying or switching electronic signals. The video discusses how current mirrors can be designed using MOSFETs, emphasizing their role in creating stable current sources for biasing amplifiers in ICs.

💡Biasing

Biasing refers to the process of setting the operating point of a circuit or a transistor to a desired level. In the video, biasing is achieved using a current mirror circuit, which provides a stable current source to ensure that the MOS amplifier operates at a consistent point, enhancing performance and stability.

💡Voltage Gain

Voltage gain is the ratio of the output voltage to the input voltage in an amplifier. The video mentions that biasing an amplifier with a constant current source, like a current mirror, provides high voltage gain. This is because the stable current source helps maintain a consistent relationship between input and output voltages.

💡ICs (Integrated Circuits)

ICs are miniaturized electronic circuits built on a single semiconductor material. The video discusses the use of current mirrors in ICs to provide stable biasing for MOS amplifiers, highlighting the importance of current mirrors in maintaining the performance of thousands of amplifiers within an IC.

💡Diode Connected Transistor

A diode connected transistor is a MOSFET with its drain and gate terminals connected together, operating in the saturation region. In the video, it is used to generate a voltage proportional to the reference current, which is then used to control the drain current in a current mirror circuit.

💡Device Parameter

The device parameter refers to the characteristics of a MOSFET, such as its threshold voltage and transconductance. The video explains that for a current mirror to work correctly, the device parameters of the MOSFETs involved must be identical or carefully controlled to ensure that the drain current mirrors the reference current accurately.

💡W/L Ratio (Width-to-Length Ratio)

The W/L ratio is a key parameter in MOSFETs that affects the current drive capability. The video explains how changing the W/L ratio of a MOSFET can generate currents that are multiples or fractions of a reference current, allowing for flexible current mirror design.

💡Channel Length Modulation

Channel length modulation is a phenomenon where the drain current in a MOSFET changes slightly with changes in the drain-source voltage. The video discusses how this effect can be minimized by adjusting the length of the MOSFETs in a current mirror circuit to maintain stable current output.

💡Early Voltage

Early voltage is a parameter that characterizes the channel length modulation effect in a MOSFET. In the video, it is mentioned in the context of how changes in drain-source voltage can affect the drain current, and how this can be accounted for in the design of current mirror circuits.

💡Cascode Amplifier

A cascode amplifier is a type of amplifier circuit that uses a cascade of transistors to achieve higher gain and better input impedance. The video touches on the use of cascode amplifiers and the challenges of biasing stability, suggesting that current mirrors can be used to improve both gain and stability.

Highlights

Current mirrors are used as current sources in integrated circuits to provide high voltage gain and improve biasing stability.

Biasing voltage stability is crucial, especially in battery-operated devices where supply voltage reduces over time.

The drain current or biasing current of a current source is temperature-dependent.

A stable current source is generated in ICs using current mirror circuits to overcome biasing challenges.

The MOSFET is a voltage-controlled device used in current mirror circuits to control drain current.

The drain current ID in a MOSFET can be changed by controlling the gate-to-source voltage.

A diode-connected transistor can generate a voltage Vgs proportional to the reference current.

The drain current ID is equal to the reference current I(REF) when the MOSFET operates in saturation.

The relationship between the reference current and the drain current can be expressed using the MOSFET's device parameter k.

The current mirror circuit ensures that M1 MOSFET generates a drain current equal to the reference current.

The W/L ratio of MOSFETs determines the relationship between the drain current and the reference current.

The channel length modulation effect can cause changes in drain current with changes in voltage Vds.

The Early voltage (Va) is introduced to express the change in drain current due to changes in Vds.

The pMOS current mirror operates similarly to the nMOS current mirror but with source connected to Vdd.

Multiple copies of a reference current can be generated using a single reference current source in ICs.

The W/L ratio can be adjusted to generate currents that are multiples or fractions of the reference current.

Current mirrors can be used to bias various amplifiers in ICs, improving their performance.

The cascode current mirror can be used as an active load to improve both gain and biasing stability.

Transcripts

Hey friends, welcome to the YouTube channel ALL ABOUT ELECTRONICS.

So in this video we will learn about the current mirrors and we will see that how it can be

designed using the MOSFET.

So typically in the integrated circuits, the MOS amplifiers are biased using the current source.

And typically in the ICs, this current mirror circuit is used as a current source.

The advantage of biasing the amplifier with the constant current source is that it provides

the high voltage gain and it also improves the biasing stability.

In the earlier videos, we have seen that actually how this current source can be implemented

using a single PMOS transistor or using the cascode current source.

But in this type of circuit, this biasing voltage should be as stable as possible.

So if there is any change in this biasing voltage, then it will affect the bias current.

And it is particularly important in the battery-operated devices where with the time, the supply voltage reduces.

Well, there are ways to minimize the change in the biasing voltage and let's say somehow

we are able to keep the biasing voltage stable.

But if you see any IC, then it contains hundreds or thousands of such amplifiers.

And each amplifier is operating at a different operating point.

So to generate a stable biasing voltage for each amplifier is another challenge.

Moreover, this drain current or this biasing current of this current source is a function of temperature.

Because this μn and VT are the temperature dependent.

So because of that, with the temperature, this biasing current will change.

So to overcome all these problems, what is actually done is, a very stable current source

is generated in the IC and using the current mirror circuit, a copy or the number of copies

of this reference current source is generated.

So this generated copy could be same as the reference current source or it could be a

multiple or the fraction of this reference current source.

So now, let's see how this current mirror circuit is designed using the MOSFET.

So we know that the MOSFET is a voltage controlled device.

And in the saturation, the drain current ID can be given by this expression.

So by controlling this gate to source voltage, this drain current ID can be changed.

So if we want to copy this reference current in this MOSFET, then first of all, we need

to generate a voltage which is proportional to this reference current.

And we need to apply that voltage between the gate and the source terminal of this MOSFET.

So that, this drain current is a function of voltage Vgs and in a way, we get a relationship

between this reference current and this drain current.

So using the diode connected transistor, we can generate a voltage Vgs which is proportional

to this reference current.

So in this diode connected transistor, the drain and the gate terminals are connected together.

That means in this case, this voltage Vd is equal to Vg.

So here, this voltage Vds is always greater than or equal to Vgs - Vt.

That means this MOSFET always operates in the saturation region.

And therefore, this drain current ID can be given by this expression.

So let's say, this device parameter is equal to k.

That means drain current ID can be given as k times Vgs-Vt whole square.

And here, this drain current ID is equal to I(REF).

That means I(REF) is equal to k times (Vgs-Vt) whole square.

Or we can say that this voltage Vgs is equal to square root of I reference divided by k plus Vt.

So depending on this reference current, this voltage Vgs will change.

And the same voltage Vgs can be applied between the gate and the source terminal of this MOSFET.

So let's call this MOSFET as the reference MOSFET and this MOSFET as M1.

So if these two MOSFETs are perfectly matched, or in other words, if the device parameters

of these two MOSFETs are identical, then for the same voltage Vgs, this MOSFET M1 should

also generate the drain current which is equal to the reference current, right?

So here, the only thing which we need to ensure is that this M1 should remain always in the

saturation, so that it can be used as a current source.

That means this Vds1 should be greater than or equal to Vgs-Vt.

So in general, let's find the relationship between this drain current Id1 and this reference current.

So let's say for two MOSFETs, except the W by L ratio, all the device parameters are same.

That means the oxide capacitance and this threshold voltage are equal.

So for the given value of the voltage Vgs, the reference current or the drain current

of this reference transistor can be given as, I(REF), that is equal to half times

μn times Cox times W divided by L(ref) times, (Vgs-Vt)^2 , while for this

transistor M1, this drain current Id1 can be given as half times μn times Cox times

W divided by L1 times, (Vgs-Vt)^2.

So this W by L1 is the W by L ratio of this transistor M1, while this W by L(ref)

is the W by L ratio of this reference transistor.

So if we take the ratio of this Id1 and the I(REF), then we can say that this Id1

divided by I(REF) is equal to W by L1 divided by W by L(ref). Or we can say

that this Id1 is equal to I(REF) times this W by L1 divided by W by L(ref).

So if the W by L ratio of the two MOSFETs are same, then this Id1 is equal to I(REF).

And by changing this W by L ratio, we can generate the new current which is multiple

or the fraction of this reference current.

For example, if the W by L ratio of this M1 is 5 times the reference transistor, then

this Id1 will be equal to 5 times I(REF).

So in this way, we can generate a copy which is the fraction or the multiple of this reference current.

So, typically in the ICs, the length of each MOSFET is kept fixed and only width is varied.

And by changing this width, we can change the W by L ratio.

Now so far in our discussion, we have neglected the effect of the channel length modulation.

That means we have assumed that in the saturation region, even if the voltage Vds increases,

then also this drain current Id remains fixed.

But actually, the drain current does change with the change in the voltage Vds.

So if two MOSFETs are identical, then this drain current Id1 will be equal to I(REF).

And it will occur when this voltage Vds1 is equal to Vgs.

So considering the channel end modulation effect, it is true when this voltage Vds1 is equal to Vgs.

Because for this reference transistor, since the drain and the gate terminals are connected

together, so the voltage Vds is same as Vgs.

That means for this MOSFET M1, to generate a same reference current, the voltage Vds1

should also be equal to Vgs.

But if there is any change in this voltage Vds1, then the generated drain current will also change.

So if this voltage Vds1 changes by ΔVds1, then let's say the change in the drain current

will be equal to Id1 plus ΔId1.

And with this change in the drain current, the drain current can be written as I(REF)

times, 1 plus λ times ΔVds1.

But this λ is the channel length modulation coefficient of this M1.

So we can say that the change in the drain current is equal to I (REF) times this

λ times ΔVds1. Or we can say that it is equal to I(REF) times this ΔVds1 divided by Va.

Where this Va is equal to 1/λ.

And this Va is also known as the early voltage.

So this is the expression of the change in the drain current due to the change in the voltage Vds.

Now while writing this expression, we have assumed that the W by L ratio of these two

MOSFETs are identical.

But if the W by L ratio of these two MOSFETs are different, then let's see the general

expression of this drain current.

So in that case, this Id1 plus ΔId1 can be written as I(REF) times this W by L of

first transistor divided by W by L of this reference transistor times, 1 plus λ times ΔVds1.

So in this way, with the change in the voltage Vds, there will be also change in this drain current.

So this change can be minimized by increasing the length of these MOSFETs.

That means in the W by L ratio, if we increase the value of L, then we can minimize the effect

of the channel length modulation.

So on the second channel, we will cover a few examples based on it so that it will get more clear to you.

Alright, so so far, we have seen the current mirror circuit using the nMOS transistor.

Similarly, let's see the pMOS current mirror.

And for the time being, let's neglect the effect of the channel end modulation.

So here, this reference current is connected at the drain terminal of this reference transistor,

while the gate and the drain terminal of this reference transistor are connected together.

So basically, it is the diode connector transistor.

And if we notice over here, then the source of both transistors are connected to the Vdd.

That means here, voltage Vsg for both transistors are equal.

Now once again, this drain current Id1 can be given as W by L of this first transistor

divided by W by L of this reference transistor times I(REF).

So the only thing which we need to ensure over here is that this M1 should operate in

the saturation region.

That means here, this voltage Vsd1 should be greater than or equal to (Vsd-Vt).

Alright, so now let's see the circuit where using a single reference current source, multiple

copies of this current are generated.

So here as you can see, using a single reference current source, multiple copies are generated.

So here, this Id1 will be equal to W by L of this first transistor divided by W by L

of this reference transistor times I(REF).

And if we notice over here, then the same voltage Vgs is also applied to this transistor M2.

That means for this transistor M2, this Id2 will be equal to W by L ratio of this second

transistor divided by W by L ratio of this reference transistor times I(REF).

And the same current is also flowing through this transistor M3.

That means here, this Id2 will be equal to Id3.

So now this current Id2 will act as a reference current for this p-MOS current mirror.

And here, this Id4 can be given as W by L ratio of this fourth transistor divided by

W by L ratio of this third transistor times Id2.

So now let's put some numbers so that it will get more clear to you.

So let's say, the W by L ratio of this M1 transistor is 2 times this reference transistor,

while the W by L ratio of this second and the third transistor is equal to 3 times reference transistor.

And let's say, for this fourth transistor, the W by L ratio is 6 times reference transistor.

So with that, this Id1 will be equal to 2 times I(REF), while this Id2 and Id3

will be equal to 3 times I(REF).

And this Id4 will be equal to 6 divided by 3 times Id2.

That means Id4 will be equal to 2 times Id2.

And since Id2 is equal to 3 times I (REF). So we can say that this Id4 will be equal

to 6 times I (REF).

So in this way, we can generate a multiple or the fraction of this reference current source.

So the only thing which we need to ensure over here is that, all the MOSFETs should

operate in the saturation region.

Alright, so now this current Id1 can be used to bias the source follower, while this current

Id4 can be used to bias the common source amplifier.

And the same is shown over here, where this Id1 and the Id4 are generated using the current mirror circuits.

So for this common source amplifier, this PMOS current mirror can be used as a load.

And using the small signal analysis, if we find the equivalent impedance which is

seen from this drain side, then it is actually the output impedance of this M1.

Let's say that is equal to Ro1.

So if we use this PMOS current mirror as an active load, then the voltage gain of this

amplifier will be equal to gm2 times this, Ro2 parallel Ro1. Where this Ro2 and the gm2

are the output impedance and the transconductance of this amplifying transistor.

So in the earlier videos, we have seen that, to improve this voltage gain, a cascode amplifier

along with the cascode current source can be used.

But in that case, there will be an issue of the biasing stability.

So if we want to improve the gain and the biasing stability at the same time, then

the cascode current mirror can be used as an active load.

So if you want to know more about it, then let me know in the comments.

But I hope in this video, you understood what is current mirror and how it can be designed

using the MOSFET.

So if you have any question or suggestion, then do let me know here in the comment section below.

If you like this video, hit the like button and subscribe to the channel for more such videos.