Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Buffer to drive a large capacitive load?

Status
Not open for further replies.

Octago

Member level 1
Member level 1
Joined
Mar 14, 2008
Messages
36
Helped
0
Reputation
0
Reaction score
1
Trophy points
1,286
Activity points
1,557
In my circuits, I want to design a buffer in order to drive a large capacitive load

For example, a source follower is often used in this case as a buffer

But the source follower always introduces a voltage drop between the two stages, which is not desired in my case.

So are there any buffers which can on one hand drive a large capacitive load, but on the other hand not introduce any DC voltage drop?
 

thanks, I just did a quick search of Pushpull class AB buffers in Internet, it seems like such buffer consists of opamp and the other stages, and the design can be quite complex

Are there any simple topology of Pushpull class AB buffer? My goal is only to drive a large capacitive load.


raduga_in said:
Pushpull Class AB buffers can be an option ..

Raduga
 

Octago said:
In my circuits, I want to design a buffer in order to drive a large capacitive load

For example, a source follower is often used in this case as a buffer

But the source follower always introduces a voltage drop between the two stages, which is not desired in my case.

So are there any buffers which can on one hand drive a large capacitive load, but on the other hand not introduce any DC voltage drop?

A simple OTA (long tailed pair), with -ve feedback is what I normally use, far superior linearity to a SF aswell.
 

    Octago

    Points: 2
    Helpful Answer Positive Rating
I think a CMOS inverter with a feedback resistor between input and output should do the job. It would have enough current drive due to the addition of gm of the pMOS and nMOS and the resistor should set the output bias to be the same as the input bias.
 

    Octago

    Points: 2
    Helpful Answer Positive Rating
Old Nick said:
Octago said:
In my circuits, I want to design a buffer in order to drive a large capacitive load

For example, a source follower is often used in this case as a buffer

But the source follower always introduces a voltage drop between the two stages, which is not desired in my case.

So are there any buffers which can on one hand drive a large capacitive load, but on the other hand not introduce any DC voltage drop?

A simple OTA (long tailed pair), with -ve feedback is what I normally use, far superior linearity to a SF aswell.

what is -ve feedback pls?

Added after 2 minutes:

spminn said:
I think a CMOS inverter with a feedback resistor between input and output should do the job. It would have enough current drive due to the addition of gm of the pMOS and nMOS and the resistor should set the output bias to be the same as the input bias.

can this one work? it has gain only in a small portion near the middle of input voltage, and what is the purpose of the feedback resistor?
 

    Octago

    Points: 2
    Helpful Answer Positive Rating
As I said, the resistor can be placed to set the output bias to be equal to the input bias. Secondly, the gain of any amplifier is always within a small range of input. The range of this input can be shifted by changing the ratio of the pMOS to nMOS width.
 

    Octago

    Points: 2
    Helpful Answer Positive Rating
leohart said:
Old Nick said:
Octago said:
In my circuits, I want to design a buffer in order to drive a large capacitive load

For example, a source follower is often used in this case as a buffer

But the source follower always introduces a voltage drop between the two stages, which is not desired in my case.

So are there any buffers which can on one hand drive a large capacitive load, but on the other hand not introduce any DC voltage drop?

A simple OTA (long tailed pair), with -ve feedback is what I normally use, far superior linearity to a SF aswell.

what is -ve feedback pls?

Added after 2 minutes:

spminn said:
I think a CMOS inverter with a feedback resistor between input and output should do the job. It would have enough current drive due to the addition of gm of the pMOS and nMOS and the resistor should set the output bias to be the same as the input bias.

can this one work? it has gain only in a small portion near the middle of input voltage, and what is the purpose of the feedback resistor?

-ve feedback, is the shortened way of writing 'negative feedback'. Feeding the outpu to the inverting input. This configures the ota as a voltage follower.
 

    Octago

    Points: 2
    Helpful Answer Positive Rating
thanks

But if I use a OTA with unity feedback as the buffer, will the problem of oscillation be critical?
 

Octago said:
thanks

But if I use a OTA with unity feedback as the buffer, will the problem of oscillation be critical?

I use them all the time, and not one of them has oscillated, ever. If you're feeding it with square waves then I wouldn't put my mortage on it not oscillating a bit, but if you're worried about it do a few sims.
 

    Octago

    Points: 2
    Helpful Answer Positive Rating
thanks very much again.

it seems like OTA with unity gain as the voltage follower is quite often used. I don't have much practical experience on operational amplifier design.

Are there any reference schematic or datasheet for this circuit?


Old Nick said:
Octago said:
thanks

But if I use a OTA with unity feedback as the buffer, will the problem of oscillation be critical?

I use them all the time, and not one of them has oscillated, ever. If you're feeding it with square waves then I wouldn't put my mortage on it not oscillating a bit, but if you're worried about it do a few sims.
 

I can't actually give you any exact details of the ones we use, but they are basically a standard long-tailled pair (differential amplifier) with full -ve feedback.

I've attached a schematic of a 5uA buffer, you may need to adjust the sizes of components to suit your drive needs and process etc. just choose sensible overdrive voltages etc, for your transistors. If you have the space though, use a 2-stage op-amp (with -ve feedback), we always use these when drivinng the pads of a chip - biased at several mA.
 

    Octago

    Points: 2
    Helpful Answer Positive Rating
You can rebias the DC level by a couple capacitor and a resistor if the frequecny is very high.
And if the frequency is not very high, it will not care the large capacitor.
So can you tell us the frequency and the capacitor value?
 

    Octago

    Points: 2
    Helpful Answer Positive Rating
The frequency is no more than several KHz
the capacitive load needed to drive will not be more than 50pF

jecyhale said:
You can rebias the DC level by a couple capacitor and a resistor if the frequecny is very high.
And if the frequency is not very high, it will not care the large capacitor.
So can you tell us the frequency and the capacitor value?
 

Octago said:
The frequency is no more than several KHz
the capacitive load needed to drive will not be more than 50pF

jecyhale said:
You can rebias the DC level by a couple capacitor and a resistor if the frequecny is very high.
And if the frequency is not very high, it will not care the large capacitor.
So can you tell us the frequency and the capacitor value?

R=1/(jwc).
if the w=2pi*10k and c=50pf, so R is equal to 300M.
So I think 50pF is a very small capacitor for a frequency less than 10K.
isn't it?
 

    Octago

    Points: 2
    Helpful Answer Positive Rating
How big the capacitor load is. I think Class AB is a potentialchoice
 

    Octago

    Points: 2
    Helpful Answer Positive Rating
not more than 10pF


tia_design said:
How big the capacitor load is. I think Class AB is a potentialchoice
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top