Mixed Signal Circuit Design || DAC Architecture Part 2
Oct 6, 2023
The next one is Charge Scaling DAG.
A very popular DAG architecture used in CMOS technology is the Charge Scaling DAG.
This Charge Scaling DAG is the most popular DAG architecture used in CMOS.
This figure is the example of Charge Scaling DAG.
A parallel array of binary weighted capacitors totaling 2 raised to n capacitors is connected to an op-amp.
The value of C is a unit capacitance of any value.
After initially being discharged, we check if all the capacitors are discharged.
Some digital switches are connected to it.
These digital switches can be connected as V-reference or ground.
An output is formed.
The output is a function of voltage division.
The capacitor circuit acts as a voltage divider.
The output is a function of that.
The second figure is an equivalent circuit.
MSB is the equivalent circuit at the time of 1.
MSB1 is connected to V-reference.
Everything else is connected to ground.
When this is connected to V-reference, a voltage of 2 raised to n-E is generated.
The equivalent of parallel capacitors is 2 raised to n-1C.
The equivalent of all other capacitors is connected to V-reference.
We can draw the equivalent diagram of this circuit.
The total capacitor array totals 2 raised to n into C.
Therefore, MSB is high and the remaining bits are low.
The voltage divider occurs between MSB, capacitor and rest of the array.
MSB is high and the remaining bits are low.
The voltage divider occurs between MSB, capacitor and rest of the array.
This is the equation for the output voltage.
Output voltage is a voltage follower circuit.
We get the output voltage here.
To find the voltage here, we divide the main voltage into resistance branch by total.
Here also, Vout is equal to main voltage, which is V-reference, into capacitance of this branch.
2 raised to n-1 into C divided by total capacitance.
Total is 2 raised to n-1 into C plus all other effects.
We can write this equation as V-reference into 2 raised to n-1C divided by 2 raised to n-1C.
2 raised to n-1C into 2 raised to nC.
2 raised to n-1C divided by 2 raised to nC.
Vout is equal to V-reference divided by 2.
The meaning of MSB bit 1 is that it reads half of the maximum value.
Which confirms the fact that MSB changes the output of DAC by 1 by 2 V-reference.
Now, we are converting this equation into a generalized form.
Generalized form is MSB bit 1.
Vout is equal to 2 raised to k into C divided by total.
Total is 2 raised to nC into V-reference.
2 raised to k minus n into V-reference.
If dk is 1 and all other bits are 0, we can generalize this equation.
Vout is equal to dk.
Vout is equal to sigma k is equal to 0 to n-1.
dk is 2 raised to k minus n into V-reference.
This is the charge scaling DAC.
Now, let's see the main limitation of this.
The main limitation of charge scaling DAC is the existence of parasitic capacitance at the top plate of the capacitor array due to op-amps.
The existence of parasitic capacitance is due to op-amps.
This will prohibit it use as high-resolution data converter.
That's why we don't use this as high-resolution data converter.
A better implementation is to use switched capacitor integrated circuit as driving circuit to reduce the parasitic capacitance effect.
Now, let's see an example.
Design a 3-bit charge scaling DAC and find the value of output for D2, D1, D0 is equal to 0, 1, 0 and 1, 0, 1.
Assume that V-reference is equal to 5V and CE is equal to 0.5V.
Draw a circuit diagram corresponding to 3-bit charge scaling DAC.
Find the value of D2, D1, D0 is equal to 0, 1, 0.

