Question

Notify Me Bookmark

This question: Moore’s law predicts that the number of transistors on an integrated circuit doubles every two years. If the microprocessor shown above dissipates 60 watts at VDD = 3 V, then find the value of VDD that will be required for the next-generation microprocessor in 4 years with transistors working at doubled frequency (2 times). Assume the device capacitance remains constant and that we want the power dissipation to remain unchanged at 60 watts. 1.5 V 530 mV 750 mV 1.06 V 864 mV

This question: Moore’s law predicts that the number of transistors on an integrated circuit doubles every two years. If the microprocessor shown above dissipates 60 watts at VDD = 3 V, then find the value of VDD that will be required for the next-generation microprocessor in 4 years with transistors working at doubled frequency (2 times). Assume the device capacitance remains constant and that we want the power dissipation to remain unchanged at 60 watts. 1.5 V 530 mV 750 mV 1.06 V 864 mV

Image text
This question: Moore’s law predicts that the number of transistors on an integrated circuit doubles every two years. If the microprocessor shown above dissipates 60 watts at VDD = 3 V, then find the value of VDD that will be required for the next-generation microprocessor in 4 years with transistors working at doubled frequency (2 times). Assume the device capacitance remains constant and that we want the power dissipation to remain unchanged at 60 watts. 1.5 V 530 mV 750 mV 1.06 V 864 mV

Detailed Answer

0 0

Please enter your email here. You will get email when this question is answered by our tutor.

Doubtrix Logo

Problem is not yet solved!

Click on Notify me to get email when question is answered.

Join Doubtrix with 7-days free trial

  • Icon Correct & Verified Answers
  • Icon No AI-Generated Answers
  • Icon Video Solution for Difficult Questions
  • Icon Work With Experts to Reach at Correct Answers
Notify me Login Sign Up

Similar/Related Questions

For the two-stage OTA below, assume that Iref = 50 μA and that all transistors are designed for |Veff|= 0.2 V. Assume: |Vt|= 0.5 V for all devices, Vdd = 3 V, M5 and M7 are 2X larger than M8, λ = 0.2V-1. 1. Calculate the total power consumption of this OTA. 2. Complete the table below for the small-signal parameters of all transistors. Transistors Ids Veff gm gds ro μf 3. What is the gain of this OTA in dB? 4. What is the CMRR of this OTA in dB? 5. What is the common-mode input range of the circuit? 6. What is the output swing of this OTA? 7. Assume that this OTA has parasitic capacitances C1 = C2 = 100fF, where C1 is the total shunt capacitance at output of stage 1 and C2 is the same for stage 2. What value of Cc is needed to set the gain-bandwidth product equal to the second-pole location (resulting in 45-deg phase margin)? 8. For this value of Cc, calculate the GBW (in MHz, careful of the 2π ) and slew rate (in V/us), where slew rate is defined as 9. What value of RZ is needed to set the zero location to the LHP at 1.7X the GBW value? 10. For the values you have selected in Table 1, will this OTA have any systematic offset? Why or why not?

Solution
0 0

Cascode Stage. Design a Cascode amplifier shown in figure below to achieve a gain of 1600. Assume VDD = 1 V, Vth = 0.3 V, λn = 0.2 V−1, γ = 0, μnCox = 100 μA/V2, I1 = 1 mA, gmro ≫ 1. a) Find bias point (VGS−Vth of transistors) to achieve the target gain such that we achieve maximum swing at Vout. b) Find W/L for each transistor. c) Calculate optimal Vb and VIN DC bias to maximize output swing. What is this maximum/minimum voltage levels at the output? What is the maximum swing? d) Notice you must generate a specific Vb externally to maximize the swing. If a lazy designer decides to connect Vb to VDD instead, what would be the swing in this case? e) To realize the current source load, we will use a PMOS active load. For PMOS, use the same parameters with λp = 0.2 V−1, μpCox = 50 μA/V2. Assume the bias current will remain at 1mA, find the new gain value. What would be the W/L for PMOS if we want to have a VOD(VGS−Vth) of 0.1V for this device. f) Repeat part (e) for a PMOS load with 2L gate length. g) It seems like making the PMOS load very large (large W) is beneficial to reduce VOD and consequently improving the voltage swing. What do you think is the drawback?
Solution
0 0

Problem 3: (Short Circuit Power Consumption) For the circuit in Figure 2, assume that the input Vin is a periodic square wave signal (called "clock") of frequency 200 MHz, and has a rising transition time (from 0−100% of VDD ) of 0.2 ns and a falling transition time (from 100−0% of VDD ) of 0.2 ns. For the inverter in the figure, find the short circuit power consumption. Use the procedural hints listed below. Procedural Hints:Calculate the short circuit duration, tsc. Short circuit currents occur when both nMOS and pMOS transistors are simultaneously ON. Use the rise/fall times given. Calculate the peak short circuit currents. For simplicity assume that Ipeak_sc_LH and Ipeak_sc_HL are equal, and calculate the latter only. To calculate Ipeak_sc_HL, , assume that the short circuit current for the H-to-L transition peaks when Vin = VDD/2 and Vout = 9 VDD /10 (this is an arbitrary choice assumed for this problem alone). Now calculate the pMOS transistor current - this will be Ipeak_sc_HL. . Question: Does the nMOS current equal the pMOS current? If not, how do you account for InMOS−IpMO ? Using your results from (1) and (2), calculate the short circuit power dissipation. Note: Assume that VDD = 1.0 Volts, (W/L)p = 3.5 /0.1, and (W/L)n = 2 /0.1. Transistor properties are given by the 90 nm model parameters handout posted on the class web-site. For the current calculation, use the transistor parameters (not the equivalent resistance parameters) and ignore channel length modulation. Assume that there is an 89 fF load on Out.

Solution
0 0