The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Analog Design >> Wideband Amplifiers design
https://designers-guide.org/forum/YaBB.pl?num=1177389742

Message started by amitsh on Apr 23^rd, 2007, 9:42pm

Title: Wideband Amplifiers design
Post by amitsh on Apr 23^rd, 2007, 9:42pm

Hi ppl,

I have to design an op-amp for the Switched capacitor S/H amp, for pipeline adc, my specs for the op-amp design are high Bandwidth of 500MHz, gain=50-60dB, settling time less than 5ns and phase margin of 45 deg. so please any body can tell me what kind of op-amp is used to design wideband amplrs.

Amit

Title: Re: Wideband Amplifiers design
Post by carlgrace on Apr 29^th, 2007, 7:20am

Usually you try to use a cascoded single stage op amp for a pipelined ADC. If there isn't enough headroom, you can use a second common-source stage but that will limit your speed. You can also use a gain-boosted cascode to improve the gain. These amplifiers are discussed in most analog IC books. The simplest is probably Johns and Martin.

Carl

Title: Re: Wideband Amplifiers design
Post by hspice on May 2^nd, 2007, 3:44am

Depending on what process you have. If smaller than 0.35, I think a good tuned two stage amplifier is OK. The benefits of it are high gain, good out swing and more fitful at SC circuit (you may find it's very important), but with dark side of high power consumsion and little bit complicated.

Good luck.

Title: Re: Wideband Amplifiers design
Post by gipo on May 3^rd, 2007, 2:49pm

Hi there.
I would start using a classical 2 stages Miller Opamp.
1. the Gain-Bandwidth is well controllable: gm,in/Cc, but you must have a high current consumption budget to get 500 MHz;
2. the open loop phase response has the minimum number of extra poles/zeros others than dominant, non-dominant poles and RHP zero; in a cascode topology the high frequency poles might likely fall at frequencies close to 500 MHz (depending on the process), thus compromising the opamp's stability; watch out your feedback network, which might likely add extra poles at the frequencies of interest, in the open loop AC response, influencing then the stability;
3. the output swing is the widest available and a good linrarity can be expected;
4. a 50-60 dB DC gain can be easily accomplished, depending on the load impedance.