I am designing voltage low noise amplifiers from long time using BiCMOS 0.35 SiGe AMS technology. I am looking for new ASIC technology and ask if the bipolar transistor is required for this kind of development. Indeed, my understanding is that bipolar transistor provides larger transconductance than MOSFET for the same bias and roughly same surface area. Moreover, 1/f noise is significantly better with bipolar as compared to MOSFET. Finally, a common way to improve both noise and gain of MOS transistor is to strongly increase their surface area leading to dramatic decreases in bandwidth. These considerations are based on my own analyze of 1nV/√Hz LNA design literature and on my own try of designing full CMOS LNA (using 0.35AMS) showing a GBW more or less limited to less than a 1 GHz and a 1/f noise exceeding the 100 kHz.
Have you any example of dif/dif voltage LNAusing fullCMOS ASIC technoloy fitted
* Low noise : en few kΩ / ore match to 100Ω
* 1/f corner frequency
Dear Damien,
From the specifications you gave in your question it is apparent you seek an operational amplifier as you speak from differential input amplifier. You speak laso from GBW not a fixed gain and bandwidth. So, it is not an rf amplifier.
Would you please clarify the amplifier configuration which you want to achieve.
I think in order to increase the GBW you can use more scaled transistors than 0.35 um Technology generation. You can use for example the 0.18 um technology. Using modre advanced CMOS transistors will enhance your GBW product.
The second solution is to use as you said a bipolar input amplifier provided that you have BICMOS technology. It is more suitable for amplifiers than the MOS transistors.
The last idea is to suppress the noise and extend the bandwidth by modifying the configuration of the amplifiers.
I would like to see the paper in the link: Conference Paper The evolution of the microelectronic bipolar junction transistor
Best wishes
Dear Abdelhalim,
Many thanks for your answer. Our application is not a RF one. We need a voltage amplifier with a fixed gain of 160V/V and a bandwidth of about 20MHz. I am thinking about using a thinner technology, as yes 180 or 120 nm, but not sure that even with such technology, MOS transistors perform the desired requirement.
For more details about our current design, please find here one of our datasheet :
http://www.apc.univ-paris7.fr/~prele/Datasheet_AwaXe_v2_v4dec2019.pdf
Welcome Damien,
By short search in the literature i found that linear amplifiers with specs comparable to your are realized in 0.18 um CMOS technology. Please see for example: Article A 71 dB 150 Variable-Gain Amplifier in 0.18 CMOS Technology
You can further search for linear amplifiers implemented in 0.18 um CMOS technology.
Best wishes
- Badges
- Science topic
- Similar topics
- Social Psychology
- Attitude