Hi, I have 2 questions about the current noise difference between BJT OPA & FET/CMOS OPA:
1. When looking at OPA134/300 and other FET/CMOS OPA's current noise density, it seems has a inverse noise shape with voltage noise... i mean the higher the bandwidth cause higher current noise (I call it white noise then f noise...). Why?
2. FET/CMOS OPA's current noise shape is different from Bipolar OPAs such as OPA277, the BJT OPA's noise and current noise density has the same shape(1/f noise then white noise). Why BJT's current noise has the same shape with voltage noise?
Thanks a lot for your reply.
The increase in noise is called f-squared noise. This noise is the result of an interaction of the input voltage noise and the input capacitive reactance. I have an attached a preliminary article on this subject. This article has not been published. I have published a series of articles on noise on en-genius (see link below).
Art Kay Senior Applications Engineer High Performance Linear
I am so honored to have you answered my question. Actually, your series of articles about OPA Noise have been translated into Chinese, I have studied #1-#7 already. Thanks for introducing #8 and the latest one. I will try to understand the f squared noise from your articles. I will reply you when I have further questions.
BTW, I have added an Word file to describe my question more clear to other visitors.
So, do you think the f-squared noise existing in all kinds of OPA? But some of them didn't mention it in their datasheet, such as OPA277, OPA211.
I’m glad you were able to make use of the articles. Thanks for the feedback.
Yes. F-squared noise exists on all op-amps. On bipolar amplifiers the magnitude of the broadband current noise is larger then the f-squared noise, so you can’t see the f-squared noise. OPA211 and OPA277 are bipolar amplifiers. In FET input amplifiers, the broadband current noise is smaller so you can see the f-squared noise in measurements and simulations.
However, in new datasheets we do not show f-squared noise. This is because f-squared noise is an interaction between voltage noise and the input captive reactance of the op-amp. So, for example, if you were developing a spice model for the op-amp, the f-squared noise would automatically be generated by the input capacitance and the voltage noise. We do not want people to erroneously count both f-squared noise in the current noise spectral density and f-squared noise from interaction with the voltage noise and input capacitance. The only source of f-squared noise is from interaction with the voltage noise and input capacitance.
A good tool to help understand f-squared noise is Tina. Try running a few Tina simulations with and without input capacitance to see the f-squared effect.
Yes, simulation and hands-on will give me more deep understanding on it. It's great to learn from you. Thanks a lot for your help!
All content and materials on this site are provided "as is". TI and its respective suppliers and providers of content make no representations about the suitability of these materials for any purpose and disclaim all warranties and conditions with regard to these materials, including but not limited to all implied warranties and conditions of merchantability, fitness for a particular purpose, title and non-infringement of any third party intellectual property right. TI and its respective suppliers and providers of content make no representations about the suitability of these materials for any purpose and disclaim all warranties and conditions with respect to these materials. No license, either express or implied, by estoppel or otherwise, is granted by TI. Use of the information on this site may require a license from a third party, or a license from TI.
TI is a global semiconductor design and manufacturing company. Innovate with 100,000+ analog ICs andembedded processors, along with software, tools and the industry’s largest sales/support staff.