Advertisement

Blog

6 Burr-Brown secrets to success you need to know about instrumentation amplifiers

My dear friend, Neil Albaugh—a veritable analog ‘Wizard’, wrote an often unnoticed handbook in January 2000 at Burr-Brown which stands the time 16 years later and will be still relevant in the future as well—Analog basics are “evergreen”. The title was The Instrumentation Amplifier Handbook and I would like to share the top 12 most important aspects of designing a functional and robust circuit using the instrumentation amplifier. You can access it online here. All images and references are attributable to this excellent handbook and there are far more details and tips about INAs in the handbook that will make you an expert in design.

Click on the slideshow image below to see some insightful tidbits about INAs or IAs, as they are sometimes called, that might give you a feel for proper usage and design using this long-time useful tool for Analog designers:

Here we see a poor choice in the use of an INA in a high impedance transducer application. The designer needs to create a low noise pre-amp for the piezo accelerometer source at its input. Well, the INA103 has under 1 nV/
	√
Hz and low distortion with excellent common mode rejection. Also, a very low noise amplifier will want to have a low source impedance. That 1 MΩ resistor, needed for extending the low-end frequency response of the transducer/amplifier combination will not bode well for this design. The INA103, chosen here, has an input bias current of 2.5uA typical and 8uA max which will pass through the 1 MΩ resistor creating a whopping 2.5V input offset at the inverting input of the INA----with the INA's gain of 100---the output saturates at the rail. In these types of high impedance sources, a better choice of amplification would be a very high impedance input JFET or CMOS amplifier.

Here we see a poor choice in the use of an INA in a high impedance transducer application. The designer needs to create a low noise pre-amp for the piezo accelerometer source at its input. Well, the INA103 has under 1 nV/ √ Hz and low distortion with excellent common mode rejection. Also, a very low noise amplifier will want to have a low source impedance. That 1 MΩ resistor, needed for extending the low-end frequency response of the transducer/amplifier combination will not bode well for this design. The INA103, chosen here, has an input bias current of 2.5uA typical and 8uA max which will pass through the 1 MΩ resistor creating a whopping 2.5V input offset at the inverting input of the INA—-with the INA’s gain of 100—the output saturates at the rail. In these types of high impedance sources, a better choice of amplification would be a very high impedance input JFET or CMOS amplifier.

5 comments on “6 Burr-Brown secrets to success you need to know about instrumentation amplifiers

  1. femia0
    May 25, 2016

    A very good article on how or not to use INAs. The link of the mentioned handbook didn't work. Can you supply a new link.

    Best regards,

    A. Mixich

  2. Steve Taranovich
    May 25, 2016

    @femla0—try the link again—it works for me

  3. Navelpluis
    May 25, 2016

    Thanks for posting. This shows that analog stuff certainly is not trivial, eh, well, that it mostly goes wrong when you do not think about all the nasty influences. Anyway, I still do love this stuff and try to earn some money with it 😉

  4. Steve Taranovich
    May 25, 2016

    @Navelpluis—glad to hear you are making a living with Analog

  5. Neil.Albaugh
    February 4, 2017

    Thanks, Steve. The “Instrument Amplifier Handbook” was only a first rough draft of my intended book. While at Burr-Brown it became clear that, at that time, instrumentation amplifiers were not well known or understood by analog designers. Op amps had become familiar by then and much literature, such as Walter Jung's “Op-Amp Cookbook”, was available to guide a user in their applications. Little was available for instrumentation amplifiers.

    I was writing it when BB was acquired by Texas Instruments. The TI policy of publications was more restrictive than at BB, requiring reviews and approval by appropriate groups before proceeding. No one at TI was willing to devote any time to doing it. The project fell into limbo and somehow a copy of the draft that had been sent to a BB customer to help him with his circuit fell into the Internet. Now the “book” is all over the 'net.

    There may still be some useful information in my rough draft  but it is pretty dated by now. I've retired to do some consulting and building a race car so I'm not going to revise it now. Someone else might want to take the IA Handbook idea and expand and update it. They have my Blessing. 🙂

    Regards, Neil Albaugh

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.