Op Amp vs Instrumentation Amp
william chan
Posts: 1,326
Today's geek question.
What is the main difference between an Operational Amplifier and a Instrumentation Amplifier?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
www.fd.com.my
www.mercedes.com.my
What is the main difference between an Operational Amplifier and a Instrumentation Amplifier?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
www.fd.com.my
www.mercedes.com.my
Comments
I'm very happy you asked!!! I've been struggling with an op-amp design that I think is supposed to use and Instrumentational Amp instead. So, I've been reading up on both the past two weeks.
In short, and Instr. amp has 3 op-amps internally (schematically). Instr. amps seem to buffer signals better, aka noise. They work very good with ADC's due to a low output impedance. Both the inputs are high impedance. I don't think they suffer from high frequency gain deterioration like a single op-amp design would.
But, with all that said, I wish somewhere someone could explain when, where, and why one would use one versus doing the (seemingly) identical job with a single op-amp. This is where I have been stumped (since last weekend). I've wanted to ask the question, but...
Would someone (who actually knows the theory!) chime in besides me??? I'm VERY interested in hearing about practical applications!
-Parsko
PS - My application uses a air temp guage from and auto, to be used in an auto, that changes in resistance from 1k-10k ohms (in the temp range I'm trying to work with). My op-amp design keeps failing, but I think it is due to TOO low of a value for my feedback loop (i'm using 325 and 3750 ohm resistors, when I think it should be 1000 times that much for both).
http://en.wikipedia.org/wiki/Instrumentation_amplifier
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe
IC Layout Engineer
Parallax, Inc.
I beg to differ (though, you are right in that it covers most of it). But, to a hobbiest, it misses the connections to the supply of the op-amps. In my reading, I seem to keep coming across every op-amp requiring a negative supply voltage (aka +12 AND -12 at the op-amp). Many times, we don't have a -12 supply available. Does this circuit specify that the -12 is actually hooked up to 0V, or -12V. I think this makes a difference. This has been confusing me when looking at op-amp schematics. I have a few links that I keep bookmarked (they are likely the same as found on wiki pages).
www.pic101.com/op_amp.htm
focus.ti.com/lit/an/slod006b/slod006b.pdf
www.analog.com/UploadedFiles/Application_Notes/34661261AN369.pdf
www.eng.yale.edu/ee-labs/morse/compo/sloa058.pdf
www.phy-astr.gsu.edu/nelson/AN-31.pdf
www.analog.com/UploadedFiles/Data_Sheets/897653854AD620_g.pdf
As a hobbiest that does not quite understand the theory, nor have the experience with many applications (or having a mentor on sight to guide me), op-amps are tough to work with. I have determined that trial and error are the best method for me. But, that wiki page doesn't answer all my questions.
-Parsko
·
The supply that you are referring to is negative with respect to "ground".· The ground connection can be imposed from a single voltage supply with the use of two resistors forming a voltage divider.· In your list of references, in the first reference (http://www.pic101.com/op_amp.htm) you·can see an example of how this is done by looking at Fig 26 and Fig 28.
·
However, the relationship of V+ to GND to V- has nothing to do with the difference between an Operational Amplifier·verses an Instrumentation Amplifier.· That's another topic in and of itself.
·
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe
IC Layout Engineer
Parallax, Inc.
You just provided the "eureka" moment I needed. It's a "relative" ground reference!!!! Wow, that makes total sense now. We need a "eureka" Emoticon (one that would indicate that I just said out loud in a crowded office "ooooohhhhhhhhhhhhh")!!!!!
Thanks!
Now, back to the subject. Why would one use an inst. amp versus an op-amp? Yeah, higher accuracy, lower noise, etc... But, at what threshold does one pull the trigger and use an instr. amp? Is there a rule of thumb? When aren't we trying to get an accurate signal? What constitutes accuracy? I'm playing a bit of devil's advocate, but for those that know less than me (which is "relatively" hard), it's a hard question to answer.
-Parsko
This book discusses more of the options than I really care to know. At the end of the day, the name instrumentaion amplifier as one package might just command a higher price. You can get 4 op amps in one package and create a better impedance relationship taylored to your exact needs. It is all about precision and low noise.
Page 425 has a section 7.10 discussing 'standard three op-amp instrumentation amplifiers' [noparse][[/noparse]Fig7.32- Classic instrumentation amplifier]
I even see a schematic [noparse][[/noparse]Fig. 7.33, page 26] all 4 units of a quad op amp being used to make an Instrumentation amplifier with output, guard, sense, and reference terminals. And of course, the one package requires only nominal power supply connections.
The output is obvious. The guard goes to the coaxial inputs eternal shield. the reference goes to the load's grounded side, and the sense shares connection with the output on the load's higher voltage side.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
PLEASE CONSIDER the following:
Do you want a quickly operational black box solution or the knowledge included therein?······
A true differential amplifier has the ability to cancel out any common mode signals that may get introduced on the input.
A conventional Op-Amp can be configured to work in a differential mode, but because the output of the Op-Amp is often tied
back to the inverting input or non-inverting input to increase/decrease the amount of desired gain, this tie-back configuration
can introduce an imbalance that causes common mode signals to pass due to non-linearity’s·and get amplified when the idea
is to have them suppressed.·
·
An instrumentation amplifier is nothing more than a standard Op-Amp configured in a differential mode with isolation buffering
to the inverting input and non-inverting input.· Because of the isolation buffering, you can·avoid the bias condition mentioned above.
·
Nut shell bottom line...., is that if you have a differential signal, then you should use an Instrumentation Amplifier to achieve your
absolute best signal to noise ratio on the output.· If you have a single ended signal, then a standard Op-Amp design is usually sufficient.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Beau Schwabe
IC Layout Engineer
Parallax, Inc.
The gem of information one cannot gleem by reading theory. Thank you very much Beau. The gem provided (me) the right search words, and subsequently, a much better understanding of electrical communication. This seems like something a CS101 professor would say on day one, and thus becomes common sense to professionals (but completely frustrating to the hobbiest!)
-Parsko
http://focus.ti.com/lit/an/slod006b/slod006b.pdf
You can buy a printed version, but it's like $60 new. Might be worth it if you're really into electronics.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
- Rick
I've been sitting in Taiwan and wanting a good text that is just about Op Amps. I even purchaces an AsusEEE minilap top to store all those huge PDF when I go to Starbucks. Not only are they free, I don't have to carry around a 10 pound text. In fact, I have about 300 pounds of text in the Asus; including the Propeller Manual, the SX stuff, and BasisStamp Manual.
Chapter 12 is entitled "Sensors to ADC" and looks extremely helpful.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
PLEASE CONSIDER the following:
Do you want a quickly operational black box solution or the knowledge included therein?······
Post Edited (Kramer) : 2/29/2008 9:06:45 AM GMT
There are some op amps, like the LM3900 that were specifically designed to operate from Single Sided supplies.
I looked at the sample designs for the 3900 and oddly enough an instrumentation amp design was not provided.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
There's nothing like a new idea and a warm soldering iron.