What is effect of mismatched op-amp power supply on output error? 3

[yinthenovice]

I am trying to amplify a signal from in the range of (0.5mV-5mV) by a factor of 10 using TL082 op amp with non-inverting configuration for instrumentation but my power supply is not exactly the same . I.e +15.3V and -14.8V instead of +/- 15V. When i used Multisim it told me this would make a difference in output , how do i quantify this error ? i.e how do i put this into error budget?

I have already accounted for input bias current and input offset voltage but still getting some deviation from ideal behavior .I really appreciate any help

 

[Skogsgurra]

Look at the PSSR in the data sheet. That will show what power supply variations do to your total error. With the small deviations you have, I wouldn't be too worried.

Are you amplifying DC or AC? The TL082 has modest band-width.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[yinthenovice]

thanks for the info , its for a DC signal i'm trying to amplify coming from a shunt resistor. I guess sometimes you want to pull your hair out when you can't explain away that extra bit of noise, especially if you're mechanical trying to learn this stuff . I also think , after looking at Power Supply Rejection Ratio (I assume thats what PSSR stands for)it wont be a big deal.

Thanks again

 

[IRstuff]

PSRR is an AC specification that describes the amount of power supply noise that would leak into the output.

However, you are measuring a DC signal and outputting a DC output. You are unbalancing the supplies by 0.5 V, and I would suspect that the output offset will more than swamp out your intended signal.

You can, of course, compensate for that with an input bias, or, if you are digitizing the signal anyway, you can digitize the output with zero input and store that away as the pedestal error.

TTFN

FAQ731-376

 

[benta]

Refine your opamp selection criteria.
The TL082 is 30 years old, has nice high input impedance, but is notorious for offset problems.
Find something up to date.

Benta.

 

[Skogsgurra]

Good catch! YesPSSR should be PSRR. Finger trouble.

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[yinthenovice]

I love this forum , you guys are amazing... In case anyone is intersted , this is what I ended up doing :

I just stayed with the power supply i had in the mean time , then i calibrated the amplifier using a much more accurate digital multimeter than I'll be using for the application. So pretty much varied my input voltage over the range i'm expecting using a variable resistor and noted the output at each input to come up with a transfer function. This way i was able to include all the other nasty effects of input bias current , input offset voltage etc . As far as i can tell this gets me to within about 1% of the "true" value .. good enough for me.

@Skogsurra , i wish i read ehhm ...Finnish/Swedish? good looking site though.

@Benta, you're probably right , there's better stuff out there and i'll replace TL082 with something better, i'm mechanical learning the hard way

@IRstuff , did something very similar to what you suggested

 

[Skogsgurra]

Cross out Finnish - I understand very few words of that language. Swedish is correct. How did you get there?

Gunnar Englund

http://www.gke.org

--------------------------------------
100 % recycled posting: Electrons, ideas, finger-tips have been used over and over again...

 

[MacGyverS2000]

Gunnar,

I'm going to guess he got there by the link in your sig... ;-)

Dan - Owner

http://www.Hi-TecDesigns.com