Author Topic: Choice of Op-amp  (Read 3024 times)

0 Members and 1 Guest are viewing this topic.

Offline legostar74

  • Newbie
  • Posts: 4
Choice of Op-amp
« on: March 17, 2015, 03:01:14 am »
Hi, I'm new to Op-amp's, I have seen Dave's video tutorials about them and learned a lot  :-+
Thanks a lot Dave for your great tutorials  :)

I need to amplify an signal that ranges from 4mV up to 104mV.
Going to read the value with the 5V - 10-bit ADC on my Arduino Mini Pro.
In the lowest range (4mV) I need an accuracy of 20uV per step.
I'm able to supply the Op-amp with 6.4V single supply.
My signal is running at 200hz, and I'm going to measure the signal one time each pulse.

I try to accomplish this by using 3 Op-amp's 48x, 221x & 471x gain.

Could you help me by suggest a good value Op-amp that works in this range?

Right now I'm trying to simulate this in Multisim with NE5532P, but I seem to get the values a bit off.
Don't know if it's the bias current that makes this? A resistor at the non inverting input seems to help.
But the gain resistors in parallel seems to be to high value for this resistor.

I have also tried LM833N with much better results, seems quite steady in my simulations.
Is this Op-amp a good choice for this application?

Is there any other Op-amp that might be a better choice?

I have seen some schematics where they put a capacitor at the gain resistor? Do I need these capacitors?

Thanks a lot for any input,

Best Regards,
Anders aka Legostar
 

Offline legostar74

  • Newbie
  • Posts: 4
Re: Choice of Op-amp
« Reply #1 on: March 17, 2015, 03:44:38 am »
Thanks a lot for your input, I'm going to try your advice with multi-rails on the simulation and see if I can get better results.

Yes it's a current sensor and R1 is the load, and R6 sense.
I use 2S lithium batteries so voltage can vary from 8.5V to 6.4V - but I simulate the lowest condition.
 
The 4th Op-amp U2A I setup as a comparator, to get interrupt for over current / short detection.

But I don't need the capacitors at the gain stage? In the spec for the Op-amp they put small capacitors parallel to the gain resistor.

« Last Edit: March 17, 2015, 03:59:37 am by legostar74 »
 

Offline legostar74

  • Newbie
  • Posts: 4
Re: Choice of Op-amp
« Reply #2 on: March 17, 2015, 04:36:30 am »
Thanks, nice to see your example :)

I only need to measure current down to 3A with 16mA accuracy.

I'm thinking about using a INA225 or INA226 later on, but my supplier do not have them.
So trying to use some regular Op-amps instead, and also learn how to use them :)
So now I want a negative feedback T-shirt :)

« Last Edit: March 17, 2015, 04:38:38 am by legostar74 »
 

Offline legostar74

  • Newbie
  • Posts: 4
Re: Choice of Op-amp
« Reply #3 on: March 17, 2015, 04:50:53 am »
Yes going to use LP2950-50LPRE3 5V .1A as Vref - and measure battery voltage at Probe4 .
Hope that's going to be accurate enough.

I'm a completely newbie so I hope it's going to work out  :-DD
Otherwise I really need a INA226 :)
« Last Edit: March 17, 2015, 04:58:00 am by legostar74 »
 

Offline NANDBlog

  • Super Contributor
  • ***
  • Posts: 4660
  • Country: nl
  • Current job: ATEX certified product design
Re: Choice of Op-amp
« Reply #4 on: March 17, 2015, 01:22:48 pm »
Yes going to use LP2950-50LPRE3 5V .1A as Vref - and measure battery voltage at Probe4 .
Hope that's going to be accurate enough.

I'm a completely newbie so I hope it's going to work out  :-DD
Otherwise I really need a INA226 :)
Most of the thing you are trying to do has mistakes. If you want to measure something and have some decent accuracy, even for 10 bit, you need to make some effort understanding, where the errors come from. The LP2950 has some 0.4% line regulation and 150PPM tempco, only enough for some 8 bits or so. The 5532 has 5mV inpu offset, if you amplify that 400 times, you will end up with 40% offset error on 5V. Also, you are planning to drive the ADC with 5Kohm source probably you end up with huge INL errors.
To top this, you are not using a rail to rail input opamp to measure some millivolts of signals.
My suggestion is to throw away any book you were learning from, and start reading application notes from Linear technology, Analog devices  and Texas instruments.
 

Offline dannyf

  • Super Contributor
  • ***
  • Posts: 8229
  • Country: 00
Re: Choice of Op-amp
« Reply #5 on: March 17, 2015, 02:20:55 pm »
Quote
My signal is running at 200hz, and I'm going to measure the signal one time each pulse.

What does that mean?

Quote
Right now I'm trying to simulate this in Multisim with NE5532P, but I seem to get the values a bit off.

Could be due to input offset being amplified - the input resistors are so small that they are not likely to have much impact on the output offset.

The solution is to use low Vos opamps. Or to calibrate it in software.
================================
https://dannyelectronics.wordpress.com/
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf