Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

How To amplify The MicroVolt Order By Opamp

Status
Not open for further replies.
F

farshchian

Junior Member level 1
Joined
Oct 23, 2010
Messages
15
Helped
1
Reputation
2
Reaction score
1
Trophy points
1,283
Activity points
1,388
How I can amplify a micro volts range to volt range?
Which type of op-map i must use?I have very low signal about 1 micro volt and i want to amplify it to about 1 Volts and measure it's frequency it's frequency is in audio range and almost is about about 1-3 KHZ,I use two stage of AD620 IC WITH rg=50 OHM each stage gain is about 1000 so i increase gain 1000000 but the OUTPUT is very noisy when i see it in o-scope how can i solve this problem?
i know there is some Op-map LNA (low noise amplifier) but i haven't them could i use simple type for detecting this signal?what is chopper method in reducing offset voltage in op-amp?

thanks to your help
 
Last edited:

With such a high gain one can expect noise at the output, a gain of 1000000 in just two stages may cause the issue, gradually increasing the gain with more IC that are properly coupled and a low pass filter at the output with fc say 4Khz will result in better performance!!!
 

What is the impedance of your signal source? A 1k resistor will have around 286nV RMS of noise over 5kHz bandwidth. The AD620 has 0.28uV of low frequency noise and 9nV/rt(Hz) of input noise which will be 636nV RMS over 5kHz. If you want to measure microvolts you need either a lower bandwidth or a lower noise amplifier. Look in manufacturers opamp selection guides (TI/Linear Technology/Analog Devices for example).

Keith
 

keith1200rs said:
What is the impedance of your signal source? A 1k resistor will have around 286nV RMS of noise over 5kHz bandwidth. The AD620 has 0.28uV of low frequency noise and 9nV/rt(Hz) of input noise which will be 636nV RMS over 5kHz. If you want to measure microvolts you need either a lower bandwidth or a lower noise amplifier. Look in manufacturers opamp selection guides (TI/Linear Technology/Analog Devices for example).

Keith
Click to expand...
Hello Dear Keith
Thanks to your Reply, about source impedance it's about 6 ohms (it's a coil) yes you are right but i can't found Low noise amplifier in my area is there any way to do it? is there any source for study about amplify this range of voltage i want to learn more about this.

- - - Updated - - -

smoothcriminal said:
With such a high gain one can expect noise at the output, a gain of 1000000 in just two stages may cause the issue, gradually increasing the gain with more IC that are properly coupled and a low pass filter at the output with fc say 4Khz will result in better performance!!!
Click to expand...
Thanks Dear smoothcriminal
Ok I am try to do it in 4 stage if you have any other Idia Please inform me.
 

Don't worry about the offset - that is irrelevant for an AC signal. A non-inverting amplifier configuration with a capacitor in series with the inverting-input resistor won't amplify the offset.

Look for low voltage noise opamps. Linear technology have several below 1nV/rt(Hz) for example.

The best amplifiers are probably going to be discrete. I have designed several magnetometer amplifiers using the Interfet JFETs which get you below 0.5nV/rt(Hz) but I cannot share them here.

Keith

- - - Updated - - -

Food for thought: http://www.linear.com/product/LT1128
 
  • Like
Reactions: FvM

    FvM

    FvM

    Points: 2
    Helpful Answer Positive Rating
Classical low noise amplifiers for very low source impedance have been often utilizing input transformers, e.g. moving coil pick-ups or ribbon microphones. Next best are discrete JFET amplifiers with multiple parallel input transistors for noise matching. LT1028/1128 is the lowest noise OP also in my list (five years old, but I guess no changes since then).
 
  • Like
Reactions: tpetar

    tpetar

    tpetar

    Points: 2
    Helpful Answer Positive Rating
Thank you Dear keith;
ok right now i use LT1028 that i found it from some old board, I want to use this amplify for magnetometer too, as i say my coil resistance is about 6 ohms and its Inductance is about 7mh i use two stage Lt1028 with 100Kohm to 50 ohm ratio that about 2000 time so totally about 4000000 times increasing the gain but in oscilloscope i see only the square wave about 100KHZ so it seems that i only make a 100KHZ oscillator.
I think i should put a filter at first? Could u explain some about this problem?
 

The most likely cause of oscillations is an unwanted output to input crosstalk in the two-stage amplifier. We would need a complete schematic to discuss the problem.

The ground design must be well considered, you'll surely want extra power supply filtering for the first OP stage.
 
  • Like
Reactions: tpetar

    tpetar

    tpetar

    Points: 2
    Helpful Answer Positive Rating
Hi Dear FvM;
Thanks to your reply my circuit like attached image i configure it on test Board, But only i see an square oscillation on oscilloscope .

 

Some points that must/should be changed:
- n.i. input of second OP must no be floating
- you'll will get rather high output DC offset, you should consider AC coupling between second and first stage
- as said, sufficient power supply decoupling might be required for stability
- if you implement the circuit on a testboard, ground layout will be very critical
 
  • Like
Reactions: tpetar

    tpetar

    tpetar

    Points: 2
    Helpful Answer Positive Rating
Sorry you are right i make a mistake in drawing but in real isn't float, it's like attached photo, and when i put a 100nf capacitor between the stage one and two only the output square wave frequency change and but already oscillate.
 

I didn't say that AC coupling would help against oscillations. AC coupling would also require a bias resistor to ground.

It might help to add bandwidth-limiting capacitors parallel to the 10 k resistors, e.g. 100 pF up to 1 nF.
 
  • Like
Reactions: tpetar

    tpetar

    tpetar

    Points: 2
    Helpful Answer Positive Rating
Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top