How To amplify The MicroVolt Order By Opamp

[farshchian]

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

 

[Ow@i$]

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!!!

 

[keith1200rs]

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

 

[farshchian]

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

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 - - -

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!!!

Thanks Dear smoothcriminal
Ok I am try to do it in 4 stage if you have any other Idia Please inform me.

 

[keith1200rs]

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

 

[FvM]

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).

 

[farshchian]

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?

 

[FvM]

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.

 

[farshchian]

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 .

 

[FvM]

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

 

[farshchian]

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.

 

[FvM]

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.