Electronics > Beginners
Looking for a high gain LNA at acoustic/ultrasonic frequencies
(1/2) > >>
raff5184:
Hi all,
I am looking for an LNA OpAmp  that can give me 60dB of gain at relatively low frequencies - about 150kHz -  narrow bandwidth signals, and powered by 5V battery in single supply mode.

I have input signals of 5mV in the  140-160 kHz range.
I found some LNAs but they are mainly for higher frequencies >100MHz. Are there OpAmps that can satisfy my requirements?

Or can I use "normal" OpAmps and design the appropriate circuit? And in this case what are the parameters that I need to consider for low noise amplification?
magic:
60dB at 150kHz is 150MHz gain-bandwidth product. That's on the fast side of audio opamps ;)
A two stage solution would enable the use of slower amps and only one would be critical in terms of noise performance.

For a single stage, AD797 with decompensation capacitor is one easy bet and it has very low voltage noise albeit a bit of current noise. Lists "sonar" and "ultrasound" among recommended applications. Expensive.

You need to know your source impedance to correctly choose between voltage noise and current noise. You also need to know your target SNR / tolerable nV/rtHz or some other number specifying how much noise you can put up with. Perhaps even an LM358 would meet your spec ;)

There is also the question of gain accuracy and distortion. If you need it low, you don't want to run close to GBW limits so two stages it is.
Kleinstein:
60 dB gain at 150 kHz is already requires a rather fast amplifier. I would prefer 2 stages for this.
The OPs usually give the GBW, that is the product of frequency and gain. Normally one some >20 dB or so of extra OP gain to compensate errors. So a single stage would need an > 1.5 GHz.

Another important factor is the source impedance - there are different LNAs for high and low impedance.

5 V supply and 60 dB gain to a 5 mV does not go together well. A question is also how much accuracy / stability of the gain is needed.
Zero999:
The bandwidth is only 20kHz, so I wonder how well a discrete amplifier would work?
raff5184:

--- Quote from: magic on September 19, 2019, 02:08:28 pm ---For a single stage, AD797 with decompensation capacitor is one easy bet

--- End quote ---
Thanks, I'll try it, AD sends free samples, even if cost is not a problem.


--- Quote from: magic on September 19, 2019, 02:08:28 pm ---You need to know your source impedance to correctly choose between voltage noise and current noise.

--- End quote ---
The source is an ultrasonic transducer - high capacitance. I'll get the measurement soon.


--- Quote from: magic on September 19, 2019, 02:08:28 pm ---You also need to know your target SNR / tolerable nV/rtHz or some other number specifying how much noise you can put up with.
--- End quote ---
  SNR = 15dB



--- Quote from: Zero999 on September 19, 2019, 02:46:43 pm ---The bandwidth is only 20kHz, so I wonder how well a discrete amplifier would work?

--- End quote ---
Why is a narrow band a problem? Shouldn't it make things easier? Or is it the bandpass filtering the problem?


--- Quote from: magic on September 19, 2019, 02:08:28 pm ---A two stage solution would enable the use of slower amps and only one would be critical in terms of noise performance.

--- End quote ---
Sounds like a good solution. But, do I need to match the output impedance of the 1st opamp with the input of the cascaded one, or can I connect them directly?
Navigation
Message Index
Next page
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod