Electronics > Beginners
Piezoelectric hydrophone low noise amplifier design questions
<< < (5/10) > >>
Sparker:
Feeling extremely unprofessional now :palm:
I intended to use the protection circuit below(1st attachment) at the amplifier input. It's an inherited variant used with other sonars here. I didn't add the 1k resistors to the simulation, but it turns out that a 1k resistor generates 4 nV/sqrt(Hz) of noise(and two of them give sqrt(2) times more), about the same amount as the op amp itself, assuming a 1:1 transformer. It kind of nullifies our efforts to make a low noise input stage.

Thinking of it, I decided to take a better look at the photos of the commercial Starfish sonar board(link) I made some time ago in hope to get insight of how this is properly done.
It's quite an interesting device. They emit 430kHz-470kHz CHIRP pulses and do DSP with an FPGA. The receive path is made of two THS4130 fully-differential op-amp stages, followed by a LTC1403 14-Bit 2.8 MSPS ADC. Probably they don't do analog bandpass filtering, but I didn't analyse the circuit enough to be sure.
The first stage opamp configuration appears to be an inverting amplifier with 220 Ohm input resistor and 47 kOhm feedback resistor, giving 46 dB gain. See the picture below(last attachment) for the recovered schematic of the input.
What I don't quite understand is why is their amplifier's input impedance so low(220 Ohm) while we typically want higher input impedanse for such a sensitive device.
Marco:
AFAICS these small thickness mode resonant fish finder piezo discs have an impedance of ~10 Ohm at resonance, so even 220 Ohm is probably high enough.

With a FDA input stage you're limited in options for low noise. Can't increase input impedance without increasing noise, so it's always a bit of a compromise. There are other architectures for low noise differential amplifiers which avoid this, but not in IC form (closest would be AD8129 but that's not exactly low noise).

Personally I'd wind a transformer for ~10x voltage gain using a nice expensive toroid and use a discrete input stage like pic (for high amplification it would need to be modified a bit, more like the one from Vojtěch Janásek which I mocked up in the second pic, or if you want you can parallel BF862s like GK described here on the forum, but I doubt you'd get away with 100x amplification in that case). Connect it to the fishfinder piezo and hit it with a hammer to see if the input protection is good enough.

PS. the input impedance of the starfish inputs is 1220 Ohm, not 220 Ohm. The 1K is simply in series with the 220 Ohm resistor as far as the signal is concerned. Don't let the coupling capacitors and diodes fool you.
David Hess:

--- Quote from: Sparker on April 27, 2018, 01:39:02 am ---I intended to use the protection circuit below(1st attachment) at the amplifier input. It's an inherited variant used with other sonars here. I didn't add the 1k resistors to the simulation, but it turns out that a 1k resistor generates 4 nV/sqrt(Hz) of noise(and two of them give sqrt(2) times more), about the same amount as the op amp itself, assuming a 1:1 transformer. It kind of nullifies our efforts to make a low noise input stage.
--- End quote ---

That is one of the reasons I said earlier that you cannot go wrong with a FET input operational amplifier; the series resistance added by simple input protection circuits raises the noise level.  Using a lower noise bipolar operational amplifier requires something more complex than a series resistor and shunt, some type of transmit/receive switch perhaps.

It is possible to make low impedance switches and overdrive protection circuits using diodes.  Perhaps easier in this case is to put a pair of depletion mode power MOSFETs back to back with a source resistor to set a current limit in place of those 1k resistors.  Below the current limit, the resistance is just the source resistor and MOSFET channel resistances.  Above the current limit, the channel resistances increases with voltage to product a constant current.

The diode method uses no resistors at all but is pretty esoteric and may produce too much noise.  I have only seen it used in 50 ohm RF systems where this was not a problem.  Check out the Tektronix 7A29 vertical amplifier and 485 oscilloscope for examples where this was used.


--- Quote ---What I don't quite understand is why is their amplifier's input impedance so low(220 Ohm) while we typically want higher input impedanse for such a sensitive device.
--- End quote ---

The piezoelectric hydrophones have a low impedance to start with because they are so large.  The impedance translates into an optimum ratio of voltage noise to current noise at the input to the amplifier.  That is why a bipolar input amplifier with lower voltage noise but higher current noise is more optimal for the examples I gave but this can be offset by raising the impedance with a transformer to take better advantage of the higher voltage noise but almost zero current noise of a FET input operational amplifier.  It is also easier to implement protection circuits for FET input amplifiers which is an overriding concern if the design is to be kept simple.

In the example schematic there is another reason; the amplifier will be very unhappy with a capacitive source on its inverting inputs (two of them for the shunt feedback differential amplifier shown) without series resistance.  I really think that is the wrong way to go about this; at the very least, a non-inverting differential amplifier should be used which is the first half of the 3 operational amplifier instrumentation amplifier.  That avoids the whole problem.  The same circuit can be built with a pair of low noise JFETs or even better, a transformer and singled ended amplifier as we discussed earlier.

Does the piezoelectric hydrophone's connection really need to be differential or could one side be grounded?  I have never messed with one.
Marco:

--- Quote from: David Hess on April 28, 2018, 11:37:58 pm ---The diode method uses no resistors at all but is pretty esoteric and may produce too much noise.

--- End quote ---

The diodes in the second schematic I posted and in a simple non inverting amplifier between the +/- inputs of the opamp are bootstrapped to nearly the same voltage, they can be much chunkier than the diodes to the rails the FDA has because their effective capacitance is reduced. The TVS/MOV to ground is in parallel with a very low impedance (1 Ohm in this case) so at 200 kHz that's pretty much irrelevant. A MOV is probably fast enough for any spikes coming from the piezo, so that would be the preferred option.

You could put chunky diodes between the +/- inputs of a discrete instrumentation opamp configuration as well of course. You can't realistically get close to the noise performance of a transformer based design though and if you use a transformer you can use a simple single ended amplifier, because the transformer takes care of common mode noise. In simulation the second schematic has a noise level of 0.86 nV/rtHz with a series resistance on the input of 100 Ohm (set to be noiseless, only amplifying current noise). This would suggest 0.086 nV/rtHz with a 10x transformer, assuming the transducer has about 10 Ohm impedance at resonance ... getting there without a transformer seems almost impossible.
Sparker:
Thanks for the help again, Marco and David!  :)

Marco, it's quite a mystery to me how the JFET circuit you've mentioned works. I understand that the op-amp helps the JFET's biasing, but I don't understand the AC behaviour of the circuit.
The problem with your diode based protection is that, maybe I didn't mention it good enough, the antena works both in TX and RX modes, so when we are transmitting the impulse, about 50...100 volts will be applied to the protection circuit, thus it should have some kind of series protection.


--- Quote ---PS. the input impedance of the starfish inputs is 1220 Ohm, not 220 Ohm. The 1K is simply in series with the 220 Ohm resistor as far as the signal is concerned. Don't let the coupling capacitors and diodes fool you.
--- End quote ---
Agree. I virtualy moved the 1k resistors to the generator side in my mind when I was typing this.


--- Quote ---It is possible to make low impedance switches and overdrive protection circuits using diodes.  Perhaps easier in this case is to put a pair of depletion mode power MOSFETs back to back with a source resistor to set a current limit in place of those 1k resistors.  Below the current limit, the resistance is just the source resistor and MOSFET channel resistances.  Above the current limit, the channel resistances increases with voltage to product a constant current.
--- End quote ---
Thanks, I think it's a brilliant idea! I will try to investigate this possibility with the depletion MOSETS I can buy here.


--- Quote ---Does the piezoelectric hydrophone's connection really need to be differential or could one side be grounded?  I have never messed with one.
--- End quote ---
It can be grounded, but concluding from what I've read about differential signaling, it offers better noise reduction. Even the designers of the starfish device use differential signals: for each channel there are two wires and a grounded shield.


--- Quote ---In the example schematic there is another reason; the amplifier will be very unhappy with a capacitive source on its inverting inputs (two of them for the shunt feedback differential amplifier shown) without series resistance.
--- End quote ---
Indeed, as the amplifier's gain is -Z2/Z1, having a capacitor as Z1 in the denominator just asks for a trouble.  :-BROKE
Navigation
Message Index
Next page
Previous page
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod