Concept Design

[ivconic]

It doesn’t have conventional timers as such but has up to 16 PWM channels and 8 programmable PIO state machines which will probably be ideal for creating the dual or triple frequency TX pulses

I have Raspberry Pi 3 + for indeed long time I think. I got it just as it hits the marker, several years ago.
But believe me or not; I haven't played with it more than 20 minutes so far, overall.
I just installed Raspbian to check it and than forgot it in a drawer. It is real power hog.
Pico, however, is low power one and I think suitable for battery operated applications.
But that's all I know by now.
I have too many "toys" here and so not too much time to cover them all.

[Carl-NC]

- RAIL-TO-RAIL INPUT
High speed operation (38MHz, 22V/µs) make them ideal for driving sampling Analog-to-Digital (A/D) converters.
The OPA350 series operates on a single supply as low as 2.5V (up to 7V).

Flicker noise is too high (1/f knee is 10kHz) and supply is limited to 5V only. I need a 10V part and lower 1/f noise.

[ivconic]

Flicker noise is too high (1/f knee is 10kHz) and supply is limited to 5V only. I need a 10V part and lower 1/f noise.

So you actually do have something already in progress?
Decisions are made, so it seems.
Then why all this talking here?
You gonna supply RX part with 10v? +-/5V?
Ain't be good to supply whole RX part with single 0-5v supply instead?
OPA suits me just because of the single supply and range 0-5v and of course other good specs too.
I can use it directly with Atmega, fer instance, without thinking much on supply and signal levels.
Of course, this has nothing to do with ongoing project here, yet I am always planning "widely".
"It is good opamp, alright, how can I use it else?" - sort of thinking and planning.
And probably that OPA is the best I can find in local shops.

[ivconic]

Given that this will be an educational rather than a practical project for most of us; it wouldn't hurt if you could explain what you said.
"..Flicker noise is ..."
What is "flicker noise"?
"...too high (1/f knee is 10kHz)..."
How did you get that result?
Is that written somewhere in the datasheet?
Why isn't it good that 1/f knee is 10kHz?

[David_1]

if someone has part of the circuit already working, can it be posted so others can test it ? looking from the outside the metal detector does seem like a box of bits put together in many different ways.

what Parts have already been decided ?

[Carl-NC]

I've begun posting prelim schematics in the AMX-RX thread.

[Carl-NC]

Given that this will be an educational rather than a practical project for most of us; it wouldn't hurt if you could explain what you said.
"..Flicker noise is ..."
What is "flicker noise"?
"...too high (1/f knee is 10kHz)..."
How did you get that result?
Is that written somewhere in the datasheet?
Why isn't it good that 1/f knee is 10kHz?

Flicker noise (also called 1/f noise) is the excess low-frequency noise above the white thermal noise. Lots of components exhibit flicker noise, even resistors, but it's more prevalent and worrisome in opamps. You get it from the data sheet. A good data sheet will list the excess noise, usually in a 0.1-10Hz bandwidth, as in the AD797 data sheet:



Note that the 0.1-10Hz noise is in nVp-p, meaning it's fully integrated over that BW. The thermal noise (usually specified at 1kHz) is given as nv/rtHz, so the integrated noise depends on your designed BW. When the data sheet doesn't list a flicker noise spec it can usually be derived from the noise plot, which is almost always given. Here is the AD797:



From this I see the white (thermal) noise is 0.9nV/rtHz, and at around 100Hz begins to rise. 100Hz is the "1/f knee", below that flicker noise dominates. Here is the plot from the OPA350:



From this you can see the 1/f knee is 10kHz (bad) and the excess flicker noise is pretty high; at 10Hz it's almost 300nV/rtHz compared to less than 2nV/rtHz on the AD797.
When you look at the spec table it says:


The only integrated noise listed is from 100Hz to 400kHz, which is not very useful. In short, I consider this a very noisy opamp and not suitable for a high-performance preamp.

[Olly]

Carl have you considered the LT1028? Thermal noise is similar to the AD797 but the 1/f knee is typically down at 3.5Hz (Max 14Hz).

[Carl-NC]

Yes, if you look at the schematics I posted in the AMX RX thread you will see the LT1028 listed as an alternative. I started with the AD797 primarily because I trust its Spice model. With some opamps the noise modelling is just wrong.

[ivconic]

Flicker noise (also called 1/f noise) is the excess low-frequency noise above the white thermal noise. Lots of components exhibit flicker noise, even resistors, but it's more prevalent and worrisome in opamps. You get it from the data sheet. A good data sheet will list the excess noise, usually in a 0.1-10Hz bandwidth, as in the AD797 data sheet:



Note that the 0.1-10Hz noise is in nVp-p, meaning it's fully integrated over that BW. The thermal noise (usually specified at 1kHz) is given as nv/rtHz, so the integrated noise depends on your designed BW. When the data sheet doesn't list a flicker noise spec it can usually be derived from the noise plot, which is almost always given. Here is the AD797:



From this I see the white (thermal) noise is 0.9nV/rtHz, and at around 100Hz begins to rise. 100Hz is the "1/f knee", below that flicker noise dominates. Here is the plot from the OPA350:



From this you can see the 1/f knee is 10kHz (bad) and the excess flicker noise is pretty high; at 10Hz it's almost 300nV/rtHz compared to less than 2nV/rtHz on the AD797.
When you look at the spec table it says:


The only integrated noise listed is from 100Hz to 400kHz, which is not very useful. In short, I consider this a very noisy opamp and not suitable for a high-performance preamp.

Thanks! That's educational!


Tough, if you said "1/f noise" I would understand.
"Flisker" confused me. I could search the net, but it is always easier to understand when someone explains directly.
"Knee" I assume is the crossover point between the 1/f noise and the broadband noise ?
It is called the 1/f corner, "knee" is another confusing term.
...
So, correct me if I am wrong; if 1/f corner is at 10kHz and here the BW is 25kHz, obviously the 1/f falls down almost to center of BW and that's certainly not good, right?
What if the BW was ... say ... 9kHz? Then this opamp would pass easily, right?

[ivconic]

Thanks! That's educational!...So, correct me if I am wrong; if 1/f corner is at 10kHz and here the BW is 25kHz, obviously the 1/f falls down almost to center of BW and that's certainly not good, right?
What if the BW was ... say ... 9kHz? Then this opamp would pass easily, right?[/FONT]

Reason I am asking this... I can still make it with OPA and run TX at 9kHz (or lower than that) instead 25kHz.

[Olly]

Reason I am asking this... I can still make it with OPA and run TX at 9kHz (or lower than that) instead 25kHz.

Don’t confuse TX frequency with the bandwidth required to adequately amplify and sample low TC target responses. I believe we need to be looking at a bandwidth of around 1MHz or so.

[ivconic]

Don’t confuse TX frequency with the bandwidth required to adequately amplify and sample low TC target responses. I believe we need to be looking at a bandwidth of around 1MHz or so.

[ivconic]

STM32L496
1.71 V to 3.6 V
Core: Arm® 32-bit Cortex®-M4 CPU with FPU
4 to 48 MHz
Up to 1 MB flash
320 KB of SRAM
3× 12-bit ADCs 5 Msps up to 16-bit with hardware oversampling
20x communication interfaces
16 timers

Hardware oversampling allows 16 bits ADC resolution.
Not that I fully undrstand the mechanism of that...
But then separate ADC chip is maybe not needed?
And other specs makes STM32L496 a very good candidate here... or am I wrong?
...
And the sweet thing is this:
https://estore.st.com/en/products/ev...eo-l496zg.html

[moodz]

STM32L496
1.71 V to 3.6 V
Core: Arm® 32-bit Cortex®-M4 CPU with FPU
4 to 48 MHz
Up to 1 MB flash
320 KB of SRAM
3× 12-bit ADCs 5 Msps up to 16-bit with hardware oversampling
20x communication interfaces
16 timers

Hardware oversampling allows 16 bits ADC resolution.
Not that I fully undrstand the mechanism of that...
But then separate ADC chip is maybe not needed?
And other specs makes STM32L496 a very good candidate here... or am I wrong?
...
And the sweet thing is this:
https://estore.st.com/en/products/ev...eo-l496zg.html

The price is right ...

There might be a coupla gotchas ( see dithering section of Application Note )

an4629-adc-hardware-oversampling-for-microcontrollers-of-the-stm32-l0-and-l4-series-stmicroelectronics.pdf

moodz