VLF MD with digital signal processing : Bee-Buzz 1

[Olly]

Very nice, but I'm not sure that an RX coil with such a low series resistance (0.5 Ohms) is practical.

[Aziz]

Hi all,

have a look.
I have adapted and arranged Dick's Ultra-Low-Noise Pre-Amp:
see source http://www.dicks-website.eu/low_nois...rt3/part3.html



This nice pre-amp with gain of 1000 and 9 V single supply operating voltage (9 V block battery), achieves with only 1x NPN (ZTX 815) an input voltage noise density (en)
en of 0.225 nV/rtHz at 1 kHz and
en of 0.214 nV/rtHz at 10 kHz

Not bad. With a cheap NE 5534A op-amp. Even with a 500 Ohm load on the output of the op-amp.
Now the critical conditions, which allow the superior noise specs:
Collector current set to approx. 9 mA
Source resistance is max 0.5 Ohm (RX-Coil with 0.5 Ohm series resistance)
Gain setting resistor RG1 is 2x1 Ohm parallel (=0.5 Ohm). Must be fixed at this value.
ESR of the capacitors must be each max. 20 mOhm.
The real gain setting resistor RG2 can be calculated according to the formula above (must be approx. 500 Ohm)
Bandwidth is well above 100 kHz. Enough for me.
Power consumption approx. 12-13 mA.

No LT1028 or AD797 required at all.


I have versions with 2xNPN and 3xNPN in parallel. These are a bit better. But the power consumption is too much for a 9 V block battery. I like the single transistor version. I don't need to go below en 0.22 nV/rtHz.

Cheers

[Aziz]

Hi friends,

I have lost a lot of time in investigating LTSpice noise calculation error. But I have found the reason finally. You have to set the operating point voltages on critical nodes correctly. If you have (large) capacitors in your circuit, the initial voltage on its node is very very important. If not set correct, you get wrong noise calculations in LTSpice.

BTW, the 0.2 nV/rtHz noise voltage density is very very difficult to achive with the NE 5534A op-amp. No, I'm not going to use other expensive op-amps (1nV/rtHz). This is not necessary.
Under very critical conditions, I can get 0.175 nV/rtHz at gain x1000 with the NE 5534A op-amp.
The gain of 1000 is too much I think. Gain of 200 - 500 would be ok I think. Let's investigate how much gain we really need. And then optimize the pre-amp for the required gain.
Me LTSpice
Cheers

[Aziz]

Hi Carl,

yep. I should have called it noise floor level instead of SNR.
When I reduce the block size to 128 (instead of 1024), the frequency bins integrate more noise as the bandwitdh of the frequency bins gets greater.


It is at least 10 dB more noise at each frequency bin.
Oh my god!, I can process only 1/3 of the frames (495 of 1500). 2/3 will be discarded as the process-thread can't make it as fast as the IO-thread delivers data.
But this is the old process-thread with 1001 features. The new process-thread with the Goertzel decoders will do that with ease.

Aziz

[Carl-NC]

Keep in mind what you call SNR isn't really SNR, it's a relative noise level and its value depends on the depth of the FFT. To get SNR, you have to integrate all those noise bins over the BW of interest. Here's an FFT or the AD6645, a chip I helped design. You can see the noise floor is around -120dB while the SNR is only -75dB. From this, you can calculate the size of the FFT to be 64k. Using a smaller FFT will raise the noise floor but the SNR stays the same.

[Aziz]

Hi all,

this is the simple loop-back test on Sound BlasterX G6 running with 24 bit resolution @ 192 kHz SR . Line-out has 3 kHz (left-ch) and 12 kHz (right-ch) signal, which is fed into Line-in with a stereo cable. So you don't see any RX signal. It is a basic loop-back test to check the SNR of the DAC/ADC system. You see, that in the frequency region up to 24 kHz, the average SNR is at -120 dB (this is quite good). Up to 48 kHz it degrades a bit. At approx. 54 kHz, there is an external noise source coming from PC I think. The SNR arrives -100 dB at approx. 75 kHz. The higher the frequency, the worser the SNR. I say it is a snake oil from Creative.



My block/frame size is 1024 samples (for FFT reason). I get 187 frames/s from the sound driver. And process all 187 frames. The display rate is reduced to 10 frames per second.
The blue line is the FFT spectrum. There is a thin blue line indicating the average FFT spectrum.
White trace is left input channel signal. Red trace is right input channel signal.

Sound Blaster G3 is much much much much worser!
Cheers,
Aziz

[Aziz]

Hi all,

my old Windows XP PC is now running too. I have full access to my old project files.
Again, problems with contacts and aging plastic. I had to remove all connectors from the main board and plug in them several times. And I had to clean and repair my CPU water cooling system. Everything done now.
I can focus to the software coding part. The glitch problem with G3 on Windows 10.
Cheers

[Aziz]

Hi all,

good news: The Li-Ion batteries on my Eee Netbook PC seem to work. It was apparently a production error: They have simply forgotton to solder the battery strips at the corroded connection.

There is an another Windows XP PC, which isn't working anymore too.
---

The most interesting part in this project will be the ground balance and discrimination. Note, that we have a lot of multidimensional information: Complex output Real/Imag or Magnitude/Phase, for each interested frequency bin. And we have two transmit frequencies at the same time. And we also measure the TX-coil as a reference signal at interested frequency bins (at least to get correct phase relation of RX to TX).

Why I want to use the dual frequency driven resonant TX tank?
- low Z (impedance) of the TX-coil -> we can push more current through the TX coil, TX-current will be amplified
- high energy efficiency -> energy will be recycled on resonant LC systems
- more info -> better GB and discrimination

This is going well beyond any simple VDI calculation. The GB and discrimination gets very easy, after I implement and show some visualisations of the calculated data. But this is a long way to go yet.
Cheers

[Aziz]

Hi all,

this is my Asus Eee PC 901. It's alive now.
I have attached an external USB keyboard.
I have changed the internal CMOS battery.
I have repaired the Li-Ion battery. The contact between two cells was totally corroded. I have cleaned it and put a clean washer between the two cells. The battery is charging now. I hope, the 6 (not 4) cells are still good enough. Anyway, I have to change them in the future.

Cheers

[Aziz]

Hi all,

no chance to get the Sound Blaster G3 to work on Windows 7. No drivers available. It isn't down compatible and won't work even at USB Audio 1.0.
(Creative, go to the hell!)

I have to wait for the Blue-Tooth keyboard. I can't work efficiently on the damn touch screen to work on the glitch problem on Windows 10 tablet PC and G3. And there is only one USB-port and this is occupied by the damn G3, so I can't attach an external keyboard.
I don't have an USB hub too.

BTW, my old Asus Eee PC 901 with Windows XP is out of order and must be repaired:
- the keyboard isn't working anymore
- CMOS battery is empty and must be replaced
- 4 Li-Ion batteries must be replaced too. They are totally damaged.

The Eee Netbook PC is a nice portable detector platform and I wan't to use it too. This is the further reason for, why I need the Blue-Tooth keyboard and numpad too.

So I want to check, whether my detector software or the hell's Creative driver software on G3 is buggy. I hope, that it's my fault so I can fix it easily.
Me Creative!

It is not a good day.
Cheers

[Aziz]

If you want a good, fast FFT 'C' library to use in your application, then I have used 'FFTW' in conjunction with QT under Windows with very good results...
It's a native 'C' library so should work with just about any processor, although it does have optimizations which can take advantage of various processor specific features.

https://fftw.org

Regards,

Thanks,

I know this site.

I also have many different implementations of FFT. Even for a realtime DSP with fixed floating numbers (using the efficient MAC opcode; I have coded this for over 20 years ago or so). Also an AI inference code for the DSP (neural network). Anyway.

I have put in the reference c-code a simple version for block sizes of power of 2 (in-place complex FFT variant). This is doing still a good job.
Aziz

[Olly]

If you want a good, fast FFT 'C' library to use in your application, then I have used 'FFTW' in conjunction with QT under Windows with very good results...
It's a native 'C' library so should work with just about any processor, although it does have optimizations which can take advantage of various processor specific features.

https://fftw.org

Regards,

[Aziz]

Hi again,

the glitch problem does seem to happen on the Sound Blaster G3 only. With Sound BlasterX G6 there isn't any problem. Could be a bad driver problem. Or my own software, which I should check this or at least solve this issue.

G6 is obviously optimized for up to 20 kHz output (audio range). Snake oil advertising.
Above it is getting quickly worse (up to 20-30 dB worser) and we have to rise the RX signals well above the noise floor level. This is the reason for the pre-amp need.

[Aziz]

Hi all,

I have purchased several parts for the next step:
- ZTX 851 (npn), ZTX 951 (pnp), ultra-low-noise transistors. Only these interesting parts are available from where I buy parts.
- enough enamelled magnet wire 1.0 mm diameter for the choke L1
- Blue-Tooth keyboard and Blue-Tooth numpad for my Tablet PC
and other parts.

Re: Blue-Tooth numpad
I'm going to remote control the detector software on my Tablet PC with this small numpad. The software will be adapted to work with this kind of keypad.

The software part is quite time consuming. I have to change a lot of things and re-implement the whole detector software. And I have a glitch problem on Windows 10 OS with the Sound Blaster G3. So I have to re-write some critical threads once again. If I am finished with these critical parts, it is time then to show some frequency spectrums. And decoded results. And how to solve the GB problem (yet another WBGB? ). And whether we need an ultra-low-noise pre-amp.

Cheers,
Aziz

[Aziz]

https://archive.org/details/the-art-...ed-2015_202008

Here is the link to a free download

Thanks!

See on pages around 500 and more!

There is a list of npn and pnp transistors with noise specs.
Long live Zetex & Diodes for making these transistors.
I will look at spice files for best candidates with lowest Rbb (base resistance).

Aziz