My design finally works!

[ivconic]

Nice job... and an interesting assembly approach. One question: Why didn't you omit the breadboards and move the microcontroller board directly onto the actual circuit board?

I thought I'd ask the same question... but in the meantime the answer came in post #30...

[Pep_T]

And here it is a view of the display. It shows the measured TX frequency, Q and I channels (whith their K and P indexes to minimizes them, controlled by the encoder), module and phase of the received signal. I am now working on the code level in order to perform better averages of the values in order to minimize noise. The magnitude is too noisy to use it as a trigger for detection, I am still using Q as a target detector.

[Carl-NC]

The magnitude is too noisy to use it as a trigger for detection, I am still using Q as a target detector.

That's what everyone does.

[Pep_T]

I tought I could do something a little more sophisticated, but it is not really important.

[Sun-Boy]

Hi Pep,
Just came across your design, it looks great, good job.
One way of getting a signal detection, if you want to get a little more sophisticated, is feed the I and Q into an RMS to DC converter, this should give you a true RSSI signal.
But if you want to keep it simple you can feed your coil return signal into a comparator with some hysteresis and a level adjust potentiometer, basically a threshold circuit.

[Sun-Boy]

Earlier you mentioned you didn't like the sound of the audio tone, are you using a square wave or distorted wave to drive the speaker?
If you make it into a true sine wave it will sound really good, you can make the square into a sine with a simple resistor/capacitor low pass filter, you can use LTSpice to simulate it, you'll lose some signal level but the tone will be a lot better.

[Sun-Boy]

I just ordered boards for my design, it's in this thread:

Creating New Prototype - Analog System

I'm going to experiment with higher frequencies, into the VHF.
Also experimenting with a Lock In Amplifier input to see how it will perform.

[Pep_T]

Hi Sun-boy,
I just took a look at your design and I am really impressed! The mechanical construction is far more detailed and organized than my device. Answaring to your questions:

1: I do not understand the "feeding I and Q into RMS to DC", since both I and Q are outputed from an integrator and they are basically DC. Can you elaborate more? I am interested;
2: It's not that I didn't like the tone per se, I did not like the way it was generated. It was a combination of two square waves, one linked to the I and the other to the Q signal. Right I just monitor Q for variation and beep a buzzer if the signal moves below or above a threshold;

I have still a couple of months before a well deserved pause from work will allow me to test (or rather, USE) the detector on the fields, I will use them to refine the detection sounds and triggers.

About your design, even if it is different from mine in terms of operating frequency, I actually have some questions but I read on the other forum where you originally posted that you do not want to share the schematic... such a shame, any chance you will reconsider?
Apart from being curios to understand how it works it might help with future versions of my own. Anyway, here's the questions I have, many referred to the image of the PCB you posted on your topic:

1. Why Ipex connectors for the coils?
2. In order to generate the TX signal, do you first generate it and then amplify it on another stages?
3. Do you perform some kind of nulling or adjustments on the RX signal?
4. What converters are you using for power supply?

Again, If you wish to not respond to these questions I will completely understand. If you want to know more about my design I posted the schematic on message #5, but attached to this message you will find an updated version.

[Sun-Boy]

Hi Pep,

No problem answering questions, I'm retired now, so I have time to work on my circuits.

About the I and Q, you can input the signals into a chip that is only an RMS to DC converter, like an LTC1966, this is my first time using it for an RSSI signal, but you should be able to get a dc level by combining the I and Q signals. I'm wondering if by only monitor either I or Q, that you will potentially get a phase response that will not give you very much DC depending on the metal that is being seen at the coil. By looking at both I and Q, you should get a 'real', RSSI.

You originally posted that you do not want to share the schematic... such a shame, any chance you will reconsider?
Maybe, we'll see if it works first.

1. Why Ipex connectors for the coils?
On the input to the receive amplifier, I wanted to keep the signal as quiet and shielded as possible, plus they are easy to disconnect, and also take up little room on the pcb.
On the transmit signals, I didn't know at the time that a shielded twisted pair would work better, so I'll try both cables and see which works better.
2. In order to generate the TX signal, do you first generate it and then amplify it on another stages?
Yes, in older designs I used a simple crystal and inverter oscillator at twice the operating frequency and then put the signal through a flip-flop to divide the frequency by two to get to my desired operating frequency, and have a true 50/50 duty cycle. I then go through a low pass filter to get a clean sine wave - I want to remove as much harmonics as possible.
On this new design I'm just using a 555 as the oscillator/vco, this way I can adjust it to different frequencies for testing.
Yes, I do use a transmit stage, I'm using an AD8397, I wanted to drive the coils with about 100mA, but am currently running a little over 50mA.
3. Do you perform some kind of nulling or adjustments on the RX signal?
Yes, I used to use the same nulling circuit that you are using in your system, but this time I'm using an all pass phase shifter and op amp dc level adjuster, I want to keep from loading the transmit signal, and buffer the output. Again, probably not needed, but I'm wanting to see how it will work.
4. What converters are you using for power supply?
We used these SEPIC devices on a project at work, LT3471, they work well, you'll need to put them in LTSpice to choose the proper inductors under load.
They allow you to adjust the outputs to higher voltages if needed, and run the batteries real low, lower than the output voltages operation.