'KROT2-XM' (mole), new project Xmega32

[reptooyep]

Hi,
First, thank you blobber for your advices.
The datas that you gave me are exactly my first setup. (Tx=50 turns 0.5mm and RX= 200 turns 0.15mm).
I put them back into my coil shell and put 3X4.7nf in parallel as RX tank and now, it works !!!
I don't realy understand why cause i didn't do anything else than with my first setup.
I'm not sure but i think cheap china ceramic capacitors don't do the job. I put old ceramic caps from the 70s and now it's fine.

If i can give one advice: For caps, cheaper is not a good choice. Prefer old fashion caps or good quality.

Now, everythin works but i have since few days a sound comming from the HP. It's a 2 HZ click with a constant volume level which stop when the detector circuit works. I changed all the ICs but it's the same. What can do such a sound in the circuit .
I'll try to take a video of the strange sound later ..

[blobber]

Yes I also have in RX tank 15 nF for C12, not need C0. I prefer foil caps, MKT are good and for TX too. In the beginning I also used ceramics, with bad results.
Your click sound is always, also if you reduce the gain and sensivity? Yes show a video!

[reptooyep]

I have a 15nf C12 (cheap chiness ceramic) but also a 15nF C0 (Good quality old fashion cap). With this configuration, i can reach a 20mv nulling.
When i take the C0 off, i have 4600mv displayed. the coil have been nulled with this setup. Maybe i have few extra turns on my RX coil but there is a big difference in inductance measurement between the construction step and the shielded finnished coil.
With the C0 and the last firmware, the detector works well but the detection depth is not the best. This afternoon, i found 2 coins in a hour (5 holes) .

[blobber]

This is not bad my friend, well done.
As I started with experimenting Krot, I also used C0, in RX tank.
But I think, it is difficult to make calculation for right offset 1500 Hz.
I also have noticed, an increase of inductance in finished coil. That's why I decided, to trust the measurement with oscilloscope and not my LC- meter. To check this, I made an easy oscillator, as I wrote earlier. With him and the values of inductivitys which I should actually have, I found the proper capacitors in TX and RX tank. In my case are the values:
TX 238 nF (220 nF nominal) 9,836 kHz; 1,1 mH
RX 16 nF (15 nF nominal) 8,333 kHz; 22,8 mH

[reptooyep]

Hi friend,
Thanks for your advices.
Later, i'll make a new coil and tune it with your oscillator and a frequency metter. In my configuration, i must have a different offset, that's why i need a tank C0.
I notice that the project works good with USB cable.. sure it works better with double shielded pairs cable but that don't explaine my poor detection depth.
I can reach 250MA on TX but i can't tune the MD. With 90mA and autotune, i get 30mv, which is not bad i think but the ground ballance function does not work (i get an error) but it works well in wood and in sand...
A piece of foil gives me bad result (ok), a coin gives me good results (ok) but junk iron gives me good results (Not OK). I should try with more tones (i tryed only 4 tones ..)
There are many menu that i did not understand, i have to explore that ..

[blobber]

Is your auto setup for ferrite working in the menue? Andrews video in Youtube is helpful, how to do. Ground balance adjust with ferrite is not working, or with real soil?
Small pieces of iron (nail, pliers, small hammer) gives a bad signal but big iron parts gives sometimes a good signal. I had dug a old and rustyspade on the beach and thought I had found the end of the rainbow .

[blobber]

For my DIP version, I use one USB cable. In my SMT version I have two separated USB cable. This setup is only less better, than with one cable.
I also can reach round 30 mV for 90mA, with both versions, when I use auto balance and adjust manually. When I increase gain to 4, I have less or no false signals and it detect coins 35 cm.
With more gain value (8 or 16), both detectors can reach coin until 45 cm in air, but I have a lot of false signals and this setup is not usable. Also not, when I increase threshold iron or no iron.
Well, I know 35 cm is not too bad, maybe with new firmware in the future it will be better. It is easy to change the firmware and setup too, I realy like this detectors.


Today I have made a little experiment with Krot SMD. I connected with a separated plug, only TX coil to Krot, than I was going to auto adjust frequency in menu, to find the right one, by pushing *0*. Of course he found the same frequency as before (with normal connected coil RX and TX).
Than, I have connected RX coil with 16nf (15nF nominal) in series, to TX input, only for this test of course. Krot also found the right resonance frequency in this tank. So it is easy to find the right offset relationship Ftx>Frx. But please note: I have the TX capacitor in my coil and not in circuit, to change the caps. I have there a small maintenance flap in all my coils, to change caps or wire.
With cap in the pcb (TX tank), it will not working. Only the 15 nF for Rx is my pcb.

[CDI]

I saw one thread on Polish forum where some guys attached coils from Tesoro Cibola/Vaquero and from Garrett Ace series. Look promising judging by test's. Next week i'm going to Vienna and i will buy one 12 x 15 butterfly SEF Detech.

[reptooyep]

Hi there,
I was dreaming of a software improvement for the krot2 tuning...
Because my coils do not have exactly the good inductance, i use a tank cap on RX.
I noticed than when you tune the TX, it finds the frequency with the biggest signal for TX.
So, i hask myself why not searching the less RX signal instead ?
What do you think of this ?

[blobber]

A new firmeware is available on Andrews home page.
Here you can find a short air test video, with my SMT version and the K2-G-S6 firmeware.


https://youtu.be/BxEUwPnf9CI


...not bad results, new sound and discrimination for bottle caps. Thank you Andrew

[moscadondres]

Hi blobber,
Can i know which frequency you use with the s6 firmware,best regards.

[blobber]

Hi blobber,
Can i know which frequency you use with the s6 firmware,best regards.

10kHz

[moscadondres]

Hi blobber
Thank you for reply but Andrew says:8.3kHz,regards;

[blobber]

Yes, but as you can see, it's also working fine. The auto frequency adjustment, will find the proper frequency, which you have chosen with your coil and capacitors.

[moscadondres]

Thank you.