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

[Geo]

Thank you.
I have the same problem but bigger, especially at "pinpoint mode".

[SimonS]

Ropcsika
Can you share with us your coil data? Resistant?, Induction ? How many turns, diameter of wire...?

[Ropcsika]

Ropcsika
Can you share with us your coil data? Resistant?, Induction ? How many turns, diameter of wire...?

DD coil, diameter 240mm
Tx: 45 turn, 1125 uH, 3,1 ohm, wire diameter 0,53mm
Rx: 220 turn, 25300 uH, 80,1 ohm, wire diameter 0,22mm

Tx and Rx frequency distance more than 1500 Hz

[SimonS]

Hmm we are close....
TX: 1238uH, 2.9ohm, wire 0.5mm (45 turns)
RX: 26700uH, 74.8 ohm, wire 0.2mm (200 turns)
possible I have problem, because I have only 200 turns on RX ????
what is mean "Tx and Rx frequency distance more than 1500 Hz"?

[Wiltran]

In terms of learning on demand "MOLE-2" and " 3-Mole "

1. Stitched controller program.
2. "Fee (Z80) + stitched controller" , board with the other hand . Assembly . sticker
+ Indicator
3. customized payment in collection (under body z80 )
4. "electronic control unit" .
5. KROT2 MD without the rod.
6. Other options ..


I got this off Andrew Mole k2 forum. What does it mean?: Is the firmware he is selling better than the freeware version?

Everywhere will stand the best version of the program, surpassing the "free" firmware in functionality!

[reiv]

I have read (using translator) over 140pages of threads about Kpot2 on Andrew`s forum and can`t find straight informations about how/what is the procedure to match all these parts marking "*" on schematic and not only them. I mean:
R3
R7/R8
R10
R11
R17
R18
R19
R21
R24
R25
D4
And they should be fitted before procedure "zero ADC" ? Is here anyone who knows this and can write here?

[sled]

I have read (using translator) over 140pages of threads about Kpot2 on Andrew`s forum and can`t find straight informations about how/what is the procedure to match all these parts marking "*" on schematic and not only them. I mean:
R3
R7/R8
R10
R11
R17
R18
R19
R21
R24
R25
D4
And they should be fitted before procedure "zero ADC" ? Is here anyone who knows this and can write here?

No guarantee that what I say here is correct, but here is an educated guess:

R3 is the output impedance of the speaker amplifier (MC34119) which works with a load of 8Ohms and up. The DC resistance of your coil and R3 should at least sum up to 8 Ohm, if it is less than 8Ohm you draw too much current from the amplifier. Also don't choose a too large (Ohms) resistor since the magnetic field is proportional to the current that flows through the coil. Since the current through the resistor can get large I'd suggest to use a 1W resistor.

R7/R8 form a voltage divider which divides the supply voltage (3.3V) by a factor of two (net is labelled U/2). This is a reference/bias voltage for the opamp where your RX signal is centered at. R7 and R8 should match each other as precisely as possible, if both are 10.3kOhm it's not a problem but if one is 10kOhm and the other one is 10.3kOhm the ratio of the voltage divider is not correct and you loose headroom for your signal. Use 1% metal film resistors if possible and measure them with a multimeter to get two resistors with almost the same value.

R10 is part of the op-amp feedback and adjusts the gain together with R12 (and C1, try to get both R10 and R12 as close to their nominal value as possible by using 1% metal film resistors and measure them with a multimeter.

R11 is optional I think.

R17 with R16 form a voltage divider to measure the battery voltage (low battery indicator), you probably need to adjust R17 if you use a supply voltage different from 4.8V. Again, if you want to measure your battery voltage accurately use 1% metal film resistors.

R18 is responsible for setting the analog reference voltage, the reference voltage is quite critical and should be as stable as possible. But I don't see any special marking on the schematic. If you have an oscilloscope, make sure you don't have much noise on AREF, AVCC and AGND. The digital ground and analog ground should join only at one point on your PCB and this joint is done using a (10uH) inductor as choke to block high frequency noise from the processor getting into the analog part. There are two distinct ground symbols on the schematic, one bar for digital ground, two bars for analog ground.

R19 adjusts the contrast on the LCD I think, the potentiometer goes to the VEE pin on the LCD, check the datasheet of your LCD module for parameter ranges.

R21 is for the background light of the LCD I think, check if your LCD display has a background light. R21 limits the current to the LED/background light I think - the LED pin on the microcontroller is pulled to ground to turn on the background light, so make sure you don't exceed the microcontroller's maximum ratings for input voltage/current on the pin. Look at the datasheet of your LCD module and microcontroller.

R24 sets the gain for the speaker, have a look at the Datasheet of the MC34119, if you need more gain because it's too quiet, increase R24 in value.

D4 is a protection diode (zener) which protects your circuit from overvoltage if the voltage regulator fails etc.

[reiv]

Thank You soe much Sled, for your replay
Ok soe:
R3-is clear
R7/8- to be more pratical. To proper mach them first measure voltage at pin40 (xmega) for example it is 3.3V, then measure voltage at pin 44 and it should be about 1.65V if it`s not change proportion of R7/8 (should be half) - it`s correct?
R10- this one don`t understand well. What should be proper gain?, You mean C18 of course
R11-optional for what?
R17/16-clear
R18- I know that there is no "*" but Andrew wrotes about this resistors to few people,and with Kpot3 there is no R18 just a 33uH. But ok, I should measure noises at those pins (40,39,38 ) to get them less match value of R18.
R19-clear
R21-clear
R24-clear
D4-clear
Thanks ones again

[sled]

R7/8- to be more pratical. To proper mach them first measure voltage at pin40 (xmega) for example it is 3.3V, then measure voltage at pin 44 and it should be about 1.65V if it`s not change proportion of R7/8 (should be half) - it`s correct?

Yes on the website of Kpot2 it says "Voltage at PIN 1 and 44 of the microcontroller should be about 1.6V!". Also measure it between R7 and R8 (I've marked it in red on the schematic here: ). Proportion is important, but 10k is a good value, if you have a handful of 10k resistors, pick two that match each other in value.

R10- this one don`t understand well. What should be proper gain?, You mean C18 of course

It's a non-inverting operational amplifier configuration, with frequency dependent feedback because of the capacitors C15 and C18.

See this as example:

Av is the gain of the amplifier. In our case we have a complex impedance R1 = R12 || C18 and R2 = R10 + C15.

This is an active band-pass filter with gain. At the center-frequency of ~10kHz it has about ~40dB gain.

Here is the Bode-Plot:

You can calculate this on your own using this web tool: http://sim.okawa-denshi.jp/en/opampkeisan.htm

R11-optional for what?

R11 serves no purpose I think and can be removed, maybe it is there to minimize the input offset so that both inputs of the op-amp see the same impedance, but not sure!

R18- I know that there is no "*" but Andrew wrotes about this resistors to few people,and with Kpot3 there is no R18 just a 33uH. But ok, I should measure noises at those pins (40,39,38 ) to get them less match value of R18.

Turn the schematic 90 degrees and look at it again, it's a simple RC lowpass filter see here:

Again you can use the awesome tool here: http://sim.okawa-denshi.jp/en/CRlowkeisan.htm

It has a cut-off frequency of fc = 15915.4943092[Hz] with a -20dB slope since it's a first order filter:

If you repace R18 with a 33uH inductor you get an LC low-pass filter, you can calculate it here: http://sim.okawa-denshi.jp/en/RLClowkeisan.htm (choose something like 0.06 Ohms for R)

[reiv]

Thank You for your time and usefull links, now i think know all

[ANDREW K2]

In the circuitry of a "*" says the nominal element can be selected when setting

[Geo]

Thank you.
I have the same problem but bigger, especially at "pinpoint mode".

Hi all.
I believe that i found the solution of my problems (not 100%).
I made another one pcb and i soldered a new uc. Detector has very good depth, good discrimination but when the sensitivity is lower than 5 then has erratic signals. From what i saw at the pcbs from Andrew, he use double side pcb with the one side to be at ground. I did not made the second side (ground).
So i found the solution...... i ordered a full MD from Andrew, completed with the coil.
When i"ll receive it i"ll inform you about the full ability of this detector.

[Ropcsika]

Concentric coil test

[Geo]

Why you use so low sensitivity ( for ferrous objects???

[Tepco]

Very nice. Did you try to compare concentric and DD response? (VDI stability, object orientation, range etc). I have built only DD but want to try concentric design. Unfortunately whole thing is not finished yet, not boxed, everything is still on working table.