Non motion detector

[ivconic]

Murphy always jokes with me, I always find data and info easily for what I don't need... and very hard for what I really need!
An acquaintance from the area has a Coinmaster 4/DB (for God's sake, how many different Coinmasters are there???)
and a B5 (or 5B) coil with a detector.
Exactly what doesn't suit me... and I don't need it!
But I noticed that the coil has the mark "B5" stamped on the left "ear".
The same coil is used by the 6DB and 6000 models (!!!???)
And here is the exact data for that B5 or 5B coil
(which means nothing to me because I need a B4 or 4B for my Coinmaster, because
the 4B and 5B coils are very different ... at least it seems so ...)
...

[Davor]

I think the channel switching detectors are the perfect candidates to be converted into correlated double sampling mixer/amplifier combos. You have switching/mixing already, so a bit of a CDS added to it should be easy, cheap, and synchronous.

[ivconic]

Have I posted this sketch before? Not sure.
I am having problems with memory lately...
Here is what I measure at Quest coil:



It looks like it is center tapped too.
Would be good to take an advantage of that situation and design TX arround it.

[Skippy]

The heat is clouding my mind ... yes, it's 222, 2.2 millihenry. And its impedance is 180 Ohms at 13kHz ( not 5 kOhm, don't know what buttons I pressed on my calculator to get that figure... )
I've edited the above post.

[Altra]

Quote:"Skippy mentioned Teknetics/Fisher coils. ...The Fisher F75 Tx oscillator using a mono loop actually boosts the output more than 2:1. So this circuit might interest some experimenters for their designs. I do not understand the math behind it's tuning and would be interested to hear how it works"

I think I understand the general principal. Just as you can take an L-C tuned circuit, and drive it via a tapping on the L, you can do the same function by 'tapping' the capacitor. In practice this means using two capacitors : the equivalent capacitance of two caps in series ( 1/C = 1/C₁ + 1/C₂ ) gives the tuning capacitance, and the individual impedances ( X = 1/(2*pi*f*C) ) determine the tapping ratio.

In the F75 implementation, the tapping-point is 50% , the two caps of 470nF behave as a 235nF cap.

When driving the circuit with a unidirectional switch to +5V, it's not going to work if you just switch to the capacitor tapping - a current path to 0 Volts is needed, which is provided by the 2n2 choke. Then when the switch turns off, there's a back-EMF from the choke, which kicks the tuned circuit the opposite direction. The choke value is high, 2n2 has an impedance of 5k at 13 kHz, so it doesn't affect the main tuning circuit much.
I don't know the purpose of the extra small cap across the coil ( 3n9 apparently? I 've never measured it ) , it's in parallel with the main 235nF, so should change the tuning C to ( 235 + 3.9 ) nF, which resonates with the 695 uH coil at 12.35 kHz. Why not at 13 kHz ? I've never simulated it, maybe with the TX coil resistance included, and the choke R & L included, it gives a better mathematical match.

Attached schematic:

Thanks for your input Skippy. The choke is 2200uh not 2n2. Your drawing explains a lot, I've messed with tapped coils before to increase the voltage. These act as an "auto transformer" boosting the output.

[Altra]

It's been almost a year since I was messing with this, so not so fresh in my mind. I used the axle style inductor from ebay.( see link) Also a photo of my test board.
Like I said in my post above, not sure how it works or is tuned. I used the simulator to adjust parts L1, C2, C3, C4. Pick your frequency and tx coil inductance and adjust the other values to get your desired output.

https://www.ebay.com/itm/233343902751?hash=item365461641f:g:w0cAAOSwKNliPkRq&amdata=enc%3AAQAHAAAA0FnQwcnohwpM4g9UzngCCForedEOPaqU2GwxVYn%2BNtjdXFonp262fhuciDD7HQ5UGis555Z7Dnbu1JRyYZVcdwXNOl4HLQc5gy6P1xHyEXUKP2bGZws4TgfL51CB94mCMOlXpXH9Bjk9ROdGYzWqDUpCiPwgds6uEPHPZdHoBahhdjeopXxjIemtS6UDgIAlD5xXuW6Wqu0xYcooMbY26%2FPPVaFjY9m5cvyuRqNfBDc4xH2E7tz7SavKQqidQqIvZoPDA%2BpTZBDwMDbFEtaf%2BgA%3D%7Ctkp%3ABFBMxM-si8Fg

[Skippy]

Quote:"Skippy mentioned Teknetics/Fisher coils. ...The Fisher F75 Tx oscillator using a mono loop actually boosts the output more than 2:1. So this circuit might interest some experimenters for their designs. I do not understand the math behind it's tuning and would be interested to hear how it works"

I think I understand the general principal. Just as you can take an L-C tuned circuit, and drive it via a tapping on the L, you can do the same function by 'tapping' the capacitor. In practice this means using two capacitors : the equivalent capacitance of two caps in series ( 1/C = 1/C₁ + 1/C₂ ) gives the tuning capacitance, and the individual impedances ( X = 1/(2*pi*f*C) ) determine the tapping ratio.

In the F75 implementation, the tapping-point is 50% , the two caps of 470nF behave as a 235nF cap.

When driving the circuit with a unidirectional switch to +5V, it's not going to work if you just switch to the capacitor tapping - a current path to 0 Volts is needed, which is provided by the 2m2 choke ( marked 222, meaning 2200 microhenry ). Then when the switch turns off, there's a back-EMF from the choke, which kicks the tuned circuit the opposite direction. The choke value is high, 2m2 has a high-ish impedance at 13 kHz, so it doesn't affect the main tuning circuit much ( I assume ).
I don't know the purpose of the extra small cap across the coil ( 3n9 apparently? I 've never measured it ) , it's in parallel with the main 235nF, so should change the tuning C to ( 235 + 3.9 ) nF, which resonates with the 695 uH coil at 12.35 kHz. Why not at 13 kHz ? I've never simulated it, maybe with the TX coil resistance included, and the choke R & L included, it gives a better mathematical match.

Attached schematic:


(edited choke value error )

[ivconic]

I am attaching proteus simulation for the double tuned Tx. It is producing 110vpp across the tx coil at 13khz. It probably not practical to use this high of voltage in a detector, but in some applications 20v - 50v may be useful. Series resistance of the coil will make a difference so you want to keep it low.

Wow! I am gratefull to you for sharing that!
Thanks a lot!
Since I have the coil I mentioned so many times so far, the one for Quest.
And I would like to engage it in some good design.
Why? I think I explained once; Quest X5 with such tiny and small coil performs unexpected well!
Extra deep and yet coil is so small.
That's what attracted me and I decided to keep it.
...
The choke 2200u? In real life; shall I inculde it in circuit?
What type of choke it may be?
There are chokes and chokes. Small ones, resistor look a likes, for example.
Also some bulky encapsulated in feritte "pots" etc.

P.S.
The other day i ordered two bathces of pcbs at PCBWay, one batch original TF900 redraw and another is CM5000 D/GEB but my "design", smaller one,
I think i posted it here; did I? Just checked; yes I did, under the #122 post. That's the one.
I picked "FedEx International Priority" method and those should arrive within 2 weeks.
So... until than.
I can add small "piggy" pcb on CM5000 pcb with your TX and see the results.
Question is; shall I change anything at RX side than?

[Altra]

I am attaching proteus simulation for the double tuned Tx. It is producing 110vpp across the tx coil at 13khz. It probably not practical to use this high of voltage in a detector, but in some applications 20v - 50v may be useful. Series resistance of the coil will make a difference so you want to keep it low.

[ivconic]

I was away and could not respond earlier. Skippy mentioned Teknetics coils. Last year I was working on a 5v vlf circuit. I wanted to boost the tx voltage. A standard Colpitts oscillator will generate around 7.5vpp with a 5v supply. A center tapped tx coil will do about 12vpp. After looking at some commercial designs, the Fisher F75 Tx oscillator using a mono loop actually boosts the output more than 2:1. I bread boarded it, and was amazed when properly tuned at the output voltage, a 12v supply could generate 80vpp! So this circuit might interest some experimenters for their designs. I do not understand the math behind it's tuning and would be interested to hear how it works

Attached is a schematic for 13khz. All the values below Q1 can be changed to affect the tuning and vpp output, The 2200uh choke is a 1/2w from ebay nothing special, other value will work also.

[Altra]

I was away and could not respond earlier. Skippy mentioned Teknetics coils. Last year I was working on a 5v vlf circuit. I wanted to boost the tx voltage. A standard Colpitts oscillator will generate around 7.5vpp with a 5v supply. A center tapped tx coil will do about 12vpp. After looking at some commercial designs, the Fisher F75 Tx oscillator using a mono loop actually boosts the output more than 2:1. I bread boarded it, and was amazed when properly tuned at the output voltage, a 12v supply could generate 80vpp! So this circuit might interest some experimenters for their designs. I do not understand the math behind it's tuning and would be interested to hear how it works

Attached is a schematic for 13khz. All the values below Q1 can be changed to affect the tuning and vpp output, The 2200uh choke is a 1/2w from ebay nothing special, other value will work also.

[ivconic]

[dbanner]

The schematic Carl posted above seems to be another version with a different coil.
The 5000D geb schematic that I am familiar with shows a standard concentric coplanar coil which we're all fond of diy.

[dbanner]

The inductance of the original 5000D geb tx coil is 2.9mH or thereabouts. The parallel capacitance is 330nF plus a second parallel cap for fine tuning.
It's all in the patent 4,293,816 " balanced search loop for metal detector". You can also probably work out the RX inductance since there is a 27nF capacitor across it. Around 35mH. That's assuming that the RX is not tuned "off resonance".

[ivconic]

Slightly modified, "improved" and good for further experimenting.
Meter scale is with center position and with adjustable sensitivity.
Will see...