Compass Yukon Coin Hustler II The Search Coil: How is it wound, what is the configuration?

Thread Starter

Tank Circuit

Joined Mar 5, 2017
28
I am getting interested in designing my own version of a digitally controlled metal detector. I know this is a daunting task, so I wish to start out with a very simple analog metal detector from days gone bye. I have this old reliable T/R (I/B) metal detector and I am in the process of reverse-engineering the circuit. I have drawn out the schematic and the only thing left is to determine how the search coil is wound. I am hoping that there is someone with a lot more experience on how manufacturers wire their search coils since I am hitting a snag.

If you know why I am getting the anomaly during my tests please chime in and assist me while I am building my replicant version.

There are 4 wires coming from the search coil, and after checking each one, I discovered that they are all connected to each. There is also a 5th wire present which is attached to the RF metal shield and it is open to all other wire connections. This coil’s 4 active wires consist of a red, green, black and white wire. After a comprehensive test of the coil’s resistance and inductance measurements, I concluded that it probably consists of two large coils and they are likely are wound in the “Double-D” configuration. I have added an insert showing how I visualize the various sections of the search coil are connected on the schematic diagram and I have added a chart showing the measured resistance and inductance on every wire (the search coil was out-of-circuit).

In the schematic, I show that the transmit coil section is made up of L2 and L3 and the receive coil is L1. I realize that measuring inductors with a DMM in resistance mode is not ideal, but during my tests, I found when testing L3 it was acting differently than what I would expect. I figure that the total resistance across all the coils would decrease a little if the wires of L3 were shorted together. When I measure the resistance between the white and the red wires, I get 9.6 Ω even if the green and black wires are shorted together or not. I checked every possible connection and read their resistance while shorting the other two wires together and the measurement never changed at all. This makes me question if I have the configuration correct.

I figured that perhaps L3 was wired in a delta or wye configuration. So I broke out the inductor tester and it shows that the total inductance drops like expected when I short out L3. The total inductance drops 40 µH from 900 µH to 860 µH. So now I have one meter showing that these coils are in series and the other disagreeing. I have assembled a simple graph showing the resistance and inductance between the various connections.

The two outside wires (red and white) naturally, have the highest resistance and inductance. I realize that if I add the sum of all the coils the total will not match. I also know that depending upon the direction on the winding it will either add or subtract to the inductance.

upload_2019-8-18_11-33-56.png

upload_2019-8-18_12-5-53.png

As I mentioned before. If I measure the resistance across the red and white wires, I get the 9.6 Ω reading. If I short the green and black wires together, the resistance remains the same.
If I measure the inductance across the red and white wires, I get the 900 µH reading. If I short the green and black wires together, the inductance drops to 860 µH.

upload_2019-8-18_11-38-10.png

The operating frequency is 100KHz pulsed every 3.540ms (282Hz)

upload_2019-8-18_11-48-54.png
Three pulses (282Hz)

upload_2019-8-18_11-50-24.png
One pulse expanded (100KHz)

upload_2019-8-18_11-51-33.png

upload_2019-8-18_11-51-53.png
 

Attachments

Last edited:

TeeKay6

Joined Apr 20, 2019
572
I am getting interested in designing my own version of a digitally controlled metal detector. I know this is a daunting task, so I wish to start out with a very simple analog metal detector from days gone bye. I have this old reliable T/R (I/B) metal detector and I am in the process of reverse-engineering the circuit. I have drawn out the schematic and the only thing left is to determine how the search coil is wound. I am hoping that there is someone with a lot more experience on how manufacturers wire their search coils since I am hitting a snag.

If you know why I am getting the anomaly during my tests please chime in and assist me while I am building my replicant version.

There are 4 wires coming from the search coil, and after checking each one, I discovered that they are all connected to each. There is also a 5th wire present which is attached to the RF metal shield and it is open to all other wire connections. This coil’s 4 active wires consist of a red, green, black and white wire. After a comprehensive test of the coil’s resistance and inductance measurements, I concluded that it probably consists of two large coils and they are likely are wound in the “Double-D” configuration. I have added an insert showing how I visualize the various sections of the search coil are connected on the schematic diagram and I have added a chart showing the measured resistance and inductance on every wire (the search coil was out-of-circuit).

In the schematic, I show that the transmit coil section is made up of L2 and L3 and the receive coil is L1. I realize that measuring inductors with a DMM in resistance mode is not ideal, but during my tests, I found when testing L3 it was acting differently than what I would expect. I figure that the total resistance across all the coils would decrease a little if the wires of L3 were shorted together. When I measure the resistance between the white and the red wires, I get 9.6 Ω even if the green and black wires are shorted together or not. I checked every possible connection and read their resistance while shorting the other two wires together and the measurement never changed at all. This makes me question if I have the configuration correct.

I figured that perhaps L3 was wired in a delta or wye configuration. So I broke out the inductor tester and it shows that the total inductance drops like expected when I short out L3. The total inductance drops 40 µH from 900 µH to 860 µH. So now I have one meter showing that these coils are in series and the other disagreeing. I have assembled a simple graph showing the resistance and inductance between the various connections.

The two outside wires (red and white) naturally, have the highest resistance and inductance. I realize that if I add the sum of all the coils the total will not match. I also know that depending upon the direction on the winding it will either add or subtract to the inductance.

View attachment 184185

View attachment 184194

As I mentioned before. If I measure the resistance across the red and white wires, I get the 9.6 Ω reading. If I short the green and black wires together, the resistance remains the same.
If I measure the inductance across the red and white wires, I get the 900 µH reading. If I short the green and black wires together, the inductance drops to 860 µH.

View attachment 184187

The operating frequency is 100KHz pulsed every 3.540ms (282Hz)

View attachment 184188
Three pulses (282Hz)

View attachment 184189
One pulse expanded (100KHz)

View attachment 184190

View attachment 184191
If no one else gives you info, I know that a Web search for ("DIY metal detector", or "Metal detector project") will give you a lot of info. I did such a search a few years ago but never built anything.
 

MisterBill2

Joined Jan 23, 2018
7,257
I am wondering about the coil and if it is really as it seems. But my very first question is does the thing respond to metal as it is? Is it really working? Can you see the actual arrangement of the coils? And does it look like anybody has "worked" on the system?
 

Thread Starter

Tank Circuit

Joined Mar 5, 2017
28
I am wondering about the coil and if it is really as it seems. But my very first question is does the thing respond to metal as it is? Is it really working? Can you see the actual arrangement of the coils? And does it look like anybody has "worked" on the system?
This is a perfectly working metal detector, I have owned it for decades and it is pristine. I don't use it much since I have several newer models that have better discriminators. It is a simple analog detector with a sealed search coil. I would need to rip it open to see if the two coils are concentric or in the Double-D configuration. Then I would probably have to start unwinding the coils to see which direction they are wound. It might help to determine if there are additional components tacked across any coils messing with my external readings.
 

MisterBill2

Joined Jan 23, 2018
7,257
OK, and thanks for the clarifications. There are ways to investigate the coils without breaking into the housing, since the system is functioning and it appears that you have a good scope of some kind.
What it looks like is that the two coils, L2 and L3, are a tapped coil in an oscillator circuit, and that there is a nulling action between the voltage from the wiper of VR1 against the voltage induced by the winding L1.
When a material that alters the phase angle between the two voltages enters the magnetic field of the coils the nulling effect is changed, and that gives the change in the amplitude of the resulting sum that is amplified.
So L2 and L3 are sections of the oscillator inductance and the physical arrangement is such that the magnetic flux coupling to L1 will be varied most by a material in the spot where the coupling is greatest. So this is a nulling type of detector. It may be that L1 is a separate winding with a different orientation from the oscillator coils, which would be functionally a tapped winding. Probably L1 is wound to produce a voltage out of fhase with the other coils.
Any digital alternatives will be on another post, in case this tab crashes.


Some postings on this site are relating to packages that have been damaged quite a bit and the original arrangement is unknown.
 
Top