Flat Spiral wound....Is my method sound?

[Ism]

Hi all, been a while. I decided to build a flat spiral wound (nugget hunting) coil for my TDI.
I am hoping to get some input regarding the construction method I am using.

Using 330/46 Litz wire, and spray adhesive (like contact cement), I wound my coil on a 2mm thick corrugated plastic disc to fit an 8" Hayes housing.
Then I sprayed the coil and another disk of corrugated plastic. Sandwiched the windings and sealed the ends with hot glue.

Next I arrayed wire filaments from some 7 strand #28 silver plated wire on both top and bottom of the coil assembly.
I know most people use one or maybe 2 filaments, but I chose not to cut any strands and array them as to get the most contact both top and bottom.
I will be covering the entire coil assembly with graphite spray.

At this point I am wondering if my shield filaments are acceptable or if I have too much. I have built a number of standard bundled wire coils, but not a flat spiral.
Any advice is appreciated.

[ODM]

No comments on coil geometry, there seem to be so many approaches toward it both by hobbyists and commercial companies alike.

As long as the filaments do not form any loops in the coil normal plane it should be a good way of connecting to the shield. Maybe not painting the litz wire itself with graphite but having some sort of cover such as a paper circle over them to minimize the amount of graphite that gets in direct touch with the coil proper. Sprays are still liquid, and they tend to soak into bundled things (like litz wire) by capillary effect and create a large surface area.

One thing to consider is solder blob size when connecting litz wire. In a previous thread about the minelab patent on masking large solder joints with ferrite beads it was also determined that having multiple smaller solder joints would be one method of getting around large solder joints with litz wire.

What about cable type, and the method of connecting the coil and screen to your detector? Where are you planning on connecting your screen, do you have a braided litz screen for your coil lead? Connecting a coax or other cable that generally has a thick circular continuous shield can add a false target for your coil. Minelab's litz coils have a proprietary cable type that has a litz-arranged shield to avoid picking the shield up as a false target.

[Ism]

Thanks for your input ODM, You gave me a few things to think about/research. I did read about the concerns of large solder points for the litz wire but haven't considered the connection to the shield.

My plastic cardboard should protect the litz coil from any graphite spray. I planned on covering the litz leads before painting.

One additional question. I know that with scotch 24 I must leave a gap in the shield. Is that the same for graphite?

[homefire]

I'm thinking the Plastic Cardboard may be more Space then you should have. Using Scotch or copper foil thats not so good. Using Gaphite top and Bottom Plates of shield I don't think the Gap is Needed. If your shielding the Coil it's self , Yes you need the Gap or you have a Transformer Winding. Shorting out the return signal.

[baum7154]

The spiral coil while quite fast in its unshielded form rapidly loses its speed when shielded because all the windings are evenly and consistently exposed to the coil shield raising undesirable capacitance. For this reason I would not double the undesired capacitance by placing a shield layer on top of the coil. Additionally I would only use a graphite shield starting with a resistance of 2K ohms per inch and adjusting it either higher or lower in resistance based on if the detector sees your hand through the shielded side with the detector set at minimum sample delay. If it does then lower the resistance till it does not. Also space the shield away from the coil perhaps starting with 3/8" and test to see if it is effective over damp or wet soil. If so, then see if the spacing can be increased and still have an effective shield perhaps a max of 1/2". If the shield isn't good at 3/8" then increment it closer to the coil till it is effective. Closer shield/coil spacing slows the coil down and lower graphite resistance can also diminish sensitivity. Just my 2 cents worth.

Good luck on your build.

Dan

[crane]

A coil that isn't shielded on the top can be almost useless for nugget hunting because it will give a signal under every bush and under tree roots and some logs. It must be shielded top and bottom.

[baum7154]

In that case the only option is to minimize capacitance with spacing.

[ODM]

A gap is necessary for low resistance shielding that is close to the coil, parallel to its field like a "sock" around the winding, for example when shielding coils by alu-foil or tape, to avoid making a shunt turn on the coil.

For high resistance shielding, like most graphite papers and thin graphite paint, this gap is not necessary as seen by the shielding on several commercial fast coils. I find it very hard to believe a lack of gap would be a cost-cutting feature especially with the graphite paper shielded coils.

Why? If you leave a single gap, there will still be at least one half of the shield area in a single conductive potential target which is massive compared to a coil. Eddy currents do not form only around the longest outer path, this is why rings have relatively better response than plates/coins. If the resistance is not high enough and requires a gap in the shield, it would be just as well to have several gaps in a more complex branching pattern to limit the individual eddy current forming areas?

Ultimately it depends on the detector and coil type, so there is no straight easy answer for the experimenter except one; "don't pot it before testing" ?

[Ism]

Thank you all for your insightful responses. I can make the gap between the coil and graphite shield 5/16 inch (8mm) top and bottom and still fit it in the coil housing.

[baum7154]

Thank you all for your insightful responses. I can make the gap between the coil and graphite shield 5/16 inch (8mm) top and bottom and still fit it in the coil housing.

--------------
it is worth a try and will be very interesting to see the Self Resonant Frequency of this coil.

Dan

[Ism]

Hi Baum, I didn't have the 1pf cap that is in the fastcoil guide, The lowest I could find was 10pf so I'm not sure if that affects the resonant frequency or not.
That said, I have resonance at 1.270Mhz unshielded.

[baum7154]

Hi Baum, I didn't have the 1pf cap that is in the fastcoil guide, The lowest I could find was 10pf so I'm not sure if that affects the resonant frequency or not.
That said, I have resonance at 1.270Mhz unshielded.

-------------------------
You can make a 1 PF gimmick cap with 2- 1" long pieces of insulated wire twisted together 3-4 times. Did you have any feed line on the coil in your test? That is a good SRF for a fast coil.

Dan

[Ism]

Dan, No, I didn't have a feed line (coax?) attached to the coil. I made the gimmick cap and now have self resonance at 2.27 MHZ.
That is, on my scope the waveform jumps to 3-4x the p-p voltage at that frequency. I have a harmonic spike around 759Khz but it is just a slight increase in p-p voltage.
That seem pretty high for self resonance on a flat spiral?

[WM6]

Hi Baum, I didn't have the 1pf cap that is in the fastcoil guide, The lowest I could find was 10pf so

If you have double sided PCB laminate Fr4, thickens of 1.6mm, then cut 44mm square piece of those double sided laminate, solder pin wire on both side and you have 1pF capacitor (you only need to isolate it from outside - best with epoxy glue).

Using 0.8mm thickness of laminate you need piece of square 22mm to get 1pF.

If you use thin sheet of mica isolator between two small copper sheets you can even build as small as 2mm square of copper plates to get 1pF.

There are other solutions too, par example by thin coaxial cable or two isolated wires spiral wound and from one side cut to 1pF value (here you need LRC meter: first measure 10pF length and then cut to 1/10 of those length - or LRC meter should be very precise and correctly calibrated).

[baum7154]

Dan, No, I didn't have a feed line (coax?) attached to the coil. I made the gimmick cap and now have self resonance at 2.27 MHZ.
That is, on my scope the waveform jumps to 3-4x the p-p voltage at that frequency. I have a harmonic spike around 759Khz but it is just a slight increase in p-p voltage.
That seem pretty high for self resonance on a flat spiral?

-----------------------------------------------------------------------

That frequency is entirely possible according to what I have heard on flat spirals. If you put a36" length of feedline either twisted pair or coax I would expect that SRF to drop significantly. The additional capacitance of the feed is the cause for the drop. Then if you put a shield on both sides with its capacitance added an even larger drop I believe will occur. I'm curious to hear where it ends up.

Dan