Pulse induction and coil Polarity!!!

[dfbowers]

While winding a few coils for my Barracuda I came across a startling discovery.
I was noticing that when building coils, some really shine and other times the performace is somewhat lack-luster.
Specifically, experimenting with an older 8" coil I made for one of my Stuart PIs, (in the 300uH range) I noticed that
by changing polarity I could get up to an extra 2" on coin sized objects just by changing coil polarity! NOT by flipping the coil or using the other side but reversing leads!!

Me, thinking I was nuts, repeated this over and over (and over) again until I was convinced that it was true. I tried different orientations, vertical and horizontal, top and bottom.. no difference, but reversing the leads made the difference.

Then, to confirm this, I wound another coil on a 10" form with a 1/4 inch groove using plain olde 20 Gauge magnet wire and repeated the experiment. Sure enough, the new coil behaved the same way. An extra 2" for nothing more than reversing the leads. With as many coils that I have ever built, I never noticed this. Now the "hot" coil had the "end" of the winding connected to ground (outer layer of the coil) and the beginning of the winding connected to the IRF9530 drain.
Neither coil was shielded, delay set around 15uS. Anyone else ever seen this?

Don

[Chet]

Hi Don

I haven't seen this but it is worth trying. Maybe the outer windings are acting as a Faraday shield reducing stray capacitance and EMI. And when you reverse them the outer windings are adding more capacitance loading onto the flyback pulse and also allowing some EMI to corrupt the target signal. An interesting finding anyway.

Have a good day,
Chet

[CAS]

Hi Don
i had experimented with this myself after reading that when constructing a basket weave coil to use the outer wire as the one to connect to gnd. I now try the coils both ways before I finish them off.
By the way Don, the shells arrived and look great. I have just shielded one with graphite for a standard fast coil and am going to use the other with an unshielded basket coil.

[Skippy]

I'm guessing this is a capacitive effect, and the decay characteristics of the coil differs depending on polarity. Perhaps if you changed polarity, AND optimised the damping resistor, your two depths may come closer?

[green]

Interesting, does the amplifier out scope trace look different when you reverse the leads? Does the effect stay the same if you increase delay time.

[Darin]

I was working with my SM PI PRO checking some coils I made the other night and noticed the same thing. I am new to detector building but did not remember reading anything that said one lead was ground the other positive, so made me start looking to see if there was a correct way or is it witch ever works best.
Thanks for posting this.

[baum7154]

While winding a few coils for my Barracuda I came across a startling discovery.
I was noticing that when building coils, some really shine and other times the performace is somewhat lack-luster.
Specifically, experimenting with an older 8" coil I made for one of my Stuart PIs, (in the 300uH range) I noticed that
by changing polarity I could get up to an extra 2" on coin sized objects just by changing coil polarity! NOT by flipping the coil or using the other side but reversing leads!!

Me, thinking I was nuts, repeated this over and over (and over) again until I was convinced that it was true. I tried different orientations, vertical and horizontal, top and bottom.. no difference, but reversing the leads made the difference.

Then, to confirm this, I wound another coil on a 10" form with a 1/4 inch groove using plain olde 20 Gauge magnet wire and repeated the experiment. Sure enough, the new coil behaved the same way. An extra 2" for nothing more than reversing the leads. With as many coils that I have ever built, I never noticed this. Now the "hot" coil had the "end" of the winding connected to ground (outer layer of the coil) and the beginning of the winding connected to the IRF9530 drain.
Neither coil was shielded, delay set around 15uS. Anyone else ever seen this?

Don

---=-----------------

Hi Don,

If the coil you wound was the spider coil pattern from the Chance PI Coil thread where the geometry keeps the successive and last windings encircling the previous windings then 'yes' there is a difference in polarity and the wire end tied to the winding finish' end must be tied to system ground with the wire end for the 'start' or innermost end must be tied to the system hot output. I have noticed a detection distance difference if these polarities are reversed in addition to the loss of the self shielding characteristic.

Regards,

Dan

[green]

Was wondering about my coils. Charted a 6 inch mono coil, (spiderweb, graphite paint). Wind pattern in magnet wire coil thread. Didn't see a difference in signal strength. Wondering if you can see a difference in the amplifier out scope trace with the coin on or very close to the coil. Guessing signal strength change would be about 25 percent of the higher amplitude for a two inch loss.

[baum7154]

Hi Green,

In looking at your 6" spiderweb coil it appears to me that it is basically a spiral wound coil with more air spacing because of the alternation between windings on the toothpicks. Very fast without shield but difficult to shield without adding significant capacitance from the shield. Regarding polarity it would be interesting to see if any spiral coil builders/users out there have noticed performance differences due to polarity of the coil.

Regards,

Dan

[dfbowers]

Interesting, does the amplifier out scope trace look different when you reverse the leads? Does the effect stay the same if you increase delay time.

The only difference in the scope trace I noticed was that there are some small noise spikes that get inverted, possibly from the clocking. Other than that, it looks the same. There has to be something else that I have not considered yet..

[green]

Hi Green,

In looking at your 6" spiderweb coil it appears to me that it is basically a spiral wound coil with more air spacing because of the alternation between windings on the toothpicks. Very fast without shield but difficult to shield without adding significant capacitance from the shield. Regarding polarity it would be interesting to see if any spiral coil builders/users out there have noticed performance differences due to polarity of the coil.

Regards,

Dan

Hi Dan
You are correct on how it's wound. Wish there was a name difference between the way I wind and yours. Maybe there is, you called yours a spider coil in reply #7 and I've been calling mine a spiderweb. Trying to use my magnet wire and it works better than bundle wound.

[dfbowers]

I think I know what's going on now. Up till now, I haven't touched anything on my bench except for the coils. I may be getting a feedback loop going on with the audio section (the other coil is my speaker). I might have some kind of regen effect going on. I should have know better as I have seen this with some of my IB projects in the past.
Don

[baum7154]

Good catch Don,

I ran into a similar kind of thing in my Chance PI Build and wound up shielding the audio isolation transformer and the internal speaker with copper foil to ground to eliminate spurious effects.

Regards,

Dan

[sinclairuser]

hi all, i have not noticed anything like this on my unshielded mono's(the only ones i tried connected backwards).
hi don, if you build a coil in one of your shells and cock it up, can you separate them again, its been potted with epoxy(single pour), is the shell toast or can you get the epoxied coil out again, i'd like to wind a new coil and reuse this shell thanks.

[Qiaozhi]

I think I know what's going on now. Up till now, I haven't touched anything on my bench except for the coils. I may be getting a feedback loop going on with the audio section (the other coil is my speaker). I might have some kind of regen effect going on. I should have know better as I have seen this with some of my IB projects in the past.
Don

That makes sense. Personally, I have never seen this issue with different results depending on which way round the coil is connected.