Need a bit of help: poor results = tuning issues or coil issues?

[malacoda]

Hi,

I've just built my first detectors - two Hammerheads - and am trying to figure out why I'm getting results that seem to be sub-par...

Here's the problem I have:

From viewing several threads in the coil section and the HH section, it seems that most of the outcomes for completed HH detector have average test results of at least:

9-13" detect distance in air for conductive metals (iron, coins, etc.), and
8-12" detect distance for low-conductance metals (gold rings, etc.)

I'm only getting a max of:
5-8" in air for conductive metals
3-5" in air for my gold ring (I suppose wedding bands are a common test item...)

During the build process all tests outlined in the project instructions passed and the results were very close to those published. I built the monocoil per the HH instructions - 2 sets actually: one I used copper mesh on as shield but failed to leave a gap between shield & coil and upon testing could not detect a gold ring at all... and the latest one is still bare. 26 AWG w/ 26 windings and a diam. of 10.5". My damping resistor was sized for the 500 uH coil . The build coil actually measured about 470-480 uH (was a few weeks ago so I can't remember the exact value).

The coil - which gave the above results - is wound so as to be flat in the vertical sense (one winding stack on top of the another).

My question: How critical is the shape of the coil - basket wind vs flat coil vs flat stack (or whatever one would call it... one winding atop another), etc. - in a PI detector? Also, could that 20 uH difference vs the damping resistor still be sized for 500 uH cause such an immense difference in depth capability?

As a total novice, I don't know if I should first focus time on the coil - and if so what should be changed/improved on it - in trying to find the cause of my poor results, or the PCB mounted electronics (which I suspect less as all the tests during the build gave good results)...

I should also note that I've tried increasing sensitivity via R35 and the sample delay and pulse width via R42 & R46 but with no truly significant change in the maximum 'air' depths I'm getting.

Any guidance/advice would be appreciated.

Malac

[bbsailor]

My question: How critical is the shape of the coil - basket wind vs flat coil vs flat stack (or whatever one would call it... one winding atop another), etc. - in a PI detector?

Basket weaves are used to obtain very low capacitance as the wire only come close to adjacent turns at the cross-over points of the weave pattern. This makes a bulky coil and one that would be more difficulkt to properly shield than a simple bundle wound coil. Winding the turns one on top of the other gets you a coil with lower capacitance than a bundle-wound coil but you need about two addititional turns to obtain the same inductance as you would need using a bundle-wound coil. I found that using wire with a low dielectric constant, like Teflon (PTFE) is the most practical way to make a fast coil. I say practical because you can easily wind it on 16 "C" screw-in hooks places in a circle of the desired diameter and add the spiral wrap to the coil while it is still on the hooks. This spiral wrap (use Polyethylene (PE) spiral wrap) acts as a spacer between the coil and the shield. Using this method on about a 10.5" diameter coil, and Teflon wire, I get about 140 pf coil-to-shield capacitance and a self-resonance near 1Mhz (coil and shield only). This method also makes a very compact coil that will easily fit into a variety of coil housings.

Low capacitance coils only have the potential to be very sensitive to low conductive metals as long as the PI design can operate at the lower delays of 10uS or a little lower (8 to 9 uS). Read the modification in the latest Hammerhead article that Carl recommends to reduce the delay below 10uS by reducing the value of a pot and resistor. Make this mod and connect your existing coil up to the your new PI machine and reduce the delay while waving a gold ring inder the coil. You should get to the point with your 480uS coil where the coil suddenly stops responding to the gold target. That means that you have reached the lower limit of the coil. I suspect that 480uH is a little high to obtain a delay around 10us.

What is the self resonance of each of your coils?

What delay are you seeking to achieve?

What type of targets are you seeking?


Also, could that 20 uH difference vs the damping resistor still be sized for 500 uH cause such an immense difference in depth capability?

No. The size of the damping resistor is determined by the total capacitance that the PI TX circuit sees with the combined capacitance of the coil, coax cable, shield, and MOSFET (COSS output capacitance). The Hammerhead RX input resistor also had an effect as, for a short time, the coil sees it as part of the damping resistor value. Reg reported that he increased the value of R12 to 2.2K to lower the gain slightly, and uses a FQPF2N30 low capacitance MOSFET to get delays around 10 uS. The 20uH difference will have no effect on the value of the damping resistor as there are other things, listed above, that have a greater impact on the value of this resistor much more than a few uH of coil inductance. With a low capacitance MOSFET and a low capacitance, bulk-wound Teflon wire coil of about 300 to 350 uH (10.5" diameter, about 19 to 21 turns (AWG 24 to AWG 26 stranded Teflon wire) depending on the wire size and insulation thickness) and spaced shield, you should be able to use a damping resistor of about 1K ohm. Make sure that what ever you use as a shield is minimally detected at the delay you want to operate at, otherwise the shield will be detected as a target and will reduce the sensitivity of your coil. Household aluminum foil only works well at longer delays for coin or relic hunting.

This additional information should help you. Let us know how it works out.

What test equipment do you have?


bbsailor

[malacoda]

more info on my situation

Thanks for the input bbsailor, I appreciate it.

Right now I'm a bit pressed for time - taking the kids to the beach next week and want to have the detectors in some type of working order so they can at least try to dig up some coins / jewelry...

As it stands, I've wound to new coils - one winding atop the other (since my housings won't fit a basket wind) - and have gotten my hands on some 3M 1.4 mil EMI copper foil shielding tape.

Since I'll need the next couple of nights to get the boards mounted in their boxes, the control pots mounted on the boxes, the boxes mounted on the stems, etc. I'll only have time to apply the tape to the inside of the housing walls, wrap the coil in some rigid foam (which will hopefully serve the same purpose as spiral wrap), temporily seal the housings with silicon RTV, and get everything mounted on the stems.

Hopefully the new shielding and windings combined with us detecting in an area that should have very little noise compared to the R&D lab where I work (where I've been doing the assembly and raw testing) will bump my initial results out bit so my kids have a half decent chance at some good beach finds.

Once I get back I'll be able to test, tinker, and tune futher. I have access to a multimeter and o-scope, don't think we have an inductance meter but I have to verify that.

I'll let you all know how things go once I get back - and'll more likely than not be asking more questions as I try to do the fine tuning at that point.

best regards,
Malac