best coil for baracuda?

[kt315]

Hi there, I've recently built Baracuda PI, and it works very well with a mono coil I've built, 23 cm diameter with 0.5 mm insulated copper wire, 22 turns, about 392 uH, but I know this awesome circuit can give more. ��

For now, I'm obtaining a sensitivity of about 10 ? 12 cm for a normal coin or ring, and some 20 ? 25 cm for some bigger object. I would like to know if someone out there did build or use this PI metal detector, and what particular coil is the best for it.

Thanks in advance. ��

just transfer good request in good thread.

[kt315]

[QUOTE=bbsailor Basket coil construction is designed to reduce the coil turn-to-turn wire capacitance that would be higher if the wires laid parallel to each other. Once advantage of reducing coil capacitance is the potential for that coil to sample faster due to less capacitance in the TX/RX circuit. When delays get below about 10 uS the type of wire needs to be considered so that the wire itself does not retain any eddy currents and then look like a target. That is why Litz wire makes a good coil wire when attempting to achieve low sampling time. The value of the damping resistor is a good indicator of how much coil and TX/RX circuit capacitance needs to be damped.

The ability to fully stimulate a target with a TX pulse requires that the coil discharge TC be five times faster than the target TC. Here is a thought experiment. A charged 2 uS target, fully stimulated, will fall to near noise in about 5 TCs or 10 uS for that target.That is why you would need to sample at less than 10 uS for a 2 uS TC target.

Now, to fully stimulate a 2 uS target requires a coil discharge TC of 2/5 or 0.4 uS. The coil discharge TC is determined by the coil inductance divided by the damping resistor value. A 300 uh coil would need a 750 ohm damping resistor to meet that requirement. Now you can see why making fast coils requires using many techniques to reduce capacitance and allow a higher value damping resistor to be used to fully stimulate small, low TC targets.

In mono coil PI metal detectors the op amp input resistor, typically 1K ohm, is ahead of two clamping diodes and while these diodes are conducting with a voltage above about 0.6V, the input resistor is effectively in parallel with the damping resistor thus effectively reducing the damping resistor value until the voltage falls below 0.6V. That is why DD coils that separate the RX and TX circuits are easier to operate at lower delays.

Optimizing coils should be done for a class of and small range of target TC ranges to have the best results. This includes the PPS rate, coil current, flyback clamping, damping resistor value, delay time, number of samples being integrated, coil size, coil sweep speed and coil's response to the soil and detecting environment.

The basket coil is just one way to attempt to reduce capacitance but there are many more things that need to be considered.

Joseph J. Rogowski[/QUOTE]

just transfer good post all-one-place.

[nonkapo]

LoL my poor Baracuda put in shame the chinese knockoffs of Teknetics G2 and T2;-))). Focused on the dismantling manuals and desert safaris these days:-). Facing another issue with my spiral coils. Feel like when fixed in the housing and cross the second cable above i'm loosing sensitivity. In other words the coil detect the external end. Will update soon and probably open separate thread for this.

Tibuck19, KT mean basket coil, not ordinary as the air is better insulation than the epoxy. That's why you loose speed when you mold the basket. On the ordinary coil even if there is some effect cant be so bad. Correct me if I'm wrong.

[waltr]

Really! over there in the USA ! That great! guess the boarders were not so hard to pass back then hahaha, I found this one over the summer with the TGSL, but its a One Penny not a half penny, it measures 33mm across and Wight 15grams, They are a great find as its a good indicator no one has detected where its found, as it rings up good! but don't get excited about its value lol I had it praised, it is considered very good condition, but because its a bank token and not actual money, its only worth 10$ CAD but for a bit of copper buried in the ground for a 168 years! Still a great find. Congrats I know I was really happy to find mine!

Yea, it is a great historical find but no monetary value just like the King George half penny and US large cent coins I've found.
Great yours is a full Penny (can't tell from pic) and much better condition than the half I found.
Picture here: https://metaldetectingforum.com/showthread.php?t=269101
The US was also having coin shortage issues particularly during and after the Civil War and many foreign coins were still legal tender.
Would love to know the story of how this coin made it down here near Philadelphia and then end in what was a farm field.

Ok, enough thread hijacking.

nonkapo,
How is the building going?

[Tibuck19]

Tibuck19,
I see a Bank of Canada Token as your avatar.
I found one of these a few days ago. Its an 1850 half penny token found on a local soccer field.

Really! over there in the USA ! That great! guess the boarders were not so hard to pass back then hahaha, I found this one over the summer with the TGSL, but its a One Penny not a half penny, it measures 33mm across and Wight 15grams, They are a great find as its a good indicator no one has detected where its found, as it rings up good! but don't get excited about its value lol I had it praised, it is considered very good condition, but because its a bank token and not actual money, its only worth 10$ CAD but for a bit of copper buried in the ground for a 168 years! Still a great find. Congrats I know I was really happy to find mine!

[waltr]

like I mentioned before not all options are equal, Personally for the Surf PI and the Cuda, I made few Mono loop coils at 320uH with 22awg to 24awg, None of the PI coils I made have been shielded, they are all filled fiber glass resin with a UV accelerator additive in it for a very fast low heat set time. I haven't tried what Kt mention of leaving a air gap around the coil loop, Might help but seems like extra work to me, As for my coils, On average, the coils diameter matches what coil depth will search, Example: the 6 inch coil has a search depth 6.5 and the ten inch is 10" search depth. But I only have made 6 or so fully finish coils that I use, Kt Probably made a ton more finish and working coils then I do ! might be wise to try his method out also.

Same here. These simple coils do work which is why I recommend doing a simple coil to start. Much easy to make and will get your detector going. Later you can try other coils but testing and evaluating each is a task to decide if it is better or not.

Tibuck19,
I see a Bank of Canada Token as your avatar.
I found one of these a few days ago. Its an 1850 half penny token found on a local soccer field.

[Tibuck19]

like I mentioned before not all options are equal, Personally for the Surf PI and the Cuda, I made few Mono loop coils at 320uH with 22awg to 24awg, None of the PI coils I made have been shielded, they are all filled fiber glass resin with a UV accelerator additive in it for a very fast low heat set time. I haven't tried what Kt mention of leaving a air gap around the coil loop, Might help but seems like extra work to me, As for my coils, On average, the coils diameter matches what coil depth will search, Example: the 6 inch coil has a search depth 6.5 and the ten inch is 10" search depth. But I only have made 6 or so fully finish coils that I use, Kt Probably made a ton more finish and working coils then I do ! might be wise to try his method out also.

[kt315]

do not pour by epoxy the winding. thats MAIN rule for to do right fast coil.
leave a big air gap between the winding and a housing. i use a paper rope and after wind scotch tape above.
it gives me 3-4mm of the air gap.

[nonkapo]

Thanks. So there is no benefit trying to match exactly the original specs of the coil. Will make one smaller for smaller items and that's all.

[waltr]

Few more questions about the coil for Bara. Taken directly from Geotech coil database here are the parameters :
Barracuda 8" 545uH 1.24. I read that a 15 inch coil was also offered, hopefully with the same parameters? Now what bothers me since original design was with Darlington how accurate are these induction and resistance for the MOSFET? I read many topics about how tolerant the Cuda is about the coils but still curious. On few places i so 450uH with more or less same resistance. So please do not curse me why i'm asking again.
Thank you.

Typically the coil is not critical on a PI detector.
Most run fine with coil L of 300 to 500uH.
Coil resistance is also not critical with most lower current PI detectors (under a few Amps).

I have found that if the coil R is too lower the detector can be unstable. If this happens then simply add some R in series with coil.
This is also helpful for shorter TX ON pulses since it seems best to have the coil current almost level off during the TX on period and series R helps this since it lowers the coils Tau [Tau(sec) = L/R].

[nonkapo]

Few more questions about the coil for Bara. Taken directly from Geotech coil database here are the parameters :
Barracuda 8" 545uH 1.24. I read that a 15 inch coil was also offered, hopefully with the same parameters? Now what bothers me since original design was with Darlington how accurate are these induction and resistance for the MOSFET? I read many topics about how tolerant the Cuda is about the coils but still curious. On few places i so 450uH with more or less same resistance. So please do not curse me why i'm asking again.
Thank you.

[kt315]

https://discogslabs.imgix.net/biblio...auto=compress&

one from his books

[kt315]

Andrej Schedrin

[kt315]

crosslinked to
http://www.geotech1.com/forums/showt...ax-good-enough

[Altair]

I am ready to spray graphite inside of my coil housing, just a big doubt I have: how many spray pass have to put for best result? Is one enough or not?
Please a little bit help, thanks!