PI design parameters

[Anonymous]

Hello all! I am an avid electronics hobbiest, although my work is M.E. I love electronics in my spare time. I have also been infected long ago with gold fever. Any how, I am playing around with designing a PI detector used for nugget shooting. I was wondering if some of you could impart what design considerations I should take into account. I am incorporating my experience with past experiments(coilguns-railguns) to build the coilswitching system I can achieve very fast switching and decay times using a diagnal halfbridge. I know gold will not produce a very long duration signal so my delay time before sample has to be fast. I intend to use a ATMEL Mega8 for pulse generation, sample intergration, and display driving, etc. my question is: what makes a metal detector a great nugget shooter? What parameters am I looking for? Hi freq, pulse width, etc. Thanks for the help!

 

[Anonymous]

Hi William,
I don't know much about these things really, but the things I would like in a nugget detector are;
Body mount control box, lightweight coil and shaft.
Low current usage 100mAh.
Watertight coil and showerproof control housing.
Volume altering audio, not frequency alone.
Earth field and background noise exclusion.
Motion response.
High sensitivity.
Low battery level indication.
Ground response exclusion.
High frequency, 7 to 10 kHz. (contingent upon the following adjustable settings )
Pulse width - 40 to 100us
Sample delay - 6 to 15us
Many other wants, but not yet realistic. Read as many patents as possible on Carl's site, and these will dictate to you the constraints and ideals with which you will need to design.
Cheers
Kev.

 

[Anonymous]

Hi William,
Kev pretty well covered the basics of what is needed in a decent gold detecting PI.
Depending upon just what size gold you want to detect, will determine the minimum delay needed. There are decent size nugget that require a delay of less than 8 usec or so to detect, so I would agree with Kev that a PI having an adjustble delay going down to 6 usec would be ideal.
A pulse frequency adjust is also a requirement if you are going to hunt with other people, to assure there is no or minimal crosstalk between detectors. It is also required in some places to minimize other external noise.
I prefer to run my PI at a higher pulse frequency so it is above the normal audio range. This helps with the signal to noise ratio, and reduces the posibility of a audible hum, directly do to the pulse rate.
I have my pulse width capabilities going down to about 25 usec and is used when the delay is extremely short, but 40 usec is about right for a minimum pulse duration.
A form of ground balance is preferred as is the ability to use a DD coil.
I prefer both audio increase and frequency change to indicate a target.
Some form of Iron ID would also be desireable. Especially, if it could determine the simple rusty can or parts of one.
Obviously, one very important issue is the ability to push the gain or sensitivity to maximum whenever possible. This requires very low noise high gain opamps and/or high resolution A/D.
Hope this helps in your design.
Reg

 

[Anonymous]

Thanks for the responses!
I have been looking at the hammerhead PI detector and would like to build one to test out. I think it will be a great learning platform for me to develop my Atmel PI detector (which of course I will share with everyone when its finished <img src="/metal/html/wink.gif" border=0 width=15 height=15 alt=""> )
I have the gerber viewer and will make my own PCB. Has anyone encountered any problems when veiwing these gerber files for the hammerhead. I keep getting this "self intersecting polygon error" when I open it, but when I look at the cordinates given fo rthe error I cant find anything wrong. Any tips from people who have built this board themselves? I will use the press-n-peel method.

 

[Anonymous]

[No message]