Find's Treasure Forums
Welcome to Find's Treasure Forums, Guests!
You are viewing this forums as a guest which limits you to read only status.
Only registered members may post stories, questions, classifieds, reply to other posts, contact other members using built in messaging and use many other features found on these forums.
Why not register and join us today? It's free! (We don't share your email addresses with anyone.) We keep email addresses of our users to protect them and others from bad people posting things they shouldn't.
Click here to register!
Need Support Help?
Cannot log in?, click here to have new password emailed to you
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
coils
- Thread starter Tinkerer
- Start date
The geometry of a coil that is required to be fast for a PI machine requires a minimum of capacitance between adjacent turns. There are many techniques, including all that you have mentioned except oval coils, that can accomplish that. While squeezing the absolute minimum amount of capacitance out of a given inductance value coil requires some pretty creative methods to reduce capacitance, practical aspects of coil winding and shielding must also be considered. One of my fastest coils uses a simple coil of 50 ft. of Radio Shack AWG 30 wire wrap with a Kynar insulation on a 10.5" diameter coil, a little under 300 uH, to fit a Hays Electronics coil housing.
Thicker insulation lowers the capacitance but also reduces the inductance slightly compared to a tightly packed coil.
It all gets down to answering a few questions to plan for the right coil.
What you are looking for?
How quickly are you trying to sample?
What power are you planning to pump into the coil?
What is the maximum voltage your coil driver MOSFET can handle?
If maximum power is not a concern, then putting in a series resistance between a few ohms and 40 ohms may be appropriate.
Remember, fast coils need a shield that is not detectable or is minimaly detectable by the PI machine at the fastest setting.
Look at using some new MOSFETS with about 50 pf in output capacitance. It is much easier to reduce capacitance and get faster PI responses with these new MOSFETS than trying to get the same reduction in coil capacitance. Search some older posts by Reg about his experience with these newer low capacitance MOSFETS.
bbsailor
Thicker insulation lowers the capacitance but also reduces the inductance slightly compared to a tightly packed coil.
It all gets down to answering a few questions to plan for the right coil.
What you are looking for?
How quickly are you trying to sample?
What power are you planning to pump into the coil?
What is the maximum voltage your coil driver MOSFET can handle?
If maximum power is not a concern, then putting in a series resistance between a few ohms and 40 ohms may be appropriate.
Remember, fast coils need a shield that is not detectable or is minimaly detectable by the PI machine at the fastest setting.
Look at using some new MOSFETS with about 50 pf in output capacitance. It is much easier to reduce capacitance and get faster PI responses with these new MOSFETS than trying to get the same reduction in coil capacitance. Search some older posts by Reg about his experience with these newer low capacitance MOSFETS.
bbsailor
Hi Tinkerer,
Bbsailor has provided a lot of good questions that really need to be answered before a lot can be said about coils at this point.
If you are just beginning to work with a PI, then I would suggest you simply start with a basic coil maybe 8" to 11" in diameter for your first one. From there, you can start to compare different designs.
However, you should remember that there are a lot of other factors that can influence your final results besides the characteristics of the coil. Simple things such as the Opamp used for the preamp can make a big difference. You can build a fantastic coil but lose a lot with the preamp circuitry, let alone the FET.
Yes, the FET does influence the speed at which the signal decays. You can gain speed but will generate more heat since the internal resistance of the FET becomes a factor.
Now, most of the critical factors of a coil, the FET, and the preamp do not become issues if you just want to build a decent PI to look for larger objects. They do become more important as you try to reduce the time before sampling. This is needed when trying to detect much smaller gold objects or other lower conductive materials. Even coil current and pulse duration become important factors in such a design.
Now, as a decent gold machine aimed at finding smaller gold, you can limit the coil current to an amp or so, use a FET rated at maybe 5A, use the NE 5534A as your preamp. Actually, this machine could be used for most anything, but if you are going to be pulsing for a longer duration (more than 100 usec all the time ) , then I would use a heavier duty FET just to be safe.
It is going to take a while to get a real handle on how PI's really work, so just dive in and have fun.
Reg
Bbsailor has provided a lot of good questions that really need to be answered before a lot can be said about coils at this point.
If you are just beginning to work with a PI, then I would suggest you simply start with a basic coil maybe 8" to 11" in diameter for your first one. From there, you can start to compare different designs.
However, you should remember that there are a lot of other factors that can influence your final results besides the characteristics of the coil. Simple things such as the Opamp used for the preamp can make a big difference. You can build a fantastic coil but lose a lot with the preamp circuitry, let alone the FET.
Yes, the FET does influence the speed at which the signal decays. You can gain speed but will generate more heat since the internal resistance of the FET becomes a factor.
Now, most of the critical factors of a coil, the FET, and the preamp do not become issues if you just want to build a decent PI to look for larger objects. They do become more important as you try to reduce the time before sampling. This is needed when trying to detect much smaller gold objects or other lower conductive materials. Even coil current and pulse duration become important factors in such a design.
Now, as a decent gold machine aimed at finding smaller gold, you can limit the coil current to an amp or so, use a FET rated at maybe 5A, use the NE 5534A as your preamp. Actually, this machine could be used for most anything, but if you are going to be pulsing for a longer duration (more than 100 usec all the time ) , then I would use a heavier duty FET just to be safe.
It is going to take a while to get a real handle on how PI's really work, so just dive in and have fun.
Reg
Thanks, Reg and bbsailor, My first detector was the Elsec 700 D underwater metaldetector back in 1979 and many others after that. In the late 80ties I built some PI detectors for personal underwater use only with several size coils, 1800 to 3000pps,5534 preamp etc.
Now I am facing a new challenge: Not far from where I live, there is a very large region with lots of gold. Mining companies and wildcat prospectors recover tons of gold, while the local indigenous people, that virtually sit on the gold , go hungry. A few detectors have come to the region, but they are much too expensive for the Indians and too complicated for use by people that can not read nor write and live with less than one dollar per day.
I would like to help these people by producing a gold detector with the following characteristics:
Rugged and rain proof
Sensitive for small nuggets
Very simple in its use - not more than 2 knobs
So basically I consider using my old design and speed it up to 10Kpps. A coil size that sweeps a decent area on each sweep but can be used between rocks and bushes. This brings me to a coil of about 5" diameter but stretched to about 12 inches 5x12 or so. External lead acid gelcell battery that could be carried on the belt or backpack (matter of cost)
I would apreciate any suggestions and help from all.
Tinkerer
Now I am facing a new challenge: Not far from where I live, there is a very large region with lots of gold. Mining companies and wildcat prospectors recover tons of gold, while the local indigenous people, that virtually sit on the gold , go hungry. A few detectors have come to the region, but they are much too expensive for the Indians and too complicated for use by people that can not read nor write and live with less than one dollar per day.
I would like to help these people by producing a gold detector with the following characteristics:
Rugged and rain proof
Sensitive for small nuggets
Very simple in its use - not more than 2 knobs
So basically I consider using my old design and speed it up to 10Kpps. A coil size that sweeps a decent area on each sweep but can be used between rocks and bushes. This brings me to a coil of about 5" diameter but stretched to about 12 inches 5x12 or so. External lead acid gelcell battery that could be carried on the belt or backpack (matter of cost)
I would apreciate any suggestions and help from all.
Tinkerer
That is very interesting, I am curious, are you in S. America? How "Hot" mineralized tdo you consider the ground there? on that will depend to a degree how simple a PI could be and work properly, of course it it is hotter, the first simple answer is to use a DD coil and learn to make that, I believe there is one on the "Geotech" webside. Don
Hi Tinkerer,
I admire you for your willingness to help the less fortunate. Few people are really willing to do that with the intensity you have indicated.
Now, without seeing your initial design, it is difficult to determine whether it would work well for the application you have selected. My guess is it could be modified but you may run into problems or limitations. My guess is the pulse circuitry would have to be changed to one of the more modern FET's for starters.
If you are willing to share the schematic, you might post it here or on the Geotech forum for discussion. If not, you might send me an email and we can discuss it privately. I can be reached at rgsniff@comcast.net .
The Hammerhead PI mentioned before can be modified to do what you want but would take a little effort. Once modified, it should work quite well, I would think.
As Don said, the modification to use a DD coil would be a good place to start. This mod would greatly reduce the ground signal normally associated with a PI operating at a short delay, which is what is needed to detect small gold. The use of a DD coil could eliminate the need for ground balance circuitry in most cases. In the worst of ground, it would still make a PI detector much more usable.
As for powering the PI, I would recommend reducing the power requirements and operating off AA batteries that are rechargeable. This keeps the weight low, the cost down, and still works very well. Eric's GQ V2 is a perfect example of what can be accomplished. Now, I don't expect a person to be able to build one to match Eric's machine, but they can improve what they have so it is more sensitive and easier to use.
A two knob control unit is practical. Actually, a switch and one knob could be done, but it is a good idea to have some adjustment over the pulse rate to minimize noise. In other words, it is possible to build what you want.
Now, if AA type batteries are used, the something as simple as a $20 solar panel could be modified to charge the batteries quite easily. If power is available, then a simple wall charger would work fine.
As for detecting gold, I would recommend reducing the delay as much as possible and still have a reliable machine. This takes some doing, but a DD coil helps in this aspect. The detection of small gold is more of a function of sampling early than using more pulses per second, so increasing the pulse rate isn't going to help that much when compared to reducing the delay.
I hope this gets you started. Again, you might post your requests and possibly the schematic over on the Geotech forum to get a greater input to your request.
Reg
I admire you for your willingness to help the less fortunate. Few people are really willing to do that with the intensity you have indicated.
Now, without seeing your initial design, it is difficult to determine whether it would work well for the application you have selected. My guess is it could be modified but you may run into problems or limitations. My guess is the pulse circuitry would have to be changed to one of the more modern FET's for starters.
If you are willing to share the schematic, you might post it here or on the Geotech forum for discussion. If not, you might send me an email and we can discuss it privately. I can be reached at rgsniff@comcast.net .
The Hammerhead PI mentioned before can be modified to do what you want but would take a little effort. Once modified, it should work quite well, I would think.
As Don said, the modification to use a DD coil would be a good place to start. This mod would greatly reduce the ground signal normally associated with a PI operating at a short delay, which is what is needed to detect small gold. The use of a DD coil could eliminate the need for ground balance circuitry in most cases. In the worst of ground, it would still make a PI detector much more usable.
As for powering the PI, I would recommend reducing the power requirements and operating off AA batteries that are rechargeable. This keeps the weight low, the cost down, and still works very well. Eric's GQ V2 is a perfect example of what can be accomplished. Now, I don't expect a person to be able to build one to match Eric's machine, but they can improve what they have so it is more sensitive and easier to use.
A two knob control unit is practical. Actually, a switch and one knob could be done, but it is a good idea to have some adjustment over the pulse rate to minimize noise. In other words, it is possible to build what you want.
Now, if AA type batteries are used, the something as simple as a $20 solar panel could be modified to charge the batteries quite easily. If power is available, then a simple wall charger would work fine.
As for detecting gold, I would recommend reducing the delay as much as possible and still have a reliable machine. This takes some doing, but a DD coil helps in this aspect. The detection of small gold is more of a function of sampling early than using more pulses per second, so increasing the pulse rate isn't going to help that much when compared to reducing the delay.
I hope this gets you started. Again, you might post your requests and possibly the schematic over on the Geotech forum to get a greater input to your request.
Reg
Hi Reg: Do you think it might be possible if he used a Pocket Uniprobe and adapt it for a DD coil and lower the Pulse Delay some, as an inexpensive commercial unit? maybe he could get a price break for some numbers and he could modify and learn to make a DD coil thru you or the Geotech, or maybe Carl would make some smaller ones up at good price. This would be more of course than building from scratch, but it would put people in the field quicker and since it is gold, then can repay as they find some? just ideas. Don
Hi Reg,
First of all many thanks for your help as well as all the other helpful suggestions. I think I take your advise and use the HH design as a starting point instead of going back to my 20 year old design.
It would be fantastic if we could make this an open project at Geotech, where everybody could help developing it.
Here are some more parameters that impose limits:
Batteries: Nicad AA
First of all many thanks for your help as well as all the other helpful suggestions. I think I take your advise and use the HH design as a starting point instead of going back to my 20 year old design.
It would be fantastic if we could make this an open project at Geotech, where everybody could help developing it.
Here are some more parameters that impose limits:
Batteries: Nicad AA
If properly used, NiCads will not show memory effect. The key is to design the circuit to run down to a low enough voltage to ensure that the battery is fully discharged. Alternatively, put one more cell in the battery that absolutely needed to achieve the necessary operating voltage.
Go to http://www.antonbauer.com, then "How to choose" then "Technical Information" and get the Video Battery Handbook, and read starting on page 47.
Lead-acid batteries also have a memory effect.
Go to http://www.antonbauer.com, then "How to choose" then "Technical Information" and get the Video Battery Handbook, and read starting on page 47.
Lead-acid batteries also have a memory effect.
Thanks for the input, Joe Gwinn,
I totally agree with you on the batteries and your post has helped me to understand that I will have to properly explain all the problems if I expect help in solving these problems. I hope I will not bore you with my lengthy posts.
So lets start with a profile of the
I totally agree with you on the batteries and your post has helped me to understand that I will have to properly explain all the problems if I expect help in solving these problems. I hope I will not bore you with my lengthy posts.
So lets start with a profile of the
While you are on a PI forum, depending on ground mineralization, have you thought about VLF, there is range of older ones especially that would be a good price, like Fisher goldbug and white's goldmasters. If you are trying to construct a simple PI that is different, but also alot of requirements to be met in constructing all components, would seem a time consumming task and getting supplies to you. Don
Hi Tinkerer,
Ok, lets take the issues one at a time. First, lets assume a 12V battery will be used regardless of the metal detector type built. If this is the case then a simple battery charger can be made using a solar panel.
One cheap panel capable of about 3 watts output is one originally made to keep the batteries charged on VW's when shipped. Here in the US, the chargers can be purchased off Ebay. In some cases, they can be purchased from new owners of VW's as they buy the car. These chargers normally attach to the windshield of a new car and plug into the cigarette lighter. The solar panel keeps the battery charged in case there is a long period between the car being built and being sold.
Generally, these panels will sell for less then $20 each. I purchased a 3 pack for about $12 each. Add a female cigarette lighter plug, a connector to match one on the battery pack and now you have a cheap, but functional charger. Actually, the original cigarette lighter plug that is installed on the panel can be cut off and simply install a matching connector to the battery. Again, this is cheap, yet effective.
These panels have a schottky diode installed somewhere. It may be in the cigarette lighter plug, so if the plug is removed, then the diode probably should be replaced. Again, I am not sure just where the diode is located.
There are some lower wattage units selling on Ebay for about the same price that are new. My guess is one could use a unit capable of about 200 ma to serve the purpose. The recommendation would be to have a spare battery and the batteries be switched on a regular basis.
Anyway, this is one possible solution to the power problem.
Reg
Ok, lets take the issues one at a time. First, lets assume a 12V battery will be used regardless of the metal detector type built. If this is the case then a simple battery charger can be made using a solar panel.
One cheap panel capable of about 3 watts output is one originally made to keep the batteries charged on VW's when shipped. Here in the US, the chargers can be purchased off Ebay. In some cases, they can be purchased from new owners of VW's as they buy the car. These chargers normally attach to the windshield of a new car and plug into the cigarette lighter. The solar panel keeps the battery charged in case there is a long period between the car being built and being sold.
Generally, these panels will sell for less then $20 each. I purchased a 3 pack for about $12 each. Add a female cigarette lighter plug, a connector to match one on the battery pack and now you have a cheap, but functional charger. Actually, the original cigarette lighter plug that is installed on the panel can be cut off and simply install a matching connector to the battery. Again, this is cheap, yet effective.
These panels have a schottky diode installed somewhere. It may be in the cigarette lighter plug, so if the plug is removed, then the diode probably should be replaced. Again, I am not sure just where the diode is located.
There are some lower wattage units selling on Ebay for about the same price that are new. My guess is one could use a unit capable of about 200 ma to serve the purpose. The recommendation would be to have a spare battery and the batteries be switched on a regular basis.
Anyway, this is one possible solution to the power problem.
Reg
I am familiar with Pi's, would have to learn about VLF all over. Again I have to slap my hand for not explaining myself better. I am not talking about one single machine, but maybe 10 to start with. built here. Importing parts is not really a problem besides the cost. US$ per pound can add up to a hefty bill. But a bag of IC's and transistors etc. that weighs a pound contains a lot of parts.
Some thoughts:
AC power that varies from 80 to 140 volts is not really a problem, so long as one designs for it. Lots of power bricks are rated for use at 50 or 60 Hz, 100 to 240 volts, and I have seen units rated for 400 Hz as well. The wall warts that contain only a transformer, rectifier bridge, and perhaps capacitor will work just fine, yielding a DC output that varies in proportion to the AC input. Put this through a simple voltage regulator / battery charger. Parts count is the issue, not efficiency.
The limited time availability, a few hours every few days, could be more of a problem than voltage variation.
Six-volt lead-acid motorcycle batteries are cheap and available and very rugged, but maybe a bit too heavy to carry all day long.
Gel-cell lead-acid batteries (from Gates) may be a better fit physically.
Six-volt lead-acid batteries are charged very simply, by imposing about 7 volts on the battery and waiting. Nothing fancy.
Solar chargers for 6-volt batteries are cheaper than for 12 volts, by about half. Isn't the problem with solar the fact that the end user needs to be out detecting during the day, not waiting for the solar charge (which will be slow)?
If you can wait 6 hours, NiMH batteries can be trickle charged to full with a current-limiting resistor only. Partial charges are OK.
What is the color and kind of the soil that will be searched? In which country is this? (The issue here is how much and what kind of ground noise to expect.)
Water-proofing to handle near immersion is expensive, because of the cost of sufficiently good hardware. Perhaps it would be better to pot the electronics in wax or pitch, which can be melted to gain access for repair, and poured back when done.
It seems to me that you could do far worse than simply taking the kit PI detector people have mentioned and adapting it. The current end users have nothing but their bare hands, and any detector at all is likely to prove immensely effective in practice, even if this detector would be laughed out of the treasure hunting community.
AC power that varies from 80 to 140 volts is not really a problem, so long as one designs for it. Lots of power bricks are rated for use at 50 or 60 Hz, 100 to 240 volts, and I have seen units rated for 400 Hz as well. The wall warts that contain only a transformer, rectifier bridge, and perhaps capacitor will work just fine, yielding a DC output that varies in proportion to the AC input. Put this through a simple voltage regulator / battery charger. Parts count is the issue, not efficiency.
The limited time availability, a few hours every few days, could be more of a problem than voltage variation.
Six-volt lead-acid motorcycle batteries are cheap and available and very rugged, but maybe a bit too heavy to carry all day long.
Gel-cell lead-acid batteries (from Gates) may be a better fit physically.
Six-volt lead-acid batteries are charged very simply, by imposing about 7 volts on the battery and waiting. Nothing fancy.
Solar chargers for 6-volt batteries are cheaper than for 12 volts, by about half. Isn't the problem with solar the fact that the end user needs to be out detecting during the day, not waiting for the solar charge (which will be slow)?
If you can wait 6 hours, NiMH batteries can be trickle charged to full with a current-limiting resistor only. Partial charges are OK.
What is the color and kind of the soil that will be searched? In which country is this? (The issue here is how much and what kind of ground noise to expect.)
Water-proofing to handle near immersion is expensive, because of the cost of sufficiently good hardware. Perhaps it would be better to pot the electronics in wax or pitch, which can be melted to gain access for repair, and poured back when done.
It seems to me that you could do far worse than simply taking the kit PI detector people have mentioned and adapting it. The current end users have nothing but their bare hands, and any detector at all is likely to prove immensely effective in practice, even if this detector would be laughed out of the treasure hunting community.
Hi Tinkerer,
I have a couple of questions about the size of gold generally found in the designated area? Are most of the nuggets extremely small, say less than 5 grains in size, or does the size of the nuggets span a wide range?
It is much easier to build a PI detector that displays little or no ground sensitivity but will not too sensitive to nuggets less than maybe 10 grains in size. Building such a detector is much simpler.
A detector that can detect nuggets down to lets say, 3 to 5 grains in size can be done with a detector having a delay of about 15 usec or so. The 3 grain size will be very weak signals at best, though.
A PI operating at 10 usec is quite difficult to build. Actually, the detector isn't the difficult part, but building a matching coil that works properly, is. Coils almost have to be built to match the machine as the delay becomes shorter. This can prove to be a problem if coils are to be changed on a regular basis.
Reducing the power to the coil makes things easier but it is still difficult.
So, I am thinking that one can try for the short delay PI but may have to go to something else if most of the gold is extremely small just to be safe.
If the gold is extremely small, some design work in the TR or off resonance detector field might just work better if done right. They would be much easier to build than a VLF. The nice thing about these designs is they can be built to run on much less power. As such, some form of different power system might be easier to build and lighter to use.
One other question, is there much trash to generate false or junk signals in the general area?
Reg
I have a couple of questions about the size of gold generally found in the designated area? Are most of the nuggets extremely small, say less than 5 grains in size, or does the size of the nuggets span a wide range?
It is much easier to build a PI detector that displays little or no ground sensitivity but will not too sensitive to nuggets less than maybe 10 grains in size. Building such a detector is much simpler.
A detector that can detect nuggets down to lets say, 3 to 5 grains in size can be done with a detector having a delay of about 15 usec or so. The 3 grain size will be very weak signals at best, though.
A PI operating at 10 usec is quite difficult to build. Actually, the detector isn't the difficult part, but building a matching coil that works properly, is. Coils almost have to be built to match the machine as the delay becomes shorter. This can prove to be a problem if coils are to be changed on a regular basis.
Reducing the power to the coil makes things easier but it is still difficult.
So, I am thinking that one can try for the short delay PI but may have to go to something else if most of the gold is extremely small just to be safe.
If the gold is extremely small, some design work in the TR or off resonance detector field might just work better if done right. They would be much easier to build than a VLF. The nice thing about these designs is they can be built to run on much less power. As such, some form of different power system might be easier to build and lighter to use.
One other question, is there much trash to generate false or junk signals in the general area?
Reg