3 to 1 ratio coil

[Reg]

Hi All,

Has anyone built a rectangular coil with a ratio of 3 to 1 or 4 to 1 (length to width ratio) and compared it to other coils? I am curious just how well one would work on a PI. I will probably build one shortly, but would like to hear from those who already have built one as to how they like or dislike it.

Thanks,

Reg

[Reg]

Works Great!!!

Hi All,

Thanks to Bill Hays, I had the opportunity to try a couple of different rectangular coils. One particular size, a 3 1/2" by 12" really impressed me. Building up a mono coil for my PI, I tested it against another coil that is about 9" by 12" and found the narrow coil to be about as sensitive. I did try a 3 1/2" by 16" coil but it didn't show any obvious improvement over the smaller one, so the smaller designed was selected.

The nice thing about this new coil design is it fits in places where other coils can't fit so it works great without losing any obvious depth. This is going to be a killer nugget hunting coil.

As a result of this test, I will be trying a big brother to this initial design if a few weeks. The new size will be 5 1/2" by 18". This is large enough that it will be easy to build up in either a DD or a mono coil. I am anticipating this new size will be very sensitive and be competitive to a much larger coil, maybe a somewhere between a 12" and 14" round mono or close to that size. However, the narrow footprint will allow its use in tight places, thus making it another ideal nugget hunting coil.

Actually, both sizes should work very well in the nugget hunting field.

After thinking about it, I can understand part of why it is very sensitve to smaller objects as well as deeper ones. Anyway, the design clearly has merit for those wanting to add a wide range of coils for all applications. As of now, I am hoping Bill Hays will be building these two size forms in the future. I am sure they will be a couple of my primary coils I will be using.

Reg

[nncunew]

Exactly

What wire gauge do you used to make this coils ???


I have 20 awg wire, is it good for this jobs ???

I am begining to make my coils for my PI, but I want make two coils one for smalls objects and another for large object. Using 20 awg, what sizes for the two coils ???

Thank,

[email protected]

[Reg]

Hi nncunew,

The 20 awg should be fine for the coils. I use a smaller wire, generally about a 26 awg since I don't pulse the coils for an extremely long time.

Two good sizes of coils are the 8" diameter and the 11" diameter. These two sizes are very common.

Now, I will use the 3 1/2" by 12" rectangular coil instead of an 8" coil in the future. This rectangular coil seems to have as good of depth as an 8" round coil, or very close to it. The narrow width appears to "sharpen" the signal from objects, thus making them easier to hear. At least, it appears that way.

According to Eric Foster, a rectangular coil with a 3 to 1 or a 4 to 1 ratio will go deeper than a round coil and give more of a blade type field.

Reg

[Tinkerer]

Rectangular DD coil

found this site with info on rectangular DD coils: http://kt315.hotbox.ru/constr_e_ddv2.html
Tinkerer

[Reg]

Hi Tinkerer,

I read the info on the site and wasn't sure about a couple of things. However, I am not fully awake yet either.

I really need to write more about DD coils and how to make them. Now, one thing nice about DD coils for a PI is they can be very forgiving. One doesn't have to be nearly as critical with the alignment on a PI coil as one would with a DD coil for a VLF type machine. Actually, one can be very sloppy in their alignment techniques and not notice much difference.

The one thing that is critical is the polarity of the windings. In other words, the phase of the receive with respect to the transmit has to be correct for the coil to work properly. A person could try to keep track of just how the windings are wound to assure the polarity, but simply testing them and swapping if necessary works just as well for me.

The key is to test the coil when connected to the machine and then make sure to use a proper target and test that target as it passes over the overlap zone located in the center of the coil. Yes, it does make a difference as to what is used as a test target. One should use a non-ferrous object such as a US nickel. If the polarity is incorrect then gold type objects will generate a negative response, rather than a positive one.

I will try to post more about DD's on the thread pertaining to them or start a new thread.

Reg

[bbsailor]

Reg,

I would enjoy reading about some of your experience with DD coils for a PI machine.

The separation of the TX circuit from the RX circuit provides a good opportunity to optimize each circuit independently. Based on the intended targets, there should be different coil designs. Your experience should be very instructive.

Wire size/type and peak TX pulse current can be optimized for the minimum delay time. Since the RX circuit is separate, the RX preamp gain can be adjusted to optimize the variables of RX coil inductance, the number of turns, wire gauge, wire type (single conductor, stranded or litz), coil diameter, and the damping resistor.

I would like to hear your experience about the combinations of the above variables that work well or where you think more research needs to be done.

DD coils for a high power PI machines (lower frequency) appear to be designed differently from low power PI machines that operate at higher frequencies. It would be very instructive if you could eloborate on the coil designs for each type.

Shielding, cable capacitance and total circuit capacitance are critical at minimum delay times. It would also be very instructive for you to include your experience with MOSFET capacitance, coil capacitance and coil shielding to achieve optimum coil speed.

Techniques for building coils to fit pre-made coil housings would also be very instructive. The mechanics of achieving coil stability are learned by trial and error. Your experience could help us shorten our trials and minimize our errors.

You could probably cut and paste many of your previous posts to make a very comprehensive DD coil construction article. This would make for a very good extended post or published article.

bbsailor

[Reg]

Hi bbsailor,

Jeez, you make me sound like I know what I am doing when it comes to DD coils. Thanks, but I am still learning also.

First, let me say that some time back I couldn't find any info at all about using or mounting a DD coil on a PI so I just jumped in and tried to build one. I was lucky, the first one worked and I have been lucky ever since.

Are my designs the best that can be done? Nope, I am still learning just like everyone else. So, I just want people to know that I am not the "expert" on the best design in the world.

Now, with that said I will touch on some of the questions you have asked regarding my experiences and what I use to build DD coils.

First, people should realize that it is not that difficult to make a DD coil work on a PI. The coil design doesn't have to be perfect, nor does it have to be exactly balanced. In fact, one can be very sloppy and make a DD coil work. That is the beauty of PI's. On the down side, some care does have to be taken to try to maximize depth capabilities.

Ok, when I first started out, I made my transmit and receive coils the same inductance. Since then I have come to the conclusion that such a design may not be the best. It will work fine, but by making the two different, there is less likely to be any direct objectionable crosstalk between the two. This idea is more of an idea than a proven fact. However, anyone who has worked with control systems and some of the stability factors knows that when it is best to keep the time constants different.

Also, I measured the ML DD coil and found they used different values also. So, the idea of making the two inductances different appealed to me. Now, I sort of match the ML inductances of 300 uh for the transmit and about 450 for the receive. These are good numbers to work with, especially for nugget hunting coils. Also, by matching the numbers, it is easier to use ML compatible coils on my PI since all coils basically match.

So, if you want to use factory coils such as those compatible with ML PI's, then one should try to match them inductance wise or there could be problems with getting things to dampen correctly and work at the shorter delays.

Now, I place the receive damping resistor in the coil connector so I can change it should I decide to make the coil operate on a totally different machine. Also, I have been modifying my machines and finding different ways of making them work a little better. In some cases, this means making some changes to the coils also. Unfortunately, my early designs had resistors mounted inside the coil housings which makes it impossible to change them. It does make the coils more like the ML compatible coils where the receive damping resistor is in the coil housing, but doesn't allow for dramatic chagnes this way.

It is extremely difficult to try to explain just what is going on at the decay time since I really can't determine certain things. Some strange things have come up over time that shows that it is extremely difficult to get things to work right at delays less than 10 usec. Most of these things are confidential and can't be discussed.

However, it borders on impossible to make things work much less than 10 usec anyway, especially when working with higher powered units.

Now, as for coils, windings, type of wire, etc, there are more strange things on here that I can't explain. One would think that basic theory would say that reducing the coil capacitance makes it easier to get the delay down. Well, that is not necessarily true. As strange as it might sound, sometimes the capacitance becomes a friend, especially when trying to make a coil work at delays of 10 usec or less.

As for the type of wire I use for the coils, that depends upon the coil, and the size of the coil. I have found my cheap Litz wire works very well for smaller mono coils. This is the cheaper served wire sold at Surplus Sales of Nebraska.

Larger mono coils made using Teflon coated wire seem to work best.

As for DD coils, I am not sure just what works best. I really haven't had time to try different combinations of wire to see what is the best to work with.

As for shielding my coils, I have used lots of things including some strange items such as metallized ribbon that has a coating of aluminum on one side. I have also used a conductive rubber tape normally used for high voltage applications. I have also used pieces of conductive material I purchased from EMF. I just cut strips and used them for the shielding.

Overall, my favorite is still the conductive copper plated tape. However, I have used a similar tape that is silver plated. Now, 3M no longer makes the original tape I used but makes a nickel/copper plated tape. It seems to work well also.

Finding a useable cable with minimum capacitance is critical also. So far, I have not found anything better than the Mogami cable. This works very well. On the down side, it doesn't have an extra outer rubber coating over the two coax's so they are more suspectible to damage. I have slid the wiring into a simple latex tubing for protection and it works ok.

As for Mosfets, I still like the lower current rated ones. They do get warmer but they are faster also. However, again, the coil and the FET's interact such that it makes it impossible to say just what wire/FET works best. In other words, one type of wire might work better with an IRF 740 FET and another wire might work better with one of the lower current rated FET's that have less capacitance.

Again, I am not sure just what is going on in the relationship between the FET, the preamp, and the coil wiring to say just what combination is the best. They all seem to interact more than one realizes. So far, I haven't been able to figure out the exact mechanism involved.

Using a coil form makes coil building much easier. If possible, this is the way to go. Right now, Bill Hays is the only active one out there that I know of. Bill is extremely cooperative and will work with most anyone to help them get the form they want. Also, his willingness to build different forms such as the rectangular ones is a real asset.

I know that Charles of NY makes some coil housings, but haven't discussed it for a while to see just how enthused he is on the project. Hopefully, he will make some of his standard forms available. Maybe he will post a suggested price also.

Now, much of what I have written may seem ambiguous but there are no clear cut answers until some specifics are in place and even then, it is difficult to pin things down.

When I get a chance, I may post a step by step procedure on building a DD coil for some machine such as the Hammerhead. The connection diagram and layout has already been written but could be refined a little to make it clearer in some aspects also.

Well, that is enough for now. A book could be written to cover all the questions you have asked. So, I just skimmed over them.

In the mean time, I encourage people to try to build a DD coil and get it to work on their PI. Once they have succeeded it will be easier to build different size and shaped coils. The basics have been laid out to build and install one. Now, it is just a matter of jumping in and doing it.


Reg

[nncunew]

Sorry

I don't understand, remember my bad english

20 awg is a good or bad wire gauge ????

do you recomend use another gauge ???

5 1/2" y 12" coils can be make with 20 awg ???

is truth that you have to wiring the coils in planar shape ???

Thank, newly

[email protected]

[Reg]

Hi nncunew,

Your English is fine. My explanation was bad.

What PI are you building? Yes, 20 awg is a good size wire for most PI detectors. Actually, you probably could use any size between 26 awg and 18 awg and it would work fine on most PI's. the 20 awg would work for most transmit coils on a VLF detector also.

The coil windings do not have to be wound in a planar fashion. Just simply wind them on a form and then tape or lace the windings so they are a tight bunch and the windings are not real loose. Any vibration between windings can make the coil noisy.

Winding coils in a tight planar fashion is done to reduce the capacitance and makes the coil work at shorter delays. However, just changing types of wire can do this also. Magnet wire will have the greatest capacitance and thick insulation will usually have a much less capacitance. Teflon coated wire is one of the best to reduce the capacitance because of its dielectric constant. Generally, a coil will work at shorter delays as the capacitance is reduced. This is true to about 10 usec and then things change.

Some people put a coat of varnish on the windings to hold or bind them together. That is ok but some varnishes or glues can have a negative effect. I am not sure which are good and which are bad, but any material between the windings can affect the capacitance of the coil. Just how much of an effect is not known because it varies with the material. Normally, this isn't a problem until you want to have a delay of less than 15 usec or so.

Reg

[Boattow]

Rectangular 3:1 coils

Hi Reg,
It sounds like you didn't see any of Eric's posts and I haven't been over to his forum to locate them but I do recall Eric discussing the 3:1 ratio rectangular coil saying that it works better than a round coil. He said the only reason he doesn't make them is because people expect coils to be round. The rectangular coil looks odd but works better as long as it has at least a 3:1 size ratio.
Good luck,
Boattow

[Reg]

Hi Boatow,

Thanks for the info but the 3 to 1 ratio idea did come from Eric. What I can't find is any specific info about just how effective it should be and I have searched the PI forum.

My initial tests using a rectangular coil were pretty crude since I don't have any way of minimizing local noise and that can have a dramatic effect on performance. To make matters worse, I introduced a problem into my test PI a couple of weeks ago that affected all results. Fortunately, I found out what was causing the problem and corrected it tonight, so hopefully, I will be able to do a little more accurate testing.

What I have learned so far is with a mono coil, I don't lose that much depth when going to a narrow rectangular coil but do pick up some sensitivity to really smaller stuff because of the narrow concentrated field. Depth capability appears to be better in rectangular form than if the windings were to be reshaped into a round coil. As an example, my test windings in my 3 1/2" by 12" are about equivalent to a 8 1/2" round coil. However, the depth capabilities of the rectangular coil seem to be a little better than a round coil. In fact, the rectangular coil about matches my 9" by 12" elliptical for depth capabilities when testing a nickel as a target.

Overall, I am impressed with a rectangular design and certainly will be using one in the future. It is quite possible that it will be my main type of coil I will use, especially when using a mono coil.

Reg

[nncunew]

Ok, Reg

I am making the sandbanks locator, but using a MOSFET.

I have been saw that when anyone use a MOSFET can increment the power of the pulse, this is helping for the use of more diameter in the coil's wire.

In the most probe that I make never obteined good result the locator do not catch a 1 inch coin (silver) for more of 6 cm. And i think that something is wrong.

Yesterday i make a coil with these specifications:

8.2 inch ID
aprox 10 to 10.5 inch OD
20 awg
28 Turns

is recomended for Gary in your WEB, but i don't probe yet.

I have not information of how calculate the optimun damper resistor, or some práctical method using a osciloscope.

Thank,

nncunew

[Reg]

Hi nncunew,

First, I have not built the Sandbanks PI so I don't know much about it. I will try to help you by giving general information.

First, my guess is your new coil has an inductance of somewhere between 500 uh and 700 uh. As such, I would start out with a damping resistor of 470 ohms and see what it looks like.

You can look at the output of the preamp (IC 4, pin 6). The signal should like much like the top trace in the picture I have attached. The bottom trace is what the signal should look like at the coil.

You will notice in the top trace, the signal goes negative for a time and then rises exponentially and levels off. You want the signal to rise as fast as possible but not oscillate. If your signal looks like the top trace, you might want to increase the resistance of the damping resistor and see if it still looks similar. You can keep doing this until the top signal becomes unstable and has a wave or two in it when the signal rises.

Now, just looking at the Sandbanks circuit, I woud recommend you try to shorten the sample and delay time also. That will make the detector more sensitive.

Reg

[nncunew]

I understanded

So good,

I undertended all that you said me, it looks very important and practical.

I will probe this aspects and report you the results.

Thank for your help.

nncunew