This is my first Instructable, and is not a complete how-to. This is intended to be an addendum to JoelDude's very helpful Instructable, Home Made BFO metal detector.

Some background on this type of metal detector:

A BFO metal detector is a "beat-frequency oscillator" type metal detector. This type of metal detector uses two wire coils, one which is large (detector coil) and one which is small (reference coil). The two wire coils are each connected to capacitors to create two circuits that oscillate (here is a good video explaining the fundamentals of this).

When everything is working correctly, the two coils, while being different sizes, are intended to have the same inductance. In use, the presence of metal near the detector coil changes it's inductance. This causes the two coils to have a different inductance, and the electronic circuit amplifies this difference and makes it audible through the speaker in the circuit. When you hear tones coming from the speaker you know there's some metal around.

Step 1: Visit These Links First! (reference and Credits)

Just like JoelDude, when I first set out to make this metal detector, I found the website below. This is an excellent starting point and as you will see if you have already been Googling around for a bit, this is where many of the other instructions for building a BFO metal detector originated.

http://www.easytreasure.co.uk/bfo.htm

But exactly as JoelDude points out, this website is helpful but not entirely complete or precise (at least for amateurs like me!), and JoelDude's Instructable was critical for me to get closer to being able to build this metal detector. JoelDude's Instructable is here:

https://www.instructables.com/id/Home-Made-BFO-metal-detector/

Other reference that may or may not be helpful

The video below was what convinced me to use a glue stick for a reference coil. Also, if I could have pulled it off, his hand-held format seems like a nice way to make this whole thing more compact and light weight.

Handheld BFO metal detector made with glue stick (video)

There was also some useful in-depth detail on the fundamentals and electronics of BFO metal detectors at the website below. While most of it was over my head, some of it helped me confirm or deny assumptions that i was making along the way.

http://www.talkingelectronics.com/projects/200TrCcts/MetalDetectors/MetalDetectors-1.html#SimplestCc

Step 2: Build the Circuit Board

Again, these are my addendum(s) to JoelDudes excellent Instructable on building a BFO metal detector. I will reference his steps as i go, so if any of what I've written here is helpful, you may find yourself going back and forth between his Instructable and this one.

Build the electronics circuit as described by JoelDude here. This step is fairly straight forward if you have any experience with simple circuit schematics. If not, you may still be able to make this with some google and dead reckoning. I too was able to find all the electronic components at Radio Shack, but like others i had to buy a mixed bag of resistors to get all the sizes i needed.

Main addendum:

1.) The parts list:

Power:
9v battery
9v battery connector (if you don't already have one)

Resistors:
Minimum 1/8 watt (higher is fine) 5% tolerance
1 x 1K
6 x 10k
2 x 39k
1 x 2.2m (2.2 Mega ohm)

Capacitors:
Minimum 16v working voltage (higher is fine)
5 x .01uF polyester (sometimes called "Mylar")
5 x .1uF polyester (sometimes called "Mylar")
2 x 220uF electrolytic

Transistors:
6 x General-purpose NPN type
(medium control switching 2N222 & 2N4401 transistors are what i used, found here)

Other components:
1 x speaker or piezo buzzer (that can operate at 9vDC)
(the speaker or piezo buzzer is driven by the circuit, don't get a piezo "tone sounder")

2.) One clarification is that, in the diagram there is (what looks like) a junction labeled "NC", this means "Not connected". So the wire from the 10k resistor doesn't touch the battery ground ("B-") it's just connected directly to the Q3 transistor base lead. (JoelDude helpfully added this note to the original schematic that did not have that indication, thanks JoelDude)

Other notes:

1.) I built my circuit on a small breadboard, if you're like me, and an electronics novice, breadboards are your best friend. They allow you to plug and unplug components from the circuit as often as you like. The only thing that can be confusing is that you will occasionally have to use a wire to jump around on the breadboard when components aren't close enough to share a rail.

2.) I used a piezo buzzer instead of a cone speaker, it's compact and easy to use directly on a breadboard. You'll notice in my images that if you go this route you'll need to make sure you have some space on the board for this item (if it doesn't have lead wires) since it will take up the most room. (the first time i built the circuit i got to this last component and there was no room on the breadboard for it and i had to start over... Gah!)

Step 3: Make the Pick Up Coil

Again, these are my addendum(s) to JoelDudes excellent Instructable on building a BFO metal detector. I will reference his steps as i go, so if any of what I've written here is helpful, you may find yourself going back and forth between his Instructable and this one.

I would suggest making this coil last. I found the pick up (reference) coil to be the item that required the most messing around with. I had to have everything else working before i could verify i had made this correctly. So i would go on to the next step and then come back here afterwards.

...But, if you've already done this, or don't care, without further ado here's my addendum to building the reference coil.

This coil will be wound around the glue stick, the number of winds is intended to match the inductance of the detector coil. For me this meant 60 turns, (but ultimately i needed about 70 turns, more on this below). This first image above shows how i arrived at the specs for this coil. The air core inductor simulator allows you to indicate the diameter of the coil, and the wire gauge, you then you drag the length slider to see the "turns" and inductance change. If, like me, you already wound the detector coil, then you have the dimensions, and turns for that coil. In the simulator i created this coil using the length slider to get the right amount of turns (in my case 11). I then noted the inductance (0.072mH). With my inductance noted for my detector coil, i then plugged in the diameter of the glue stick, and dragged the length slider around until i was close to 0.072mH of inductance, the simulator indicated 60 turns, so that's how many times i wound the wire around the glue stick (leaving a lot extra on both ends for more winds if i needed them)
(by putting double sided tape around the glue stick i was able to keep all the winds from unraveling as i went.)

Having the winds on the glue stick i then set about testing my metal detector. I sanded the ends of the winding wire on both coils and spliced them to some bell wire to insert into my breadboard circuit and turned it on...Then i got... nothing.... just... nothing... BUT THEN! I found that if i waved this reference coil around very close to the detector coil i could get my speaker to make some sound (I was inspired to do this after watching this video). This was how i determined that my electronic circuit was built correctly, even when i did not yet have a reference coil that matched the detector coil inductance. To further test my reference coil I held metal screwdrivers inside the glue stick casing, which created different tones, this assured me that my reference coil wasn't way to far off from the inductance i needed.

To get the reference coil windings correct, I attached the metal cylinder to the black plunger of the glue stick and using the glue stick's thumb wheel, screwed it down inside. While it was making all kinds of tones i would unplug the coil and wrap or unwrap a few turns and plug it back in. Then I would use the thumb wheel to move the metal cylinder in and out and see if i was closer to a balanced coil. Eventually I got to lower and lower pitches, at some point it was low enough that turning the thumbwheel and moving the metal cylinder was able to make the tone totally stop. At this point i waved some metal (a big piece) around the detector coil, and sure enough the tone was back... It was detecting metal! +1 for homo sapiens!

Main addendum:

1.) Because of the fact that this component has the most flexibility in how it will be made, i made it last, so i could try it while it was plugged into the circuit with the whole thing was turned on.

2.) Like i mentioned at the beginning, a glue stick ended up being the perfect item to wind this coil around. The reason is that you can use the glue sticks moving plunger to attach a piece of metal, that will be the "tuning element" for this reference coil. This arrangement is a nice way to simplify the arrangement that EasyTreasure and JoelDude use.

3.) This tuning thing is necessary because it seems to be nigh impossible to simply wind a reference coil that will exactly match the inductance of your detector coil. This moving piece of metal inside the reference coil is what will allow you to tune it whenever you turn on the metal detector. By moving the chunk of metal inside the reference coil in and out, you are changing the inductance of the coil. By listening to the output sound you know when you have brought the reference coil in tune with the detector coil. For an example of this watch the video here.

4.) While i don't know why, i noticed that moving around and crossing my leads coming off the reference coil caused the behavior of the coil to jump around a lot. So just beware that if you can get your work area and wires to stay put while you'r testing and changing the windings of this coil you may have less frustration than i did!

5.) While i used the guidance of coilgun.info's inductor simulator at http://www.coilgun.info/mark2/inductorsim.htm to determine how many turns i needed on my coils. I found that I had to add about 10 or so more winds on my reference coil than the simulator said to. What really surprised me was that i could arrive at this working solution largely by trial and error! I was able to wind and unwind the reference coil and move the tuning chunk in and out until i found a sweet spot where it all worked.

- This difference between the simulator and my final coil might be because one of the windings on my detector coil overlapped some of the others. I don't know, but this unintentional overlapping wire on my detector coil might have changed the inductance enough for the simulators inductance info to be slightly different than my actual real world coil.

- Quick note on a mac this inductor simulator does work but you need to run it in Safari, and you probably will have to download the java update and install it.

Other notes:

1.) None of the windings overlapped on this coil, but they were not all perfectly next to each other. Seemed to be fine.

2.) It's fine to have a bunch of extra winding wire coming off the coil as you test how many winds you'll need, just make sure they are straight(ish), and not looping around (creating more magnetic fields!).

3.) The cylinder of metal that i attached to the plunger was a just something i had lying around. It happen to fit well so i used it. If i hadn't found it i would have glued together a stack of washers or a stack of nuts or something, i don't believe the shape matters all that much.

Step 4: Make the Search Coil

Again, these are my addendum(s) to JoelDudes excellent Instructable on building a BFO metal detector. I will reference his steps as i go, so if any of what I've written here is helpful, you may find yourself going back and forth between his Instructable and this one.

I found the detector coil to be the easiest part to build, even though it was the most work. I cut some circles out of 1/4 inch plywood, glued them together and then drilled some holes in those to reduce the weight. It was important to have a form to wrap the wire around, so the two pieces glued together make a simple form.

I then sprayed some spray mount on the edges and started winding my magnet wire. I didn't worry to much about the size i just ended up making it 24cm in diameter (that was the largest size i could get out of my plywood piece) and i wound it the wire 11 times (after looking at several instructions from across the the web that seemed like a rough average of windings that folks were using for their detector coil). An experts might have more insight into how the detector coil design will affect the overall sensitivity of this type of metal detector.

Main addendum:

1.) While the windings largely don't overlap, one or two windings did, this didn't seem to be an issue.

2.) I used Spray Mount to make the surface sticky and the wire held fine for me.

3.) This small bracket to hold a handle seemed to be the easiest way to get some angle from the handle to the detector coil. I also ended up using an old metal broom handle for the shaft (cut in half). I thought that the extra metal might have an effect on the detector, but it seemed fine.

Other Notes

1.) All my winding wire was 28 gauge enameled coated magnet wire that i bought on amazon, I couldn't find it at Radio Shack or anywhere else nearby. If i made this again i would have tried to scrounge the magnet wire from an old transformer or something, it wasn't expensive, but still...

2.) Where i've needed to splice the winding wire to longer lengths of wire, I just did a sloppy twist, twist, twist, twist thing with the other wire and then taped it down real good. for all the other wire i just used some bell wire i had lying around. The bell wire mates well with the solid core enameled winding wire and inserts easily into a breadboard.

3.) DON'T FORGET TO SAND THE ENDS OF THE WINDING WIRE BEFORE SPLICING IT TO OTHER WIRES. The enameled coating looks almost invisible as is, so it's easy to forget that it's essentially insulated wire until you sand off that coating to expose the copper.