metal detector

Hi Ivan
http://www3.telus.net/chemelec/Projects/Metal/Metal.htm

I should have said I don't know how to use them very well.  As far as the workings of them I have spent many hours studying all types of them and played with a lot of different ideas.  Like this one.
http://www3.telus.net/chemelec/Projects/Metal/Metal.htm
and also the simple BFO type.  
The one this thread was on is just a simple CD4011 duel oscillator similar to one Harry used to have here made by one of his members that used a cmos chip as well. I found that trying to drive a little speaker from the same power supply shifts the frequency and stops it. My 9v I am testing it with is low too so that might also be the problem.

I got on this kick playing with a distance sensors with a single transducer. I found I could set the TX/RX pulse what I though was perfect on the scope and I never got a usable signal back from the receiver and I wanted to see if I could figure out what was going on but I never did.  I still enjoyed playing around with circuits like that and I guess I am sill doing it for a while at least.

Hi Jalex,
if you "know almost nothing as to how it should work", start by reading a book or web pages about metal detectors. They made a huge progress since the WWII. It is wasting your time when you are reinventing known principles.
The LM386 probably requires a grounded resistor or potentiometer directly on its output AND DC separation of preceeding circuits by a capacitor. This way the LM386 can maintain the proper bias on its input.
You can try an opamp with a current booster, too.

BR from Ivan

Hi Harry
I put together a simple cd4011 circuit and it works pretty good but I haven't figured out how to come up with a simple audio amp with a couple transistors yet. I tried an LM386 and it's didn't work the way I wanted either.  It works fine with my little signal tracer but when I add an amp it either has no sound or messes up the detector. I added a 22Ouf  cap to see if that helps and it does but I am still not happy with it.  Now that I have a real good metal detector I don't know how to finish it off. I also like playing with the TX and RX type detectors but I haven't made them work yet. Using the same coil for TX/RX I must be getting my switch timing off or something as all I have ever gotten was garbage from the amp. I think a lot of it comes from the ringing of the coil after the TX pulse goes away too and I haven't had much luck suppressing that either. Of cour


se I never had a metal detector so I know almost nothing as to how it should work.

Try putting a voltmeter on the output of the last gate and you will see a beat-note of just 1Hz or less so you can have a much better sensitivity, only by changing the operator interface.

Another method I liked was to have two coils, one fed with RF and the other positiond 1/2 over the other at the point there is a deeeep signal cancelation. Metals disturb the cancelation. You can pinpoint exactly there an object is with this method because there is a centre cancellation. But you cannot tell if it is ferrous or non-ferrous metal.

It used to be common to wind a strip of aluminium foil around the coil, but with a small gap so as not to form a closed loop (shorted turn). This will reduce capacity effects by about 95% so the frequency does not alter when the coil is placed near soil.

Best regards Harry - SM0VPO

Hi Harry
I saw a metal detector schematic once that used a Fariday shield but from the drawing I wasn't sure how to build it. 
I was mostly just playing but I did come up with a detector.  I tried using an LM567 PLL as I have lots of them but It didn't work. It was working like an FM radio and I was picking up noise from my fluorescent desk lamp where the pick-up coil was. I thought I could use the lead to get it locked and then monitor the vco voltage to see when there was metal near it.  I kind of thought it might be a little better than std audio type.  Then I tried the common CD4011  Oscillator. One for the pick-up coil and one for the compare. Then tied the two outputs to the last gate. It seems pretty good and runs at around 180Khz.

I will just throw in a little comment here. In 1980 I built a bit of test gear for detecting the frequency of an oscillator to 0.01Hz accuracy. This can be used to make a VERY sensitive metal detector, but you need a LOT of care with the oscillator loop. You need a Fariday shield around it. But that you can make using aluminium foil, as long as it is not a closed loop around trhe coil.

Use the same oscillator as the drawing and use the CD4046 VCO to make exactly the asame frequency oscillator. Connect both oscillators to the two comparator inputs and feed the output of the comparator to a simple voltmeter.

Adjust the VCO so the meter varies up/down continuously from 0v to 10v nice and slowly. This way you can see the slightest change. You can even adjust the VCO so the meter stops to get the absolute most sensitivity.

Using speakers and such you do not hear a difference if it is less than about 30 Hz. With the NOT-LOCKED LOOP (NLL) you can see easily a change of 0.1Hz. If the needle moves slower or faster you can tell which way the frequency is changing: up or down, thus identifying ferrous or non-ferrous metals.

Best regards from Harry - SM0VPO

Hi
Ivan
I came up with this idea while experimenting with PLLs. After looking at the net I see where other people have done it but a lot of the pages said they had trouble with them. I thought it would be a great way to be able to see exactly what the frequency is doing when you run the vco voltage to a meter. I think it's another thing that looks simpler than it is.  I  really don't know anything about metal detectors but was interested in a way to tell the difference between ferrous metal and non-ferrous metals and I thought this system would show that better. 
 The pin that threw me the most was the reference pin. The LM565 was designed for FSK,  I think and that must be the difference. 


Thank you for the help on the video  sync stabilizer I was working on.  I had some videos with bad sync and it worked just great.  I opted  to stay with the original design and got the NE564 chip after I saw the amount of changes I would need to make.  Both those PLL's lock solid as a rock. It strips out all the sync,  blanking, and color burst then re-inserts new ones and makes the video look great again. Of course it doesn't enhance the video but that's ok.

Hi Jalex,
IMHO you will be missing some pins on 4046. It differs from 565 quite much. In fact you will have to design a new schematics.

BR from Ivan

http://www.electroschematics.com/6616/homemade-metal-detector/
 Hi 
I would like to try to build this circuit with a 4046 PLL as I have lots of them. Would that be possible? I have the spec sheets on both but there are a lot of pins explained differently. I don't see a reason why it wouldn't work but I could be missing something.