Whites 6 1/2" VLF concentric coil

[nalag46]

Whites 6 1/2" concentric coil

Hi there. Just to let you see really what the finished but unworking coil looks like. I believe I have it connected the right way? The start of the TX winding going to one side of caps and Green cable, the other end to bucking coil start and opposite side of caps and Shield/Earth cable. The bucking coil end going by a shot length (feedback cable) to the White lead. The original feedback length was approx. 3 turns of the RX coil, have played with it, also a longer length and the shorter shown but none make any difference. I was wondering about winding a smaller 90 turns coil (like it has been suggested previously) and see if that one might resonate? What do you think?

[Skippy]

If the current setup does run at 5KHz, and you need it to run at 6.59KHz, then the main TX coil would need to be about 85 turns to get the L right. It's worth a try, but it doesn't make much sense to me - I don't see why it's running at 5KHz now, for one thing. You also haven't tried my suggestion about adding capacitance to the TX L-C circuit, to see what frequency changes occur. This would give you an independant way of calculating the inductance of the TX coil (see a previous post).

[nalag46]

Ok, we seem to be making some progress now. The coil needs to definitely run at 6.59khz. I have wound a new 87 turns coil with extra windings at 89, 91 and 93.5 turns and there are some signs of life now athough I am yet to get it to null. I have found that the Coinmaster 2 DB coil has the five leads but the shielded one that goes to the one side of the tuning caps does not connect to TX or Bucking coils, it simply goes to pcb earth and that is where I was going wrong, trying to couple all of them together. This can all be seen quite clearly on the circuit diagram of the 2 DB I posted. The Coinmaster 2 is identical to my Treasuremaster in every component shown (except for the Hex Inverter Cmos 4069, using instead TTL 7404 which is same thing) but the 2 DB I have here is yet another variation of the model as it includes a variable sensitivity control but otherwise it is all the same. I did try adding other caps to the existing ones on my coil but they do not make any difference and I have measured the inductance of the working 2 DB TX coil at 1.66 mh and that is the same on my new wound 90 turn coil (I do not know how accurate that measurement is as it has been taken with my new multimeter that has the feature but the fact that both readings are the same must be a positive indicator). So, as things stand, by holding the auto reset button down and obtaining some erratic threshold, the coil will pick up metal (largish) close to it. Letting go on the button, tuning is of course absent and it cannot pick up. What sort of feedback length would I need to play with and what else should I look out for to get it to null?

[Skippy]

You're just making things more mysterious...I guess that the two inductance readings match counts for something, however the actual value of 2.6mH is way out, I'm sure the actual coil L is more like 8mH, but I presume the 72nF of capacitance is doing its bit to confuse things.
Re. the null. First, I suggest you need to double-check the relative direction of the main TX and Buck coils is correct. And then you need to assess if the null is even remotely close, I'm not sure how far out it could be. I suppose you could add turns to the bucking coil. When I say add, I would include the option of connecting the extra turns in-phase or out-of-phase, so they add to or reduce the turns effectively. Doing this won't change the operating frequency by much at all, as that is 95%+ determined by the main TX winding. We don't even know how many turns there are on the buck , anyway, it was just calculated from some dodgy measurements. It may have 20 - 40 turns, so adding or 'removing' a couple of turns seems a reasonable start. A good use for your new L meter would be to measure the buck coil L on its own, it's probably 300 microhenries, that sort of area (from memory).

[nalag46]

I am sorry if things look more confusing. I have checked the coils direction and I am sure I got them right. The bucking coil resistance is 2.2 ohms, inductance is 0.062mh (the actual length size is 6.5m or 32 turns more or less. I have checked it by undoing the winding and measuring it end to end) I got to get to the bottom of this now. I am not quitting but I am getting a bit bored with the lack of progress. If it were a DD coil, it would have been sorted ages ago as you only need to add/take some resistance and adjust the coils overlap to obtain the right null. With this one, well, I don't know what else is going to take?

[Skippy]

That's the trouble with concentric coils. In mass production, with machine-wound self-bonded coils, they can be very precise, and once the fine-tuning of the turns is worked out, they are very repeatable. But home-building them is less predictable. It's good that you have only 32T on the buck, we originally estimated 43T. The 87T : 32T ratio doesn't look so bad. As for progress, well, you haven't got a 'scope, (didn't) have a L-meter, no cct diagram, possibly a faulty detector, and lots of mistakes were made. So the fact that it now partially works is some progress. Based on a common ratio of main TX turns to Buck turns of 3.5:1 (I think I used that one early on) it would be probable that less turns on the buck, rather than more turns, would be the thing to try first.

[nalag46]

Well matters cant get any more confusing. Taking turns off the buck coil and trying different lengths is not a simple job specially if it has to be wound back again (it is covered in adhesive and can break very easily)Anyway, I did that, tried 3 less turns, no change. Then, looking at the Coinmaster 2 circuit diagram where there is no bucking coil indication, I thought I tried without connecting the coil and the result is the TX coil works quite well like that (picks up a coin clearly at 6 inches from the center of the coil) however, all is not straight forward, as the disc. control does not alter the metal sensitivity and also, as my hand gets close to the circuit board it acts as the TX coil itself by reacting as it was picking up metal. Can I get away without the bucking coil and instead adjust the TX length either way to optimize sensitivity/threshold etc?

[Skippy]

"Can I get away without the bucking coil?"
Why did the manufacturer go to the trouble of using one then? Without a 'scope, I don't know what to suggest. Is your hearing good? Maybe you could connect a crystal earpiece to the RX coil output, and hear if the high-pitched whistle gets quieter at any particular setup. I don't know whether the DC levels from the demodulator are worth looking at, at least you can measure them without a 'scope.

[Jerry]

Can I get away without the bucking coil and instead adjust the TX length either way to optimize sensitivity/threshold etc?

In a word, No. At least not with a concentric coil.

Jerry

[sdesic]

Hi guys

Can this be applied to Coinmaster 1 DB series 2 coil ?
I have friend with this detector and he wants to upgrade coil - to make a larger one.

Thank you.

P.S. sorry to hijack this thread