Delay times

[Ray Huffman]

Re: Forgot to mention

Reg,
The pic helped in comparing with what I see on my scope. I have built a concentric coil and used 2 different wire sizes, with seperate cables I can switch them out for experimental purposes. on one coil I can get about a 5 or 6 usec delay and with the other its is up to 15 usec obviously higher capacitance. I am now going to build a new one using the same 22ga hookup wire (with the 5 usec delay)in both coils and look at the delay times and try to get them both about equal. Dont know if that is important or not9equal delay times). I was trying for higher sensitivity in the receive coil so had about double the number of turns and used 24 ga enameled wire. I have the rx circuit and the tx seperate. I also think the concentric gives me a better option for experimenting with differnet coils for both.
Ray

[Reg]

Re: Forgot to mention




Hi Ray,

I am not sure what you are saying about the concentric coil and "can switch them out". If this means, you tried to use the inner or receive coil as a transmit coil to check the delay time of that coil, then yes, I would expect the two times to be quite different. This is especially true if the "receive" coil has "double" the turns. Obviously, doubling the turns will increase the inductance considerably which will result in a much longer decay.

If you built the concentric as explained on this website where you have a second xmit winding wound on the inner receive coil, make sure you have the polarity correct on this secondary xmit winding. It will make quite a difference of how the coil works, decay wise.

To be honest, I find that if the signal at the receive coil is stable at the time the sample is taken, then the coil will work fine. So, obtaining the ideal curve isn't a necessity for me.

To explain what I am saying, I have attached a pic of the signal I get from a ML DD coil on my PI. As you can see, the signal isn't pretty, nor is the decay curve ideal. However, the signal is stable at 10 usec or even at 8 usec. So the coil should work fine. Actual testing indicates it does.

Would I use a coil with this type of response? Yep, as long as it is stable when I need it to be.

Could the signal be improved on the attached pic. Yep, but the coil is primarily designed to work on a ML PI. I have just built an adapter to see if it would work on my PI. Rather than try to modify a coil that would probably not work on the ML, I would just use it the way it is.

Reg

[Ray Huffman]

Re: Forgot to mention

Reg
thanks for the pic. I understand what you mean from looking at it.
one other thought though, do I see about a 12 cycle of overshoot on your scope. This reason I ask this is I have tried to damp the signal to get no overshoot. It looks like a 12 cycle is probably ok.
Ray

[Reg]

Re: Forgot to mention

Ray,

Nomally you can obtain a faster decay response on a DD or concentric coil by slightly underdamping the xmit coil. At least, that is what I have found. On the pic posted of waveform from the ML coil, the xmit coil is underdamped too much in my opinion but it works ok. Since the coil is really built for use on a Minelab PI and not for my PI, I would just use it the way it is so it could be switched back to the ML without any changes when needed.

I posted the pic to show what can be done with other PI coils and what will work and does not have a perfect decay response.

Reg

[Eric Foster]

Re: Forgot to mention

Hi Reg and Ray,

I have not found that overshoots or undershoots that occur before sampling make any difference to detection performance. As long as any ringing is damped out by the start of the sampling period, then all is OK.

Eric.

[Ray Huffman]

Re: Forgot to mention

Thanks all.
I will take that into account when using a damping resistor. I probably lost a lot of sensitivity by too much damping.
Ray