PI Detectors....eliminating magnetic response?

[bchampig]

Hello,

Does anyone have any tips for eliminating a PI detectors response to magnetic effects, or hot rocks?
Is the secondary delay of many PI detectors the key to this problem? I have built the Goldscan and Hammerhead detectors, and the secondary delay adjustment doesn't seem to help with this problem. Perhaps I need to adjust the secondary delay more precisely(multi-turn pot) to 'tune out' the magnetic response from moving the coil in air, or waving a magnet near the coil with the transmitter off?

I am using a microcontroller for timing, so I have lots of flexibilty. I appreciate any suggestions or ideas.

Best Regards,
Brian

[Reg]

Brian,

I am not sure which hotrocks you are trying to eliminate but many of them can't be eliminated. The magnetite ones can as can ferrite magnets but only if they are a short distance from the coil.

The second sample is designed to eliminate slow responses, which are what these magnetic lag signals are. So, if the balance between the main response and the later response is correct, the hotrock and ferrite problem should be gone. The two signals need to be the same amplitude for this to occur so the two signals can cancel.

One big mistake is improper shielding that alters the sensitivity of the initial signal, thus altering the balance. Another error is sampling too soon and not letting the preamp recover completely, thus the gain is not what one thinks it should be. To check if either of these conditions are the problem, simply advance the delay a little at a time and see if the magnetic effect is eliminated at some point.

A few years ago, Eric Foster provided a post displaying the lack of gain at short delays that can cause this condition. Hopefully, a search will display his findings. Right now, I don't have a link.

Reg

Here is the link to the info I mentioned. Eric posted using the name ferric toes.

http://www.geotech1.com/forums/showt...d+noise&page=2

Look for the info about the gain loss. Eric displays a pic of what is happening and an explanation of why he went to a dual preamp.

[bchampig]

Thanks for the info Reg. Up to this point I have done some limited testing using a refrigerator magnet, and shutting the transmitter off. I was getting a strong response from the magnet and couldn't eliminate it by varying the secondary delay. I will do more testing at different target ranges, as well as play with gains.

Best Regards,
Brian

[Reg]

Brian,

I wouldn't try to shut the transmit off and eliminate the earth field effect. You should be able to let everything run and eliminate the magnet signal. If it does, then all is working correctly.

Reg

[bchampig]

Thanks for the suggestions Reg,

I was just trying to isolate the problem by making the detector only react to the magnet, and not any other metallic objects in the area. Probably not a realistic test without the TX running.

It did get me thinking about a selectable mode on the detector that could possibly be used to check for hot rocks...basically shutting down the TX, and if there is a magnetic object in the ground, the signal would still be there. Am I wrong in assuming that this would work? I'm new at this and not even close to an expert on ground and hot rock composition.

Thanks,
Brian

[Reg]

Brian,

You are trying to make a poor magnetometer. I suppose if you worked at it hard enough you might get something to work.

If you build the right PI, many if not most hotrocks will be ignored, as will ferrite magnets. So, neither will be a problem. Most of the rocks that then will sound off on the PI, probably wouldn't make much of a signal if any on your magnetometer.

Now, with that said, you can still go through with the experiments and maybe find something that will help you. Since I only tried to do what you are trying to do briefly a long time ago, maybe you can find a better solution.

Reg