BLACK HOLE ANTENNA

[Davor]

Sorry to dig out a topic after so much time, especially because it had gone completely OT, but has anyone seen any measured evidence on sensitivity improvement of the "black hole" design? I know of sensitivity improvement on loop antennas with increased Q (proving the point), but it is not at all dramatic.

As mikebg says, the joke is that there is no joke, but how funny it is happens to be a very elusive thing.

[lonelyWOLF]

hi mikebg..

R33 should be 1M..

regards..

[MotoLupe]

The PI Controller Block of the Enhanced Houndog

Sorry, I was refering to the PI Controller Block of the enhanced Houndog Detector from this Link http://www.geotech1.com/forums/showp...90&postcount=8

[mikebg]

Thanks for the clarification, Looks like I am in for a voyage of discovery, I would prefer Tesla over Eddison R&D but I can brute force it if necessary, do you have the specifics on the PI block of you enhanced Houndog circuit. Or any recomended circuitry.

Motolupe, I can't understand the question. For what purpose should be designed the recomended circuitry?
Here is the second page of circuit 428_1:

[MotoLupe]

My Bad I meant DOD Coil

Thanks for the clarification, Looks like I am in for a voyage of discovery, I would prefer Tesla over Eddison R&D but I can brute force it if necessary, do you have the specifics on the PI block of you enhanced Houndog circuit. Or any recomended circuitry.

[mikebg]

MotoLupe, "DODd" means 4 windings: L1 - transmit, L2A and L2B - receive and L3 for reference voltage. Below is given an example for use of coil "d" or L3 in meteorite locator.
However we can realize extracting of carrier frequency without winding L3, for example using voltage induced in L2A.

[MotoLupe]

DODd Head & Houndog

Would it be possible to continue the converation In a thread geared to the houndog like Houndog Revised and be construction oriented for entry level designers and builders. I like it because it's different and elegant, I saw your rectangular coil orientation and saw I could expand on the detection area, I am planning to move onto a PI unit, I want to build up to it and not get bogged down in the "Ferris Bueller, Ferris Bueller" physics until later, It's learning in reverse, Anyways either way is fine.

[mikebg]

Let's move in other thread

As noted MotoLupe in posting #36, the title "Black Hole" appears misleading for our discussion in this thread.
I will continue the discussion in a thread having more suitable title: "Differential coil", despite the most suitable title is "Planar gradiometer".
http://www.geotech1.com/forums/showt...eferrerid=2910
http://www.geotech1.com/forums/showp...6&postcount=25

[maikl]

Maikl, the oOo loop configuration is widely used for demining because of its excellent pinpointing properties. Please read patents US3,002,262 and US3,823,365 for details.

Below is illustrated spatial response of this configuration. It seems as derivative from response of a conventional loop configuration. There is a sharp null when target is in the plane of symmetry between RX loops. The processing of such spatial response is described in patent US4,700,139, but with oOo no need of differentiation.

I'm using DOD loop configuration to locate meteorites. They are not deep buried, but my meteorite locator can detect coins at 26 cm depth in bad ground.

Hi mikebg. Mine are large objects on the surface (not deeply buried)... When I thought of poor pinpoint the following: whether the target area in 2A or 2B in the area (see picture). Detect coins at 26 cm depth in bad ground is not bad. Which detector is used with the DOD (OOO) loop configuration? Whether it comes to a well-known brand or your "handmade"?

[mikebg]

I built the Houndog and quite frankly it took a scope to set it up, and I was able to get it to work reliably in the Mode 3 state where it oscilated and target squelched it. I added a comparator to the output to invert and create threshold trigger tweaked on every aspect. I am thinking I will add a Pic or 8051 Microcontroller and digital pots for auto zeroing, also I can take advantage of the Frequency Based Discrimination it exibits and modulate power for depth determination.

MotoLupe, the principle of Houndog is regeneration (positive feedback). The drawback of this principle is for target identification (TID). The soil properties are involved in the loop of regeneration. When soil changes, it changes frequency of Barkhausen.

The path of target signal is shown below as block diagram, valid for all kinds metal detectors. Note that soil operates as two blocks causing phase shift. For TID we need two signals which exist in points Y and X. If we know them, we can calculate H(s) (impulse response or frequency response of target), ie we will recognize the target. A regenerative circuit can not deliver such information, because it not operates with carrier frequency. The Houndog is a variant of wide band Noise induction metal detector suitable for beginners. I think you should design a wide band metal detector without such drawbacks.

[mikebg]

Hi mikebg. What you get with the DOD loop configuration? Have you got something on the depth of detection? Or get a better sensitivity to small metals? DOD loop configuration provides a wider field of detection but the downside is bad pinpoint (especially in the larger coils). And finally I think MotoLupe wants more data for his case.

Maikl, the oOo loop configuration is widely used for demining because of its excellent pinpointing properties. Please read patents US3,002,262 and US3,823,365 for details.

Below is illustrated spatial response of this configuration. It seems as derivative from response of a conventional loop configuration. There is a sharp null when target is in the plane of symmetry between RX loops. The processing of such spatial response is described in patent US4,700,139, but with oOo no need of differentiation.

I'm using DOD loop configuration to locate meteorites. They are not deep buried, but my meteorite locator can detect coins at 26 cm depth in bad ground.

[mikebg]

Why LM386 as a preamp? This chip is quite noisy, and about any ordinary opamp will have better performance, also higher CMRR. Maybe just "low part count" solution

Tepco, the LM386 is enough low noise for metal detecting. The problem of metal detectors is that sensitivity is not limited by noise generated in components, but by three too large signals in input: AIR, GND and EMI. Despite planar gradiometer L2A & L2B suppresst these signals, the remained part of them is larger than component noise.

Before to reinvent the simple input circuitry with LM386, I have designed the following circuit having in input UA733 for symmetry and increased gain. However there was large common mode signal in input because I not use shielding and I have not grounded "center tap" (connection between RX windings).

Note that the circuit in right can operate without connection to COM if there is shielding.
Then radio amateurs adviced me to connect "center tap" to COM and to use only LM386, but in circuit with increased gain (near to 60 dB).

[MotoLupe]

Houndog Improved Circuit & Head

Thanks for the speedy reply, I had allready done an entire Houndog site search and that's how I found it mentioned in the Black Hole Thread, Do you have the PI block completed for the Improved Houndog circuit, and do you have the construction details for the DODd head. I built the Houndog and quite frankly it took a scope to set it up, and I was able to get it to work reliably in the Mode 3 state where it oscilated and target squelched it. I added a comparator to the output to invert and create threshold trigger tweaked on every aspect. I am thinking I will add a Pic or 8051 Microcontroller and digital pots for auto zeroing, also I can take advantage of the Frequency Based Discrimination it exibits and modulate power for depth determination.

[maikl]

Hi mikebg. What you get with the DOD loop configuration? Have you got something on the depth of detection? Or get a better sensitivity to small metals? DOD loop configuration provides a wider field of detection but the downside is bad pinpoint (especially in the larger coils). And finally I think MotoLupe wants more data for his case.

[Tepco]

Why LM386 as a preamp? This chip is quite noisy, and about any ordinary opamp will have better performance, also higher CMRR. Maybe just "low part count" solution