Phase delay for different targets

[pebe]

Is there any definitive information published that relates phase delay to type of target. I have not had much luck searching.

The only information I have been able to glean is that iron targets seem to be between 0º and 90º, and more precious metals between 90º and 180º. But I'd like something a bit more accurate.

[Qiaozhi]

Is there any definitive information published that relates phase delay to type of target. I have not had much luck searching.

The only information I have been able to glean is that iron targets seem to be between 0º and 90º, and more precious metals between 90º and 180º. But I'd like something a bit more accurate.

Have a look at this -> http://www.geotech1.com/pages/metdet.../US4486713.pdf

[mikebg]

Is there any definitive information published that relates phase delay to type of target. I have not had much luck searching.

The only information I have been able to glean is that iron targets seem to be between 0º and 90º, and more precious metals between 90º and 180º. But I'd like something a bit more accurate.

Pebe, I made so much postings in the forum for this, that is impossible to show where are. Because you measure phase lag relative to TX voltage, yes you are right, but iron targets have more than 90 deg at increased frequencies. If you have time, read this thread
http://www.geotech1.com/forums/showt...eferrerid=2910

[pebe]

Have a look at this -> http://www.geotech1.com/pages/metdet.../US4486713.pdf

Thanks Qiaozhi. It was from Fig2 of that puplication that I got my original information. What I was hoping for was a bit more detail, like what are the phase shifts for gold, silver or other precious metals, and how do pull tabs and sundry rubbish fit into the picture.

[pebe]

Pebe, I made so much postings in the forum for this, that is impossible to show where are. Because you measure phase lag relative to TX voltage, yes you are right, but iron targets have more than 90 deg at increased frequencies. If you have time, read this thread
http://www.geotech1.com/forums/showt...eferrerid=2910

I have gone through all 9 pages but can find nothing there which helps me.

[mikebg]

I have gone through all 9 pages but can find nothing there which helps me.

Pebe, this image in posting #88 explains amplitides and phases of all targets and substances. Note that line 3 for iron is combination of lines 1 and 2.

[pebe]

Yes Mike, but that is not the information I want.

I am renovating an old IB MD and have completely redesigned the front end and the 'Disc' control. I would like to be able to calibrate it so that I know what type the target is for any phase delay that I measure - if that's at all possible.

I don't want to go into the technical whys and wherefores - I just want simple answers if they are to be found out there anywhere.

[Carl-NC]

One thing that Mike's plot show is that phase angle is highly dependent on frequency. So there is no fixed relationship between target type and phase angle.

The best approach is to test various targets and see what their phase responses are.

- Carl

[Qiaozhi]

Thanks Qiaozhi. It was from Fig2 of that puplication that I got my original information. What I was hoping for was a bit more detail, like what are the phase shifts for gold, silver or other precious metals, and how do pull tabs and sundry rubbish fit into the picture.

Unfortunately it's not that simple, as silver and gold can appear all over the place. If all targets were consistent in size, shape and orientation, then you might have a fighting chance. But the real world delivers gold in the shape of rings, and also pulltabs in the shape of rings, and places some Roman coins in the pulltab area. I would think the best you can hope for is to manually use some typical targets to find the correct location on the DISC control.

[mikebg]

Yes Mike, but that is not the information I want.

I am renovating an old IB MD and have completely redesigned the front end and the 'Disc' control. I would like to be able to calibrate it so that I know what type the target is for any phase delay that I measure - if that's at all possible.

Pebe, this is possible, however the phase that you measures depends mainly on phase shift in RX LC tank. This is the bad thing in every VLF metal detector. Use for refence target a ferrite core to measure phase shift in RX relative to current in TX coil. Then you should adjust GND balance so, that to eliminate ferrite core in one end of knob GND BAL. The other end of GND should eliminate salty water, but it is preferable to use DISC knob for this.
DISC knob should eliminate salty water or nails in one end and pull tabs in other end.
At what frequency operates your machine?

[pebe]

Mike,
The machine is a Micronta 4003 and uses a (measured) Tx frequency of 14.5kHz. I bought it cheaply on ebay not for metal detecting but for experimenting with the circuitry. My ultimate aim is to design a micro-based unit that requires virtually no setting up – purely as an excercise.

When I got it I realised the coils had not been nulled – and I did not want to saw the head apart to change the coil positions. Apart from the awful setting up procedure needed for the GND control, I realised that there was no way accurate way that the DISC control could be set in the combined presence of a badly nulled signal and a GND signal. That also applied to all the circuits I had seen. So I set about modifying the circuit so that nulling as well as setting the GND control, could be done electronically. It involved taking a sample of the Tx waveform and altering its phase, amplitude, and polarity. The existing GDN control now alters phase and the TUNE control alters amplitude and polarity. The resulting waveform will be aapplied to the Rx preamp so it would cancel the unwanted signals. Then I would have a DISC control that would accurately determine the phase of a received signal.

That means I cannot use the GND balance control in the way you suggest. It also has an untuned Rx coil.

So far, the circuitry has been designed, spiced, and made up on two small boards to be added to the main PCB. Hopefully, I should have test results after my holiday visitors have left in about 2 weeks. But this is just one of the many jobs I am doing and it has no urgent priority.

[mikebg]

Wideband or narrowband RX?

Pebe,
the AIR signal (your term is "badly nulled signal") can be minimized by two cores: a small ferrite core and an aluminium piece or foil placed in correct position over search head and fixed with scotch tape. You should also find the correct size of aluminium.
Afther well nulling, the gain of RFA (Radio Frequency Amplifier) remains limited mostly by GND signal (because it is much larger than input noise). You will compensate the GND signal in the input. If this is achieved, you can increase the gain of RFA untill output noise are near to threshold of saturation.
If you can achieve this, you are near to ideal case when the sensitivity is limited only by input noise.
However the ideal case is at minimal input noise, because then is allowed maximal possible gain of RFA.
The input of your RX is wide band because RX coil is not connected as tuned circuit. That means RFA will operate at more input noise and interferences than at ideal case when the bandwidth is no more than 16Hz.
I hope a day you to describe as block diagram your design solution.
The attached figure shows output of RFA at ideal operating wideband detector. The visual width of noise band seems as 1/8 to 1/6 of saturation limits Vp-p max because there are invisible noise peaks. The output of narrow band RFA seems different.

[pebe]

Mike,
The method you suggest for minimising the ‘AIR’ signal is pretty crude and I was looking for something better.

Attached is the block diagram of my modification. It has been made on two small veroboards that are to be fixed to the main PCB. The top circuit provides a sinewave with variable phase and amplitude that is injected into the existing preamp in order to cancel the received Rx signal in the absence of a target signal.

The bottom circuit gives three signals at the mode switch that decide which axis the phase sampling will work at. VLF is the normal ‘All metals’. Setup is for adjusting the phase and amplitude as above. ‘DISC’ gives discrimination over a 90º range.

The preamp is an audio amp – not an RF amp. Its output filter is a lowpass filter with a 3dB point at 72kHz. I may have to tighten that up to decrease noise – but that’s for later when I have got it running.

[mikebg]

I want to explain terminology that I used for Micronta 4003.
Although the term RFA is taken from my hobby - amateur radio, I do not believe it differs from other areas of electronics.
I think the attached here circuit diagram is:
1. A RFA (Radio Frequency Amplifier) because amplifies the frequency radiated by transmitter. The received TGT (target signal) has carrier frequency 14.5kHz and bandwidth BW<16Hz. The circuit is not audio frequecy amplifier despite the TX frequency 14.5 kHz is audible for young people.
2. A band pass amplifier because the RX coil forms tuned tank circuit operating as second order band pass filter as is shown in the equivalent network below. Despite LC tank is dumped by three resistors, which extends the passed band much more than 16Hz, the circuit remains band pass amplifier. Furthermore, the band is limited additionally in output stage from above by C3 and below by C4.

[pebe]

Mike,
Long, long, ago…. when I was taught radio principles, they were considered to start at about 30kHz and extend upwards to the visible spectrum. Frequencies below that were known as audio frequencies or supersonic frequencies. No audio amplifier could be (or still can be) considered ‘Hi-Fi’ unless its response extends to 20kHz, so 14.5kHz is definitely an audio frequency. The amplifier doesn’t care if it is receiving its signal from a transformer loop or from a microphone – it is seen as just an electrical signal.

But insofar as the term ‘radio frequencies’ now seems to refer to all electromagnetic radiations up to about 300GHz, then your use of the term is correct. However, when the term covers such a gamut of frequencies it serves no purpose to call the amplifier an RFA – why not just call it an amplifier?

It is common usage to define an amplifier by the range of frequencies it covers, ie. audio amplifier, video amplifier, HF amplifier, VHF amplifier, UHF amplifier, etc. Examples are everywhere, and it would be to everyone’s advantage if we all sang from the same hymn sheet.

2. I mentioned that Rx coil was untuned. Now when I look closer at the circuit I realise it is series tuned with C2 to about 12kHz – although with R4 damping across C2, tuning is probably quite flat. I haven’t been there yet!

My remarks about the circuit being a low pass filter were only intended to point out that with the roll-off introduced by R3/C3, I was not likely to get the noise peaks you had referred to. If I do then I will increase C3. But as I said, I have so far only built up the circuit – I have not tested it yet, and I certainly haven’t looked for troubles beyond Q1.