TGSL Experiments

[sevar12]

http://www.google.bg/imgres?um=1&hl=...9,r:1,s:0,i:65

can this be usable to set coil for TGSl

[Jerry]

Multi turn Ground Balance for TGSL

I have been reading for several years that a multi-turn GEB control is better than just a single turn ground balance. When I built my TGSL, I elected to install R-22, the 100K pot on the front panel instead of the circuit board. With the coils I have made, I have always been able to null out a piece of ferrite at about the center of rotation or about 50k between the wiper and ground.

When using the TGSL in the field, I start out by placing a piece of ferrite on the ground and turning the GEB until it just nulls out. The problem with this setup has been the front panel knob can (and has) get bumped and throw the ground balance way off. Also I did not see much off a difference in the actual setting of the control knob position when going from place to place. Not enough resolution with a single turn pot.

I changed out the single turn GEB for a combination ten turn front panel control in series with an internal trimmer which is a PCB style ten turn pot. I do not have a ten turn 100K pot so I used a 20K ten turn in series with a 100K ten turn trimmer pot. This probably gives more resolution for setting ground balance than is really needed but now I can really see the changes almost every time I null it too my piece of ferrite when I move to another location.

I calibrated this by setting the front panel at mid rotation which is five turns from either extreme and then adjusting the internal trimmer so there is 50K between U-102 pin 6 and ground. This gives +- 10K of adjustment with the front panel GEB control.

The more I use the TGSL the more I like it and when I am out swinging it I keep thinking of ideas to try.

I dug 142 coins today for a total of $4.82 US. Deepest was over six inches and it had a loud and clear signal.

Jerry

[simonbaker]

I have been reading for several years that a multi-turn GEB control is better than just a single turn ground balance. When I built my TGSL, I elected to install R-22, the 100K pot on the front panel instead of the circuit board. With the coils I have made, I have always been able to null out a piece of ferrite at about the center of rotation or about 50k between the wiper and ground.

When using the TGSL in the field, I start out by placing a piece of ferrite on the ground and turning the GEB until it just nulls out. The problem with this setup has been the front panel knob can (and has) get bumped and throw the ground balance way off. Also I did not see much off a difference in the actual setting of the control knob position when going from place to place. Not enough resolution with a single turn pot.

I changed out the single turn GEB for a combination ten turn front panel control in series with an internal trimmer which is a PCB style ten turn pot. I do not have a ten turn 100K pot so I used a 20K ten turn in series with a 100K ten turn trimmer pot. This probably gives more resolution for setting ground balance than is really needed but now I can really see the changes almost every time I null it too my piece of ferrite when I move to another location.

I calibrated this by setting the front panel at mid rotation which is five turns from either extreme and then adjusting the internal trimmer so there is 50K between U-102 pin 6 and ground. This gives +- 10K of adjustment with the front panel GEB control.

The more I use the TGSL the more I like it and when I am out swinging it I keep thinking of ideas to try.

I dug 142 coins today for a total of $4.82 US. Deepest was over six inches and it had a loud and clear signal.

Jerry

Hi Jerry - interesting. Your GB procedure seems unusual to me. It seems you are setting the TGSL to be insensitive to a piece of ferrite rather than to variations in the actual soil you are working with. That may in fact be very useful, but I always thought the idea was to set the GB control to the soil by pumping the search head up and down over the ground to create a "ground signal" which you then eliminate with the GB control.

On another note - I think it would be interesting to test the importance of the GB setting by trying settings that are "off" calibration and seeing how it affects your detecting; how does it affect false positives (beeping when no target is present); how does it affect false negatives (no beep when a target is present).

Regards,

-SB

[Jerry]

Hi Simon,

I started out ground balancing by pumping the coil up and down and turning the GEB for the least change in sound. I found that I lost just about all depth in doing it that way. I think I had too big of an offset on U-107 and it appears that balancing to the ground was over correcting the ground balance. This was why I added the threshold control that I wrote up a few weeks ago.

With the single turn GEB pot, it seemed like it was being set to exactly the same point each time. Well I am tuning to the same piece of ferrite but at the same time it is sitting on native soil so that should be in the matrix also. Changing to the ten turn pot for a GEB control does actually show changes in the ground balance when moving from place to place. Also I found out that there is big difference in balancing to the ferrite alone as in an air test and balancing to the ferrite placed on the ground.

I am going to revisit the pumping up and down to use the ground only but the method I described is getting me excellent results. Yesterday I dug a dime at six inches and the signal was such that I am certain I would have heard it even deeper.

I just reread your reply and regarding the false signals, the TGSL runs very stable and quiet when hunting. I have the threshold turned all the way up and the sensitivity backed off so the detector just quiets. I have also tried it with sens all the way up and thresh backed down but to not like the signals that way as much. Just a personal preference as I do not see a lot of difference in depth. But I do recommend the threshold control.

Jerry

[simonbaker]

Hi Simon,

I started out ground balancing by pumping the coil up and down and turning the GEB for the least change in sound. I found that I lost just about all depth in doing it that way. I think I had too big of an offset on U-107 and it appears that balancing to the ground was over correcting the ground balance. This was why I added the threshold control that I wrote up a few weeks ago.

With the single turn GEB pot, it seemed like it was being set to exactly the same point each time. Well I am tuning to the same piece of ferrite but at the same time it is sitting on native soil so that should be in the matrix also. Changing to the ten turn pot for a GEB control does actually show changes in the ground balance when moving from place to place. Also I found out that there is big difference in balancing to the ferrite alone as in an air test and balancing to the ferrite placed on the ground.

I am going to revisit the pumping up and down to use the ground only but the method I described is getting me excellent results. Yesterday I dug a dime at six inches and the signal was such that I am certain I would have heard it even deeper.

I just reread your reply and regarding the false signals, the TGSL runs very stable and quiet when hunting. I have the threshold turned all the way up and the sensitivity backed off so the detector just quiets. I have also tried it with sens all the way up and thresh backed down but to not like the signals that way as much. Just a personal preference as I do not see a lot of difference in depth. But I do recommend the threshold control.

Jerry

I think your results are interesting. I am wondering -- even though you find differences in your pot setting for different occasions, do those slight differences (slight because you have a 10 turn pot) make significant difference in your detecting?

I think ground-balancing to ferrite is very practical because it essentially immunizes your detector to amplitude changes in the oscillator or null signal. Maybe that is the most important adjustment for our TGSL, and any minor phase differences due to soil are not a big issue for this model MD. Something that deserves more testing and checking in my opinion.

Cheers,

-SB

[Jerry]

Hi Simon,

Well the original project had the GEB control located on the PC board so it was intended to preset the ground balance to a piece or ferrite and leave it there. From that I inferred that ferrite was a good compromise ground balance target but I wanted the ability to fine tune that.

The ten turn control does make a difference in performance for the deeper targets. I have one bothersome target in my coin test garden that must be located near some iron or something because that particular one is difficult to pick up with my other commercial detectors. I could not get it at all at first with the TGSL but after the GEB mod and fine tuning, I am able to hit it.

This is all preliminary since I have used it in the field one time for about three hours and I will form a better opinion as I get more experience using it in the real world.

Jerry

[SVEN1]

My Progress with the TGSL

Been awhile since I had some time to play with my TGSL-EDU detector build.........
The 2 tone PCB mod came in several months back and I assembled most of it, as always there was one part I didn't have on hand. Only place I could find it was in the UK, so I ordered several. Over the past few weeks worked on this machine on and off after finally getting settled from a move.
Pre-assembled the 9.5" DD search coil which is ready for tuning once I have the PCB with the mod up and running. The coil now set aside, concentrated on the control box.
Decided to use one of the empty Fisher CZ6 boxes, built a slide in cassette that hold the faceplate, PCB and search coil jack. Slides right into the control box. Two screws thru the faceplate and two screws from the back of the box into the search coil plug upright hold it in place. Headphone jack placed on the side as well as a small piezo buzzer for a speaker that emits target sounds thru a 5/16" hole. Power will be two 9 volt batteries for the time being. As always, just waiting for a part, a toggle switch. Decided to add a two stage ground balance control, one pot for coarse-initial tune and a second pot for fine tune-----100K and a 10K pot.

[satdaveuk]

Its looking smart well done'
If it was me and my first TGSL build , would have got it up and running before doing any mods because if you have a problem with it working its one less thing to have to consider when fault/finding.
Saying that, we all have are own ways of doing things.
All the best my friend

[SVEN1]

Its looking smart well done'
If it was me and my first TGSL build , would have got it up and running before doing any mods because if you have a problem with it working its one less thing to have to consider when fault/finding.
Saying that, we all have are own ways of doing things.
All the best my friend

Hi,
It already was debugged, up and running before the mod, just never got around to completing, making a control box and coil for it. Got side tracked building my PI's.
I had 4 Fisher CZ6 boxes I made up the same way at the same time to use on four builds if needed.
Keeping my fingers crossed it powers up and works after the mod.....LOL
Will let you guys know.........

[ashefr10]

Hi, good job, you will not be disappointed.

[simonbaker]

I think ground-balancing to ferrite is very practical because it essentially immunizes your detector to amplitude changes in the oscillator or null signal. Maybe that is the most important adjustment for our TGSL, and any minor phase differences due to soil are not a big issue for this model MD. Something that deserves more testing and checking in my opinion.

Cheers,

-SB

Rereading my last post, I think I made a false conclusion. I concluded that ground-balancing to ferrite would also "immunize" the ground channel to unwanted amplitude variations in the oscillator -- meaning the ground channel would not respond to oscillator amplitude changes.

It struck me that the above statement does not seem consistent with my theories that the null signal may be a composite of magnetic coupling and non-magnetic coupling (such as capacitive). Thinking about it some more, I suspect that any ferrite-induced signal would probably be in phase with magnetically coupled component of the null signal, and so ground-balancing ferrite would indeed prevent the ground channel from responding to that component.

However, if the null signal has significant non-magnetic coupling, it could have a phase quite far from the ferrite signal. In that case, it would seem that variations of the oscillator amplitude could possibly register significantly in the ground channel by modulating the null signal.

Just wanted to make that note for the record, since this thread is about discovering metal detector principles through experiments with the TGSL.

----------------------------------------

To sven1 -- nice work -- I like the battery pack.

-SB

[Davor]

Your observation is unfortunately correct in extreme conditions such as salt water beaches, but oscillator amplitude is not a direct culprit. The 2nd harmonic is. Also the capacitive coupling would not exist in case you use a balanced Tx, which is also free from 2nd harmonic.

So IMHO the next big thing in VLF is a balanced Tx.

[simonbaker]

Your observation is unfortunately correct in extreme conditions such as salt water beaches, but oscillator amplitude is not a direct culprit. The 2nd harmonic is. Also the capacitive coupling would not exist in case you use a balanced Tx, which is also free from 2nd harmonic.

So IMHO the next big thing in VLF is a balanced Tx.

Q - Why would amplitude modulation of the Tx oscillator, at say 10 Hz, not be a problem if the null signal has a phase that could register in both the DISC and GB channels?

I still think the "second harmonic" is mainly a problem if you have a low freq additive signal coming into the RX coil which could shift the null signal up and down, even if you have a full-wave SD. With a half-wave SD, the problem exists regardless of the harmonics as you pointed out previously.

I don't see "second harmonic" modulation of the oscillator by the ground as a big problem yet because I think the low frequency component is attenuated so much by the RX tank. However, I hope to do a full sim to see how much it might contribute. But I still believe that if the ground is modulating the second harmonic, it will modulate the fundamental even more, so eliminating the second harmonic modulation wouldn't help much in that case. But perhaps there is another effect you are referring to.

The capacitive coupling I'm talking about is a signal that couples from the TX leads to the RX leads because they lie next to each other in the cable -- or maybe it happens in the head or shield, I don't know. I did some experiments which showed very different null signal depending on the type of cable you use, and I believe the difference was due to some kind of non-magnetic coupling, because of the way it affected the phase of the null signal. I'm certainly no expert on it, but I think it is a real consideration in nulling our coils and interpreting the null signal of our coils.

Regards,

-SB

[Davor]

Q - Why would amplitude modulation of the Tx oscillator, at say 10 Hz, not be a problem if the null signal has a phase that could register in both the DISC and GB channels?

Actually it is not producing much response in GEB if rig is properly ground balanced. Would it influence target phase reading? In a way yes, but not by much. Coupling between ground and Tx is not that big to produce much of a modulation depth. So on the fundamental frequency, even with some frequency/phase modulation the effect is quite mild. True, amplitude stability would not hurt anything.

I don't see "second harmonic" modulation of the oscillator by the ground as a big problem yet because I think the low frequency component is attenuated so much by the RX tank.

Trouble with 2nd harmonic is that it spoils balance and enables Tx noise enter Rx by back door through offset shifting the phase shifters, and in a flat rate, unrelated to the targets in any way, and in effect determines your noise floor. The most interesting feature of the 2nd harmonic is that it is much more sensitive to just about anything that happens with Tx oscillator, and you may freely attribute most of the falses to the 2nd harmonic being playful in your rig.

The capacitive coupling I'm talking about is a signal that couples from the TX leads to the RX leads because they lie next to each other in the cable -- or maybe it happens in the head or shield, I don't know.

Exactly! The only way of cancelling these is balanced operation. If for example you have a single ended oscillator providing Vtx with one end of the coil grounded, the average voltage of the coil, cable, etc. is Vtx/2. This Ttx/2 will couple capacitively, and in a case of weak coupling at -60dB, you'd still have a residual at ~10mV.
In case of balanced operation, you have Vtx in counterphase, and the average is 0V. This is much better at any coupling.

[starmoml]

Hello. show me breakdown on the circuit schematic. PLEASE help me