TGSL Tuning & troubleshooting

[simonbaker]

Hi,
I have made my TGSL according to the recipe "TGSL - complete details".

Some hundreds of messages ago I had problmes with oscillation in the receiver stage. That calmed down when I replaced one LM358 with one from another manufacturer.

The current state is, that the detecting distance of a 2 eur coin is 1cm, yes 1cm in the air. I can null the coils to 10mV, but not a bit lower. It gives a short beep when I sweep the coin just past the overlapping section of the coils, exactly as in the videos in the Youtube. I feel, that it's just about to work, but I don't know where to get the missing 29cm of detection distance.

As cable between the coils and the device, I'm using a USB cable, connectors cut off. It has two twisted pairs inside a common shielding. Does twisting make any harm or should all wires bew straight routed inside the shielding? I have made sure, that one twisted pair goes to the TX coil and another to the RX coil. So, the TX and RX signals go in separate twists.

I can compare the oscilloscope views to those in the TGSL thread's, I just don't know to which of the sets in the TGSL threads are "correct ones". Hints?

I'm not far enough to ask this, but want to clarify it to prevent later banging my head into the wall: when talking CW or CCW positions of the potentiometers, does CW mean the clockwise rotation of the arrowed wiper of the schematic symbol, see CW_CCW.JPG?

All hints appreciated,
Andy

This link has attached PDF file with a good set of scope pics to start with. The scope pictures on the left were a home-built with DD coil, the "graphs" on right are same measurement with original TGS and concentric coil.

http://www.geotech1.com/forums/showp...postcount=3896

So many things to check. Plus and Minus voltage rails are first, especially minus, which should be around -4 to -6 volts. Plus rail about 8 volts.

Then oscillator freq and amplitude. Should be about 14.5 Khz, and 10 to 16 volts peak-to-peak.

Make sure coil shielding has a gap (not continuous around coil). Make sure the Receive (RX) coil and Transmit (TX) coils shields do not electrically contact and create a continous loop.

Check RX coil leads are correct phase -- you can try reversing them to see how it changes the results (you may have to null again).

Then I would work my way from "front" (coils) to "back" (speaker) checking the various signals using oscilloscope, comparing to reference pictures.

There are two "channels", the discrim channel, and the "ground balance" channel. If one is not working right, can cause problem.

It is easy to put JFets in wrong way, double-check that.

Then check the signals at the gates of the JFets of each channel's detector with switch in discrimination mode (not all-metal). The signals should look like almost square waves that have a max at about zero volts and a min at about V-. They should have different phases, which varies with the discrimination knob.

Turn the discrimination knob and the phase of the discrim gate signal should shift relative to the TX signal. It can shift maybe 90 deg.

You can check the Ground Balance gate signal similarly. Turn the GB knob, and the gate signal phase should shift relative to the TX signal. It shifts typically 45 to 50 deg, then may disappear at one end.

If the peak of either gate signal drops below zero suddenly as you turn the knob, then you may have your coil null point set on the wrong side. This causes the filtered voltage on the Source side of the JFet to be negative instead of positive, which causes the JFet to be forward biased at times, dragging down the gate voltage -- and probably causing the detector to function poorly. Voltage at JFet Source should be positive, maybe up to .5 volts. To fix, shift the coils slightly to other side of null point. But this assumes your circuit is working perfectly.

Then what I do is set my oscilloscope to a very slow sweep, maybe .2 cm/second. I put a probe on each output of the LM308s and set the vertical sensitivity to maybe .1 volts per cm or whatever is necessary. Then sweep the coin over your coil. You should see pulses in both channels. If the both pulses go postive, you should hear a tone in the speaker. If the pulses go in opposite directions, you are discriminating out the coin. If both pulses go negative, some phase is backwards.

Try turning the discrim knob to make the pulses go in the same direction. If they can't, then it sounds like your phases are wrong in the front end. You could have wrong capacitors in the discrim or GB circuits, or coil leads reversed, etc. Check carefully the signal phases.

If the pulses are just tiny when you sweep the coin in both channels, then maybe you have problem with ICs or wrong resistors, etc.

It takes time to understand the circuit to find problems, but that is part of the fun (maybe).

It is a lot easier when you have one project finished and working well, then you can compare. But first one is always hard.

Let us know what you find.

Regards,

-SB

[Mullihaka]

Thanks for help! Now I have a load of information to go ahead.

I still havent figured out the operating principle of the detector, so I would need to know how CW and CCW is interpreted with different pots.

There are three potentiometers. When saying CW, is it towards:
- pin 1 of U101b or R19 for disc pot
- pin 6 of U102b or ground for ground balance pot
- plus rail or ground for sensitivity pot

Andy

[simonbaker]

Thanks for help! Now I have a load of information to go ahead.

I still havent figured out the operating principle of the detector, so I would need to know how CW and CCW is interpreted with different pots.

There are three potentiometers. When saying CW, is it towards:
- pin 1 of U101b or R19 for disc pot
- pin 6 of U102b or ground for ground balance pot
- plus rail or ground for sensitivity pot

Andy

It depends on how you wire it and how you want it. Just test it with your ohm meter and make note where you wired it into the circuit.

People shouldn't explain procedures to you with CW or CCW because people may connect their pots differently.

For guidance, I personally would want this on my detector:

* Sensitivity - CW makes more sensitive. So CW would make wiper voltage go more to ground.

* Disc pot - CW would reject more metals. Good question, I'm not sure -- I think maybe CW makes wiper go to LF353 pin 1. I would try it, if not the way you want it, reverse the leads.

* GB pot - I don't think I care which way the pot turns, as long as I can find a point for ground balancing. Again, reverse the leads if it feels wrong to you.

Not a great answer, sorry. Hopefully others will give you better advice on what is standard. Molzar has the original TGS so he would be a good reference.

-SB

[dfbowers]

Check your soldering job, a bridge or bad joint! Double check all component values! I am always my own worst enemy!

Don

[Mullihaka]

Coil nulling: are methods mutually exclusive?

Getting more distance: now the device can detect 2 euro coin at 10 cm.

I have checked everything Simonbaker recommended. The graphs look much the same as those taken by Max. Frequency is 15.0 kHz.

There are several instructions about nulling the coils.
One of them asks the voltage at the source of TR5 to be adjusted as close to zero as possible (method 1).
Second one asks the phase difference between TX and RX coils to be adjusted to 200 degrees (method 2).
Third one asks the voltage at RX coild to be adjusted as low as possible (method 3).

I have tried all nulling methods with variable results. Common for all of them is, taht when I swipe the coin cross the coils, it gives two beeps: one when crossing the coil's outer edge, second one when the coin crosses the overlapping area.

In pictures below, trace 1 is TX coil (collector of TR1), trace 2 is pin 7 of U101a and trace 3 is source of TR5. In first picture the nulling has been done according to the method 1, which resulted +25 mV at the source of TR5. The phase difference is about 29 degrees (8% * 360 deg).
In the second picture, the coils vwere adjusted to get minimum voltage at pin 7 of U101a. In this case the voltage at the source of TR5 is -25 mV and pahse difference about -120 degrees (-33% * 360 deg).

I have the coils resting one on other, without any fixing yet. They don't have the shielding yet. The coils are resting on a styrofoam board, about one meter above the floor; so there's no metal above or below them. At the side of the setup (30 cm away) there are large metal cases (my instrumentation). Can they disturb the nulling? Is it even possible to null the coils without fixing the main parts of them ie am I trying to do something impossible, if I have them just resting free one on other?

It has been mentioned also, that overnulling can cause troubles. How to prevent overnulling ie what to look for?

Is it so, that if I null the coils according to the method 1, it will automatically fulfill the requirements in method 2 i.e. the phase difference is automatically correct?

Do these pictures make sense?

Andy

[Qiaozhi]

Getting more distance: now the device can detect 2 euro coin at 10 cm.

I have checked everything Simonbaker recommended. The graphs look much the same as those taken by Max. Frequency is 15.0 kHz.

There are several instructions about nulling the coils.
One of them asks the voltage at the source of TR5 to be adjusted as close to zero as possible (method 1).
Second one asks the phase difference between TX and RX coils to be adjusted to 200 degrees (method 2).
Third one asks the voltage at RX coild to be adjusted as low as possible (method 3).

I have tried all nulling methods with variable results. Common for all of them is, taht when I swipe the coin cross the coils, it gives two beeps: one when crossing the coil's outer edge, second one when the coin crosses the overlapping area.

In pictures below, trace 1 is TX coil (collector of TR1), trace 2 is pin 7 of U101a and trace 3 is source of TR5. In first picture the nulling has been done according to the method 1, which resulted +25 mV at the source of TR5. The phase difference is about 29 degrees (8% * 360 deg).
In the second picture, the coils vwere adjusted to get minimum voltage at pin 7 of U101a. In this case the voltage at the source of TR5 is -25 mV and pahse difference about -120 degrees (-33% * 360 deg).

I have the coils resting one on other, without any fixing yet. They don't have the shielding yet. The coils are resting on a styrofoam board, about one meter above the floor; so there's no metal above or below them. At the side of the setup (30 cm away) there are large metal cases (my instrumentation). Can they disturb the nulling? Is it even possible to null the coils without fixing the main parts of them ie am I trying to do something impossible, if I have them just resting free one on other?

It has been mentioned also, that overnulling can cause troubles. How to prevent overnulling ie what to look for?

Is it so, that if I null the coils according to the method 1, it will automatically fulfill the requirements in method 2 i.e. the phase difference is automatically correct?

Do these pictures make sense?

Andy

Both methods 1 and 3 are incorrect for setting up the coil balance.
Also, method 2 should have 20 degrees for the offset, when using a DD coil, not 200 degrees.
If you are getting 2 beeps when you pass a coin over the coils at a distance of about 7cm, then you have the RX coil connections swapped over.

[Mullihaka]

Thanks Qiaozhi!

So, for my DD coil the phase shift in the first picture is in right direction but a bit too much (29 deg instead of 20)?

Andy

[simonbaker]

Getting more distance: now the device can detect 2 euro coin at 10 cm.

I have checked everything Simonbaker recommended. The graphs look much the same as those taken by Max. Frequency is 15.0 kHz.

There are several instructions about nulling the coils.
One of them asks the voltage at the source of TR5 to be adjusted as close to zero as possible (method 1).
Second one asks the phase difference between TX and RX coils to be adjusted to 200 degrees (method 2).
Third one asks the voltage at RX coild to be adjusted as low as possible (method 3).

I have tried all nulling methods with variable results. Common for all of them is, taht when I swipe the coin cross the coils, it gives two beeps: one when crossing the coil's outer edge, second one when the coin crosses the overlapping area.

In pictures below, trace 1 is TX coil (collector of TR1), trace 2 is pin 7 of U101a and trace 3 is source of TR5. In first picture the nulling has been done according to the method 1, which resulted +25 mV at the source of TR5. The phase difference is about 29 degrees (8% * 360 deg).
In the second picture, the coils vwere adjusted to get minimum voltage at pin 7 of U101a. In this case the voltage at the source of TR5 is -25 mV and pahse difference about -120 degrees (-33% * 360 deg).

I have the coils resting one on other, without any fixing yet. They don't have the shielding yet. The coils are resting on a styrofoam board, about one meter above the floor; so there's no metal above or below them. At the side of the setup (30 cm away) there are large metal cases (my instrumentation). Can they disturb the nulling? Is it even possible to null the coils without fixing the main parts of them ie am I trying to do something impossible, if I have them just resting free one on other?

It has been mentioned also, that overnulling can cause troubles. How to prevent overnulling ie what to look for?

Is it so, that if I null the coils according to the method 1, it will automatically fulfill the requirements in method 2 i.e. the phase difference is automatically correct?

Do these pictures make sense?

Andy

D004.TIF looks most like the result that works best for me. But only if the RX leads are correctly connected.

Qiaozhi has good idea to try swapping the RX leads and try nulling again.

I find that if the coin is very close to the coil, you may get double beeps in any case. But if further away and you get double beeps, it is a clue that the leads should be swapped.

I also feel that after nulling, the voltage at the Source of the JFets should be slightly positive (between 0 and .5 volts) -- your null phase determines whether positive or negative.

You will most likely need to start building more coils, making slight variations in size, inductance, etc. For some reason, that seems to be how people succeed in achieving the best MD. Some coil just works much better than others.

It is worth trying to finish some coils by shielding and gluing in place. Measure the maximum air depth before and after finishing. My last coil, after gluing, was worse than before gluing. I used Elmer's glue (water based) -- maybe bad idea.

Ivconic and Max are masters of making good TGS units. Copying them is probably best advice. My homebuilts so far have not achieved their depth, so I cannot give best advice. However, I have done a fair amount of experimenting.

I wonder also about nearby metal and even EMI from house wiring, computers, etc. If you can, take everything outside and see if different.

It is difficult to get your first MD to work well, takes luck. Better to make a few PCBs and coils so you can compare and have more confidence it is not bad part or bad coil.

-SB

[Qiaozhi]

Thanks Qiaozhi!

So, for my DD coil the phase shift in the first picture is in right direction but a bit too much (29 deg instead of 20)?

Andy

The first picture looks correct. However, bear in mind that the RX signal will be inverted if you swap over the connections. In which case you will need to readjust them to achieve the same phase relationship. The phase offset does not have to be exactly 20 degrees. The important point is that you must be able to achieve ground balance.

Both Max and Ivconic claim to have achieved greater than expected detection distances for this design. In general you should expect the detection depth for a 1 Euro coin to about the same as the coil diameter. Some detectors can achieve better than this guideline, but this is only in an air test. It's a different story when you get out into the field. Look at this link for a realistic guide to detection depth ->
http://wwwdark.w.interia.pl/testy.htm#tesorss

[Dennis the Mennis]

Hi,
I have made my TGSL according to the recipe "TGSL - complete details".

Some hundreds of messages ago I had problmes with oscillation in the receiver stage. That calmed down when I replaced one LM358 with one from another manufacturer.

The current state is, that the detecting distance of a 2 eur coin is 1cm, yes 1cm in the air. I can null the coils to 10mV, but not a bit lower. It gives a short beep when I sweep the coin just past the overlapping section of the coils, exactly as in the videos in the Youtube. I feel, that it's just about to work, but I don't know where to get the missing 29cm of detection distance.

As cable between the coils and the device, I'm using a USB cable, connectors cut off. It has two twisted pairs inside a common shielding. Does twisting make any harm or should all wires bew straight routed inside the shielding? I have made sure, that one twisted pair goes to the TX coil and another to the RX coil. So, the TX and RX signals go in separate twists.

I can compare the oscilloscope views to those in the TGSL thread's, I just don't know to which of the sets in the TGSL threads are "correct ones". Hints?

I'm not far enough to ask this, but want to clarify it to prevent later banging my head into the wall: when talking CW or CCW positions of the potentiometers, does CW mean the clockwise rotation of the arrowed wiper of the schematic symbol, see CW_CCW.JPG?

All hints appreciated,
Andy

Hi Andy ,

There are some more 'builders' like Stephano and me still working on the TGS(L) with simular problems , poor detection depth and wondering what could the problem.
Have you thougt about the 'off resonance' frequency of the Rx coil ?
Perhaps you can gain some centimeters overthere ?
I completely missed that part even I thought I had read all the posts.
On the other hand I had made also a set of coils way out of range as what should be considered ideal with a detection depth of 10 cm on a 1 euro coin.
The last set of coils I've made seems to be 'by the book' but I'm still strugling with some problems ...
Time at the moment time is my greatest problem , but I'm still reading these posts hoping someone has the solution ...

regards

Dennis the Mennis

[Dennis the Mennis]

But on the other hand if you could manage the aprox. 20 -30 degrees phase shift you should be close to the desired 'off-resonance frequency'...
Did you had to tune a lot in the Tx- and Rx parts with the capacitors?

[Mullihaka]

Coil nullin, conditions to met?

Hi Dennis TM,

The coil recipe was as follows: 255mm diameter, 137mm height at middle, TX 100 turns (5.7 mH), RX 105 turns (6.2 mH), wire diameter 0.25mm.

I used 0.3mm wire, maybe that left the impedances bit lower: TX 5.3 mH and RX 5.8 mH. The TX frequency is 15.1 kHz. I then added the recommended 9% to that and got 16.5 kHz for RX coil resonance. With the existing RX coil inductance, the capacitor C6 should be 16.2 nF, therefore I added 1 nF in parallel to the 15 nF. I don't have the shieldings (yet). I need to get the device to work on bench before going further.

Simonbaker and Qiaozhi have given valuable hints, but I'm afraid I still haven't digested everything. I'm still not sure how to null the coils. Based on everything I have read here, there are several criteria, which all should be met simultaneously:

1) First minimize the level at pin 7 of LF353. Then push the coils a bit more together so that the RX signal lags the TX by 20 degrees.
2) After completing (2), the voltage at the source of TR5 should be close to zero. According to Simonbaker, 0 - 0.5V.
3) Also the sampling on the GEB channels should happen at the zero crossing point of the RX signal. See attached JPG. I understand this so, that the lower red "X" (GEB sampling start) should be exactly at the zero crossing of the RX singal (upper red "X").
4) Ferrous targets should decrease the initial delay (<20 deg) and non-ferrous increase (>20 deg).
5) Ferrous targets should increase the RX level, non-ferrous decrease.
I still haven't managed to get all these happen simultaneously. I can get one or two at any time.

It would be good to see the nulling in practise (Youtube?) to know, if my procedures are anything like those resulting a good detector.

Andy

[Qiaozhi]

Hi Dennis TM,

The coil recipe was as follows: 255mm diameter, 137mm height at middle, TX 100 turns (5.7 mH), RX 105 turns (6.2 mH), wire diameter 0.25mm.

I used 0.3mm wire, maybe that left the impedances bit lower: TX 5.3 mH and RX 5.8 mH. The TX frequency is 15.1 kHz. I then added the recommended 9% to that and got 16.5 kHz for RX coil resonance. With the existing RX coil inductance, the capacitor C6 should be 16.2 nF, therefore I added 1 nF in parallel to the 15 nF. I don't have the shieldings (yet). I need to get the device to work on bench before going further.

Simonbaker and Qiaozhi have given valuable hints, but I'm afraid I still haven't digested everything. I'm still not sure how to null the coils. Based on everything I have read here, there are several criteria, which all should be met simultaneously:

1) First minimize the level at pin 7 of LF353. Then push the coils a bit more together so that the RX signal lags the TX by 20 degrees.
2) After completing (2), the voltage at the source of TR5 should be close to zero. According to Simonbaker, 0 - 0.5V.
3) Also the sampling on the GEB channels should happen at the zero crossing point of the RX signal. See attached JPG. I understand this so, that the lower red "X" (GEB sampling start) should be exactly at the zero crossing of the RX singal (upper red "X").
4) Ferrous targets should decrease the initial delay (<20 deg) and non-ferrous increase (>20 deg).
5) Ferrous targets should increase the RX level, non-ferrous decrease.
I still haven't managed to get all these happen simultaneously. I can get one or two at any time.

It would be good to see the nulling in practise (Youtube?) to know, if my procedures are anything like those resulting a good detector.

Andy

I think I have spotted your mistake.
The sampling pulse should not start at the zero-crossing of the RX signal. It is the middle of the sample pulse that must be at the crossing point. Also, for a DD coil, the sampling must take place at the negative-going transition, not the positive.

And one last point ... all targets should cause a phase-shift to the right.

[Stefano]

Time at the moment time is my greatest problem , but I'm still reading these posts hoping someone has the solution ...

That's exactly my same problem!!! .....

Thanks Qiaozhi for your hints: always very appreciated

[simonbaker]

I think I have spotted your mistake.
The sampling pulse should not start at the zero-crossing of the RX signal. It is the middle of the sample pulse that must be at the crossing point. Also, for a DD coil, the sampling must take place at the negative-going transition, not the positive.

And one last point ... all targets should cause a phase-shift to the right.

This is a tricky issue which Qiaozhi and I occasionally have slightly different interpretations of. And I don't have concrete proof for my ideas. But...

My opinion is that the null signal is very deceiving to use for phasing decisions because it does not really indicate the phase of the actual received signals. The actual received signals are very tiny and not visible and have their own phase which is not directly related to the phase of the null signal. I believe you can vary the phase of the null signal by shifting the coils here and there without much changing the phase of the received signals. But don't believe me yet -- it needs to be demonstrated.

However, based on my opinion, it means the "sample pulse" needs to be properly phased relative to target signals, in other words, don't worry so much about the null signal for that issue. Of course the sample pulse phase moves around when you rotate the discrim pot, so we always need to mention the location of the discrim pot when we show the "sample pulse", otherwise we are comparing different circuits.

As it turns out, there is no adjustment in the TGS circuit to "bias" or shift the basic phase range of the discrim "sample pulse" -- just the discrim knob itself. So for now, we just build the circuit and hope for the best. I'm actually trying to experiment with some ways to adjust the range over which the discrim control works to help troubleshoot if that may be the cause of poor performance.

Anyway, what this all means is that I think the phase of the null is mainly important for biasing that voltage at the Source side of the JFet. However, I realize that must be an oversimplification because of all the importance placed on nulling. But from my tests so far, that's how I see it.

I know I need to learn more and understand better, so if I completely change my opinion later, don't be surprised. But the main reason I make this point is that I think that with some coils you can null for the rest of your life and they'll never work any better. It's something else about the coil that is deficient and it is a mystery right now. So I would follow one of Ivconic's recommendations which is: null your coil by swinging a coin above it for maximum depth. If you don't get good depth, build another coil!

I hope to know more some day. I'm in that group that has not made a TGS with more than 15 to 20 cm depth.

-SB