TGSL Experiments

[simonbaker]

Actually Simon since posting ive rechecked the amplitude on my TGSL and on the TX im getting just short of 20vp-p with a near perfect sinewave apart from a very small blip to the right, which is proberly nothing to worry about so by pushing that any higher do you think there would be any benefit, im proberly treading on ground you have already covered.
I think the extra 5 volts was gained when I did ivonics preset mod in the TX stage but had forgotten and had the old 14vp-p in my head, must be old age creeping on .

Hi satdaveuk:

I don't know how feasible it is to increase the TX voltage/current, but I think it is one of the things we should try and find out!

Because that test oscillator has such a squirrelly waveform, I'm not sure it is a good choice to experiment with -- but maybe the current waveform in the coil is nicer, that's what matters I think.

One problem we might encounter with higher voltage is our null signal will get larger and larger and at some point, depending on phase also, might cause some problems. But these things can be looked at.

Also, to get higher voltage/current from a 9 volt battery, generally you need a "high-Q" coil driven in a certain way that allows higher voltages/currents (which the test oscillator might in fact do). But the problem is that a "high-Q" TX circuit might be too sensitive to soil fluctuations, etc. But no harm in trying! I'm interested in high voltage/current experiments to see if any benefits are out there.

Cheers,

-SB

[Dennis the Mennis]

Here it is, almost the New Year, and I'm back tinkering with a basic question -- RX coil grounding, nulling, and null phase.
And already I'm somewhat flummoxed!

Let's just talk about RX coil grounding and nulling, for the TGSL coils; ignore phase for now.

We have two basic grounding configurations:

1. Ground the inverting RX lead in the coil head.

2. Don't ground the RX leads at all (see dfbowers wet-grass configuration).

Before looking at "null phase" vs. "null amplitude" in detail (a popular obsession), I thought I'd see basically how the null signal changes when you compare the two grounding configurations. Because if grounding affects the null phase, then we probably need to study both cases individually (and hopefully come up with some common theory to give us restful nights).

I couldn't remember past experiments of how the RX grounding affected the null, so I tried it again.

To start with, I'm using some unshielded coils dfbowers sent me, and I'm using a dfbowers PCB also (to keep my constructions out of the equation).

My initial tests show that grounding or not grounding the RX lead get minimum null at slightly different overlaps. A pretty small difference, but it's there. Grounded configuration requires slightly more overlap to achieve minimum null.

So why? Can someone else confirm this (remember I'm testing unshielded coils)?

I expected phase changes, but shouldn't the coils still null at the same overlap? Especially since they are not shielded.

-SB

Hi Simon ,

I've just checked and can confirm that I've had the same results :


http://youtu.be/Q0gSS0ds6zQ?hd=1


In the attached link there is a small clip :

It starts with the unshielded coils where the ground is connected to the Rx coil as you mentioned.

When the Rx voltage gets higher I disconnect the GND and need less overlap to achive the minimum null .

So this all together is exact the same as described in your post.

In this case I've used USBII cable instead of Belden 8723 , the cable GND was still connected the GND of the pcb during the test ....


kind regards ,

Danny

[dfbowers]

Hi satdaveuk:

I don't know how feasible it is to increase the TX voltage/current, but I think it is one of the things we should try and find out!

Because that test oscillator has such a squirrelly waveform, I'm not sure it is a good choice to experiment with -- but maybe the current waveform in the coil is nicer, that's what matters I think.

One problem we might encounter with higher voltage is our null signal will get larger and larger and at some point, depending on phase also, might cause some problems. But these things can be looked at.

Also, to get higher voltage/current from a 9 volt battery, generally you need a "high-Q" coil driven in a certain way that allows higher voltages/currents (which the test oscillator might in fact do). But the problem is that a "high-Q" TX circuit might be too sensitive to soil fluctuations, etc. But no harm in trying! I'm interested in high voltage/current experiments to see if any benefits are out there.

Cheers,

-SB

Simon,

I have always had the same curiosity that you have when it comes to higher power TX coils. Fortunately, the Nautilus DMCiib that I have allows me to play endlessly with TX power in real conditions. The coil voltage is adjustable from 5 to 44 volts, but not sure how that equates to "ampere-turns". The fact that it is "adjustable" is probably more meaningful in experimenting than knowing what the actual power is. So, if I had choice in any detector feature, I would go with adjustable power.

In reality, the detection range is not much farther than the TGSL , coils sizes being equal. With the ground in the equation, detection depth is determined mostly by ground conditions and not TX power. Cranking up TX power on the Nautilus increases detection depth - up to a point. Then, it just kills batteries quicker with little or no gain in detection depth. Taking the Nautilus to sandy soil improves things dramatically - but that is the case with the TGSL and sandy soil too!!! So, good things can be said about low power TX. Still, I'm not satisfied either.. more power!!

Now here is a question that someone may be able to answer for me:
If the definition of TX power can be expressed in AMPERE-TURNS, then how does this relate to coils with a high flyback voltage? The Nautilus uses what looks like a Harley oscillator with a center-tapped coil, plus two other taps that allow a higher flyback voltage across the whole coil. Or.. is the flyback voltage just a byproduct of high coil current being shut off and the collapsing magnetic field across a higher resistance.

Here is a good example- The TX section of the Tarsos. It uses an autotransformer arrangement that produces a huge flyback .. over 100 Volts! And look at the RX section- no preamp gain before the demodulator.

Don

[simonbaker]

Hi Simon ,

I've just checked and can confirm that I've had the same results :


http://youtu.be/Q0gSS0ds6zQ?hd=1


In the attached link there is a small clip :

It starts with the unshielded coils where the ground is connected to the Rx coil as you mentioned.

When the Rx voltage gets higher I disconnect the GND and need less overlap to achive the minimum null .

So this all together is exact the same as described in your post.

In this case I've used USBII cable instead of Belden 8723 , the cable GND was still connected the GND of the pcb during the test ....


kind regards ,

Danny

Thanks for confirming. (Such a nice scope you have...)

There are a few explanations I can think of but I'll be doing more tests (I hope... holidays over...)

On repeating the test I noted that it is a very small overlap change using unshielded coils, hard to even measure. So I'm not too obsessed with it yet, however, I would like to be able to explain it with a final model.

Hope you had good holidays there!

-SB

[simonbaker]

Simon,

I have always had the same curiosity that you have when it comes to higher power TX coils. Fortunately, the Nautilus DMCiib that I have allows me to play endlessly with TX power in real conditions. The coil voltage is adjustable from 5 to 44 volts, but not sure how that equates to "ampere-turns". The fact that it is "adjustable" is probably more meaningful in experimenting than knowing what the actual power is. So, if I had choice in any detector feature, I would go with adjustable power.

In reality, the detection range is not much farther than the TGSL , coils sizes being equal. With the ground in the equation, detection depth is determined mostly by ground conditions and not TX power. Cranking up TX power on the Nautilus increases detection depth - up to a point. Then, it just kills batteries quicker with little or no gain in detection depth. Taking the Nautilus to sandy soil improves things dramatically - but that is the case with the TGSL and sandy soil too!!! So, good things can be said about low power TX. Still, I'm not satisfied either.. more power!!

Now here is a question that someone may be able to answer for me:
If the definition of TX power can be expressed in AMPERE-TURNS, then how does this relate to coils with a high flyback voltage? The Nautilus uses what looks like a Harley oscillator with a center-tapped coil, plus two other taps that allow a higher flyback voltage across the whole coil. Or.. is the flyback voltage just a byproduct of high coil current being shut off and the collapsing magnetic field across a higher resistance.

Here is a good example- The TX section of the Tarsos. It uses an autotransformer arrangement that produces a huge flyback .. over 100 Volts! And look at the RX section- no preamp gain before the demodulator.

Don

Yes, interesting questions and observations. Do you have the Nautilus schematic to look at? I'm wondering what they do when they increase TX voltage; do they dial back the RX gain at the same time? This would essentially give you a better signal-to-noise ratio, which wouldn't be too noticeable in quiet areas. Should help near power lines though I would think. Or can you adjust the gain manually until optimum? Maybe it is the stinking 6th order depth law (or whatever) that defeats the extra power.

I guess VLF and PI coils have to be analyzed differently. With VLF, ampere-turns is our main way to adjust the magnetic field. The rate of change is fixed by our oscillator frequency and sine wave shape. I assume the autotransformer design gets more current-turns, otherwise that high voltage really doesn't really buy you anything, right? I find center-tapped coil designs a bit mysterious at this point in general as to how to exploit them.

With PI coils, my understanding is that how fast you shut off the current is an important factor for goosing the target signal. So it seems to be both the current-turns and the speed of shut-off that gives the flux rate-of-change. Huge flyback voltage is desired since it is symptomatic of a fast shut-off, as far as I know.

It does seem there should be some advantage to high voltage/current coils, at least for fighting against EMI noise.

-SB

[dfbowers]

Yes, interesting questions and observations. Do you have the Nautilus schematic to look at? I'm wondering what they do when they increase TX voltage; do they dial back the RX gain at the same time? This would essentially give you a better signal-to-noise ratio, which wouldn't be too noticeable in quiet areas. Should help near power lines though I would think. Or can you adjust the gain manually until optimum? Maybe it is the stinking 6th order depth law (or whatever) that defeats the extra power.

I guess VLF and PI coils have to be analyzed differently. With VLF, ampere-turns is our main way to adjust the magnetic field. The rate of change is fixed by our oscillator frequency and sine wave shape. I assume the autotransformer design gets more current-turns, otherwise that high voltage really doesn't really buy you anything, right? I find center-tapped coil designs a bit mysterious at this point in general as to how to exploit them.

With PI coils, my understanding is that how fast you shut off the current is an important factor for goosing the target signal. So it seems to be both the current-turns and the speed of shut-off that gives the flux rate-of-change. Huge flyback voltage is desired since it is symptomatic of a fast shut-off, as far as I know.

It does seem there should be some advantage to high voltage/current coils, at least for fighting against EMI noise.

-SB

I don't have the complete schematic for the DMCiib but here is one that I believe uses the same coil and TX arrangement. Notice, along with adjustable TX power they provide external controls for nulling as well.

Circuit looks like a single channel designed to be used in salt water.

Don

[simonbaker]

I don't have the complete schematic for the DMCiib but here is one that I believe uses the same coil and TX arrangement. Notice, along with adjustable TX power they provide external controls for nulling as well.

Circuit looks like a single channel designed to be used in salt water.

Don

Thanks for schematic. Definitely interesting center-tapped coil usage among other things.

-SB

[simonbaker]

nulling to death

For those of us with noisy workshops, we can't do depth tests, but we can do nulling phase tests! Yea!

Why? I don't know either. But people seem interested in this subject as it might impart magical powers to their MD if they get the null phase just right...

Anyway, here are some results with unshielded coils. I slowly changed the overlap and measured the null voltage (at LF353 pin 7) vs. the null phase (relative to the TX signal).

I did this with the RX lead grounded and then with the RX lead ungrounded (true differential input, dfbowers wet grass configuration).

I also wanted to see how this compared to a simple theoretical model. This model is:

A sin(x) + B sin (x + 90), where A represents the pure coupling of the two coils and changes from a large number to zero to a large negative number as the coil overlap moves through the null point, and B represents a small capacitive coupling of the TX and RX leads that is 90 degrees out of phase with the TX signal.

I changed A from 8 volts to -8 volts, and let B = .3 volts, which is approximately the null signal voltage minimum observed.

The attached photo shows the results.

I plan to repeat these tests with other coils which I will then put different shields on to see the effect.

My hunch is that the radical shift of the null phase near minimum can be explained by the presence of another signal approximately 90 deg relative that is added in, and is not "magnetically" coupled and therefore not affected by ground balancing, and that the exact phase of the null signal is probably not critical. But I could be wrong.

It looks like the simple theoretical model may be improved by adding a small magnetic phase bias to the magnetic component, so instead of shifting from -90 to 90 it goes more like -85 to 85.

Anyway, more tests to come...probably...

-SB

[simonbaker]

For those of us with noisy workshops, we can't do depth tests, but we can do nulling phase tests! Yea!

Why? I don't know either. But people seem interested in this subject as it might impart magical powers to their MD if they get the null phase just right...

Anyway, here are some results with unshielded coils. I slowly changed the overlap and measured the null voltage (at LF353 pin 7) vs. the null phase (relative to the TX signal).

I did this with the RX lead grounded and then with the RX lead ungrounded (true differential input, dfbowers wet grass configuration).

I also wanted to see how this compared to a simple theoretical model. This model is:

A sin(x) + B sin (x + 90), where A represents the pure coupling of the two coils and changes from a large number to zero to a large negative number as the coil overlap moves through the null point, and B represents a small capacitive coupling of the TX and RX leads that is 90 degrees out of phase with the TX signal.

I changed A from 8 volts to -8 volts, and let B = .3 volts, which is approximately the null signal voltage minimum observed.

The attached photo shows the results.

I plan to repeat these tests with other coils which I will then put different shields on to see the effect.

My hunch is that the radical shift of the null phase near minimum can be explained by the presence of another signal approximately 90 deg relative that is added in, and is not "magnetically" coupled and therefore not affected by ground balancing, and that the exact phase of the null signal is probably not critical. But I could be wrong.

It looks like the simple theoretical model may be improved by adding a small magnetic phase bias to the magnetic component, so instead of shifting from -90 to 90 it goes more like -85 to 85.

Anyway, more tests to come...probably...

-SB

Just wanted to add the information that I used a dfbowers PCB for these tests, and that the more negative phase side of the null corresponds to more overlap, the more positive phase side corresponding to less overlap of the coils. The cable was a USB cable of fairly good quality, with a foil shield, a braided shield, and copper shield wire. The TX and RX pairs are not individually shielded however.

-SB

[der_fisherman]

Just wanted to add the information that I used a dfbowers PCB for these tests, and that the more negative phase side of the null corresponds to more overlap, the more positive phase side corresponding to less overlap of the coils. The cable was a USB cable of fairly good quality, with a foil shield, a braided shield, and copper shield wire. The TX and RX pairs are not individually shielded however.

-SB

So if I understood you correctly, it would be better to err on the side of less overlap when setting the null? At least according to your recent testing.....

By the way, I am not asking for guarantees or anything, I just want to make sure I understood correctly....

Thanks

Andy

[simonbaker]

So if I understood you correctly, it would be better to err on the side of less overlap when setting the null? At least according to your recent testing.....

By the way, I am not asking for guarantees or anything, I just want to make sure I understood correctly....

Thanks

Andy

Hi der_fisherman:

No, I haven't concluded anything about how/where to set the null point, just sort of developing a theory about what we see.

I would actually like someone with a noise-free environment (cellar???) to do a similar test only measure the air-depth carefully as the phase changes.

Theoretically, I don't see why the null phase should affect the depth much if you are near the minimum voltage. But it will affect voltage on capacitors on C12, C15 (TGSL).

It also has some implications for the ground balance setting, but I need to study that some more. It may be more important which side of the null you are on than the exact phase of the null signal, and even then it may make less difference than most people believe. Of course there are many considerations besides air depth and simple ground balancing, such as margins of error, etc.

Regards,

-SB

[golfnut]

I want to add a nylon thumb screw which pushes on the coil to fine adjust the null next time.

S

[simonbaker]

I want to add a nylon thumb screw which pushes on the coil to fine adjust the null next time.

S

I like the idea (assuming the null is really that critical.. stay tuned..). I haven't found a clean construction design to do it myself.

Cheers!

-SB

Attached: nulling machine with foil target. No particular relevance...

[simonbaker]

Redo Null Phase Tests

One more time with feeling...

Realized that for my last null phase test chart, I had the RX coil leads reversed (you know, gives a double beep).

So I fixed it and redid the tests. Didn't really change anything.

However, I added a test with a 4 x 9 cm foil target at zero depth just to see what it does to the null signal.

My conclusions so far are:

- The "grounded RX Lead" configuration has slightly larger null signal and is shifted in phase by a few degrees from the ungrounded configuration.

- Metal objects near or in your search coil (hardware, circuitry, shields) may change the null signal by increasing it and may change the ground balancing point(s) by compressing the range of the magnetic component of the null.

More experimenting to come (maybe).

-SB

[simonbaker]

Commenting further on my last graph of the null phases:

1. I shouldn't have said switching the RX leads didn't change much. The graph looks pretty much the same, but now the phases corresponding to more overlap of the coils are the more positive phases, the opposite of the previous graph with the RX leads reversed. I guess this isn't any surprise.

2. Note that the previous two graphs were for a configuration using a standard USB cable. My next graph will compare data using a Belden cable to the USB cable for the non-grounded RX case, for the heck of it. I don't expect too much difference, possibly not even in the margin of error of my data points, but let's see.

3. By the way, I know my theoretical model of two sine waves 90 degrees apart is not exactly correct because my null signal becomes quite distorted near the minimum null voltage, doesn't look like a nice sine wave.

In fact, to measure phase, I measure the positive-going zero-crossing difference for one estimate and the negative-going zero-crossing difference for another, and average the two, because the waveform is so non-symmetric. Maybe dumb approach, but that's what I do.

To normalize the zero crossings, I center both waveforms (TX and RX) vertically on my oscilloscope center x-axis gridline, making the top and bottom peaks equidistant from the x-axis. This may not work so well for the RX null signal at minimum when the top half of the waveform may look sine-like and the bottom flattened, or vice versa, etc.

But hopefully averaging the two null-crossing phase differences help reduce the errors.

I noticed that Dennis the Mennis, in his tests, seems to have a nicer looking null signal as it moves across the minimum. Perhaps it is just noise in my workshop causing distortion.

Dennis the Mennis and I seem to be studying this null phase characteristic simultaneously, so it will be interesting to compare his results to mine, and the combined data may be helpful.

By the way, I'm not doing this to make a big deal out of null phase. In fact, I generally thought the null phase was not critical and depended on peoples shields, cables, etc. But it seems to be a topic of interest and so deserves some study in hopes of debunking any myths, and learning useful practices, etc.

-SB