Building Wolf Treasure [Tesoro Lobo]

[mermelado]

Phase between transmit reference and receiver changes with frequency. I can watch the phase increase and decrease while sweeping the excitation. I would like to post my results except that it is difficult to really pick out the individual traces, one obscures the other as they overlap, but I will try some more.
One thing to look out for, is that the receiver signal gets advanced when going into the JFET switches. The leading or trailing edge of whichever phase reference will appear to be leading the amplified rx signal peak significantly, at the point t where the demodulator output changes sign. Very tricky. I will try and show that too, but I am low on fuel so it will have to wait until next week. Time to go solar.

I still haven't received the pcb from ebay, so I'm not in a hurry

[rolling rock]

Phase between transmit reference and receiver changes with frequency. I can watch the phase increase and decrease while sweeping the excitation. I would like to post my results except that it is difficult to really pick out the individual traces, one obscures the other as they overlap, but I will try some more.
One thing to look out for, is that the receiver signal gets advanced when going into the JFET switches. The leading or trailing edge of whichever phase reference will appear to be leading the amplified rx signal peak significantly, at the point t where the demodulator output changes sign. Very tricky. I will try and show that too, but I am low on fuel so it will have to wait until next week. Time to go solar.

[mermelado]

Aaah, but C5 (and some other components that got penciled in) is not part of the original design. However, after changing the transmitter frequency, fiddling with bandpass filter components, and having to build a diy coil, playing with C5 may very well be a quick fix. I am trying to adjust anything else, to avoid having to add a parallel resonant capacitor. I figure there is a good reason the design is what it is.

I believe spice simulations help me get the circuit in motion, but I do not know what exactly will happen as a properly nulled coil is introduced, in the real world.
Will the receiver phase get advanced (I think) or retarded, and exactly how much, I do not know. I hope to find out in person. Keep in mind, the factory Lobo came with a concentric coil, not a DD, and they will each align somewhat differently, and well, everybody knows that.

The only reason I can think of for the original design to have no C5 and consequently the RX coil acts as a broadband coil. is that they had planned the possibility of using different coils. Perhaps with the intention of lowering TX to detect more common targets. I wonder if this will not vary the phase and null point? I have not seen the original Lobo design, so I can only speculate.

[rolling rock]

Thank you very much for that magnificent explanation, which will help me to take my first steps. But I confess that I do not understand that the RX coil is not tuned. In the schematic I've seen, C5 is in parallel with the coil, which in theory forms a resonant LC circuit.

Aaah, but C5 (and some other components that got penciled in) is not part of the original design. However, after changing the transmitter frequency, fiddling with bandpass filter components, and having to build a diy coil, playing with C5 may very well be a quick fix. I am trying to adjust anything else, to avoid having to add a parallel resonant capacitor. I figure there is a good reason the design is what it is.

I believe spice simulations help me get the circuit in motion, but I do not know what exactly will happen as a properly nulled coil is introduced, in the real world.
Will the receiver phase get advanced (I think) or retarded, and exactly how much, I do not know. I hope to find out in person. Keep in mind, the factory Lobo came with a concentric coil, not a DD, and they will each align somewhat differently, and well, everybody knows that.

[rolling rock]

I think I was wrong about that negative peak. Look for the positive peak to slightly overlap the G phase reference. Too much overlap and you will lose ability to cancel large silver items which may be a non-issue, but I don't know what happens after that.

As I wrote previous post, I was thinking that there is an inverter in the signal path.

And I was correct in that assumption. However, I was looking at the LOBI DIY schematic which shows U-10 and U-11 having + signs on their - inputs, and vice versa, and I got confused. Yes there is an inverter in the signal path.

[mermelado]

I think I was wrong about that negative peak. Look for the positive peak to slightly overlap the G phase reference. Too much overlap and you will lose ability to cancel large silver items which may be a non-issue, but I don't know what happens after that.

As I wrote previous post, I was thinking that there is an inverter in the signal path, which is incorrect. I don't have any laboratory or equipment, at this time, so I am having to trust what I see on my laptop. Sometimes I get the + and - signs mixed up.
Ooh, it's when I get Right vs Left all mixed up, well that can be a hairy mess.��

Thank you very much for that magnificent explanation, which will help me to take my first steps. But I confess that I do not understand that the RX coil is not tuned. In the schematic I've seen, C5 is in parallel with the coil, which in theory forms a resonant LC circuit. Regarding the phase verification with the ferrite discrimination and rejection test, I confess that I built a TGSL, which is probably the worst performing of the whole forum, in fact, I am thinking of posting a thread explaining how not build a TGSL, but still, what always went well was the rejection of the ferrite and the discrimination, no matter how you put the phase, both in front and behind or even in almost total cancellation. That is a mystery that one day, with time, I will try to solve


Edit:
I just found online a PDF related to the construction and calibration of search coils. I share it here, in case someone can be useful.

[rolling rock]

I think I was wrong about that negative peak. Look for the positive peak to slightly overlap the G phase reference. Too much overlap and you will lose ability to cancel large silver items which may be a non-issue, but I don't know what happens after that.

As I wrote previous post, I was thinking that there is an inverter in the signal path, which is incorrect. I don't have any laboratory or equipment, at this time, so I am having to trust what I see on my laptop. Sometimes I get the + and - signs mixed up.
Ooh, it's when I get Right vs Left all mixed up, well that can be a hairy mess.��

[rolling rock]

Hi:
In a previous thread the coil construction is specified, but not the offset between TX and RX is this 20 degree for DD coils like TGSL?

I think that what needs to be done is to have the amplifier output negative peak to approximately align with the leading edge of the ground cancel reference signal turned fully CCW (to the left). There should be enough overlap so that we see a slightly negative output from the demodulator, which changes to positive as the control is rotated to the right. I might be off by 180 degrees so that the positive peak is what you're after, then you would reverse one of the coil's two wires. Whether or not 20 degrees separation occurs, I can't say, and that angle may differ depending on the type coil i.e. DD or concentric. All the Rs and C's around the receiver amp U7 affect amplitude and also phase, so there are a lot of ways proper phase can be upset.. .
For my detector I am working with the Lobo rx amplifier, but then I am attempting to borrow the phase reference circuit from Tesoro Aztek. Aztek uses a differential coil and amplifier, and initial phase setup works differently from Lobo, but I found an easy way to rotate the references to get everything to line up correctly. Sorry, but my easy fix will not work quite so easily with the stock Lobo circuits. I will try and post some screen shots from my Lobo LTspice simulations when I can find a place to plug my laptop into the internet. It is quite revealing to be able to watch signal phase relations moving around as frequency is varied. One popular method of checking alignment is that you should be able to cancel a ferrite rod or toroid with your ground control turned way to the left.

Lobo does not use a tuned receiver coil, it relies on selecting R and C values around the rx mplifiers to form a bandpass filter. But we need to phase the rx output, and that probably will not happen by accident.

[mermelado]

Hi:
In a previous thread the coil construction is specified, but not the offset between TX and RX. is this 20 degree for DD coils like TGSL?

[rolling rock]

Is it necessary for the output signal to be 50% high and 50% low at 555 pin 3?

You will never get perfect 50% duty cycle. I am accustomed to seeing almost perfect 50% dc on my computer simulation, but the exercise comes with a disclaimer*. So with that said, could you show us what the rx amplifier output looks like in the same screen as the ground clock reference?
Those would be at U7-7 and U5-1. Turn the ground pot to zero, then do the same for the 100% position.


*You know, the "spice simulation is not real blah blah" and I agree 100%.

[mermelado]

Hi: any body know, what's the practical diferences vs tsgl? Thanks.

[kt315]

they want to get 20kHz while Lobo is 17500, they changed the values of R and C for this and get the asymmetry.

[rolling rock]

Thanks for your rolling rock advice.

Is it necessary for the output signal to be 50% high and 50% low at 555 pin 3?

Hmm... how much wiggle room are we talking about? 'By design', your circuit should give you very nearly 50-50 dc. The fact that I can detect that not being the case, suggests that you perhaps have a defective '555. Don't tell me, you got if from China (not to beat up on the Chinese, those scurrilous, counterfeiting rascals).

Seriously, any self respecting '555 (no matter what flavor!) should be able deliver 50-50 duty cycle at 17500 or even 20000 Hz. (Just my humble opinion but surely, I am uninformed.) So, that really was that a lopsided waveform? Good, I thought maybe it was just because I was sitting sideways, because I was sitting sideways, therefore I was a little concerned. Too much, really. Just remember, KT315 saw it first.

If you have nice symmetrical square waves coming out of your phase references (U5-1, U5-7, U6-1) then you may have nothing to worry about. That is for you to decide. Non-symmetry may cause some performance loss, but it won't break your machine.

[Op04]

I forgot to remind you, vlaser version I had trouble poping sound, silent after change 4016 to 4053, the picture is taken before I put some modification

[Op04]

Hi,
It's there on post #57 I don’t have more picture, I have to pull from the case, sorry can't do that now, last version from vlaser, the pcb smd enlarged to fits Dip IC, not for long only working for a month, so I don't want to discuss here, it was the third board adds to my fail. I hope you make this version working, I still build a PI project now, I believe you will success building lobo.