lobo improved

[maxbiri]

Hi porkluvr

Thans for shematic, i will study it.
If you need same measures i can take them on my board, i have a dual traces oscope.
I measured current of oscillator and have these results:
3mA at 5 volt with a indirstorted signal of about 9 v PP
40 mA at 12 volt with again a signal of 9v pp
these are current values that feed final transistor of TX, all circuit, without final tr, drink 15 mA.
Seem that change from 5 to 12 volt have not good result; i found this very strange but need more test.

Bye Max

[porkluvr]

Foma posted a scat elsewhere, but I couldn't locate it to post its link. But, since I've made some additions, I may as well post it here - although I don't like it's large size.

I will try anything that works. I have problem because I want to build Lobo - but designing around a Silver Sabre coil, and with a Golden Sabre oscillator (tuned to 21.6 kHz) using +/-5V oscillator supply. The TX reference is C coupled to an active low-pass filter which drives the discrimination and GCX phase reference stages. I use GS oscillator because I can heterodyne the TX coil frequency against oscillator of LTC1044A voltage converter, and extract 500 +/- 200Hz audio (?!). Lobo TX is not good for that due its coil not being a frequency determining device.

I'll eventually, probably, wind a Lobo coil and replace the tank caps. But design basics are the same whether I do or not. Consider that all I had before were "?" for tank capacitance values; so what's the difference If I've redesigned the coil circuits?

Even though I have gotten away from the original design (and because of that) I need data from your project for reference. I use your "New Image Lobo" along with scat posted by Shawn Goddard as starting point... . Using 25pf cap instead of 250pf, while using 8.2nf at the coil seems to make everything good, in my perception. That is what I now use as baseline for my design. I may have ill-conceived notion about what is "good", but that's where I am. I was running out of ideas. If it doesn't work then I will make changes.

I take off r40 and schortcut r37 and removed r49.
Seem there arent difference respect before.
About 250pf condensator i think is correct, if you look at diablo schematic, that have the same input stage, they use again 250 pf.
Bye Max

R49 not so much affects total discrimination as it affects taper of potentiometer it is attached to; degrees phase reference change per rotation of dial. You'll still have almost exactly the same amount of phase change between 0%~100%, but the gradient will be different in-between.

You could parallel a 27p or 30p cap across the existing 250p for a quick test, but I'll shut up.

Well, anyway, good luck. If I ever get mine built I'll be sure and let you know how it turned out

[simonbaker]

Hi all

@ porkluvr
where is russian schematic with 8.2 nf on rx coil? i havent it.
you are right r36 and 50 are 10 k on pcb, i forget again 100k on schematic, i will correct it.
I change c 13 from 100pf ceramic NPO type to polyester 2.5% and c 47 from 47 pf ceramic to 47 pf polyester 5%, i'm no able to find 50 or 51 pf, i think is not a problem few pf of difference.
I take off r40 and schortcut r37 and removed r49.
Seem there arent difference respect before.
About 250pf condensator i think is correct, if you look at diablo schematic, that have the same input stage, they use again 250 pf.
Now i' m schielding coil with grafite, alu dont work well.
This evening i hope restart with new test and try take same photo from oscope.
@ simon
At this first step i'm driving coil at 5 v, after when i will solve problems i will drive tx stage at battery power, if you look at my schematic i put a jumper for select 5 v or battery power for driving bc307.
@ foma
I have tried even as you suggest, using 1.3 mH as tx with 47 nf and rx 10 mH with 4.7 nF, it work in yhe same way, when correctly set.

Bye Max

Thanks for info; I understand your approach with voltage for now, so we shouldn't worry too much about depth yet.

Looking forward to continued development of the circuit. Hopefully it will eventually work like original Lobo, because that is a detector I would like to try building some day, so very interested in your success.

Regards,

-SB

[maxbiri]

Hi all

@ porkluvr
where is russian schematic with 8.2 nf on rx coil? i havent it.
you are right r36 and 50 are 10 k on pcb, i forget again 100k on schematic, i will correct it.
I change c 13 from 100pf ceramic NPO type to polyester 2.5% and c 47 from 47 pf ceramic to 47 pf polyester 5%, i'm no able to find 50 or 51 pf, i think is not a problem few pf of difference.
I take off r40 and schortcut r37 and removed r49.
Seem there arent difference respect before.
About 250pf condensator i think is correct, if you look at diablo schematic, that have the same input stage, they use again 250 pf.
Now i' m schielding coil with grafite, alu dont work well.
This evening i hope restart with new test and try take same photo from oscope.
@ simon
At this first step i'm driving coil at 5 v, after when i will solve problems i will drive tx stage at battery power, if you look at my schematic i put a jumper for select 5 v or battery power for driving bc307.
@ foma
I have tried even as you suggest, using 1.3 mH as tx with 47 nf and rx 10 mH with 4.7 nF, it work in yhe same way, when correctly set.

Bye Max

[FOMA]

Hi maxbiri.
I did so:
L (Tx) - 1,33mH ; R(Tx) -11om
C = 47H f = 17,5 kHz

[simonbaker]

Hi Simon

I forget said that all my test are done feeding tx stage at 5v so all circuit can be powered at 9 volt, 6 AA or AAA battery or even a single 9v block so i presume a light device.
With new power stage i have a stable simmetrical +/- 4.95 volt, less than 5 mv difference.


Bye Max

Ok, 5v quite different from 8v -- I think you have 40% less coil current and lose some amount of depth because of it, no?

Regards,

-SB

[porkluvr]

Try this?

Maxbiri, I wish I had thought to ask you what frequency you have set.
I have the freq.pot set at 50% and have tweaked parallel (parasitic) properties of the 10mH coil so that I have a good peak
using a 4.7nF tank cap. Adjusted drive pot for a -reasonable- combination of amplitude and waveform symmetry.
TX setup is complete (fo=20.4kHz).

Here is how I would set the RX phase:
The russian (?) drawing shows 8.2nF parallel in the receive tank. Give it a try.
Maybe the cap marked 250pF is actually supposed to be 25pf? I try that.

According to your schematic, you have placed 100k resistor across the GCX pot. Remove it, or put it from the BOTTOM terminal to the wiper.
That makes a big difference. 100K gives 35 degrees of adjustment. 300k increases the range to about 48 degrees.
A whopping 55 degrees of GCX adjustment results from removing the 100k altogether.

Probably better to use 10k resistor instead of 100k on the LM393 input(s).

Now, take a look!!! One of the coil's leads may need to be reversed - one day maybe I understand why. Not today.
***********************************
The red trace is the RX output. Purple (it looks brown) is GCX ZCD output, knob turned CCW. Orange is GCX ZCD, knob turned CW.
(I hope CW and CCW are not reversed. I find this all very confusing.)

I think this is pretty close to what is supposed to happen, except that I would delay rx signal maybe 3 more degrees?? Changing the 25pF to 33pF
would take care of that. Maybe it's not necessary, but would assure that received signal of neutral soil is within range of GCX CW adjustment?

Putting a 250pf capacitor across the 48k7 resistor shifts the rx signal about 45 degrees to the right. I think that would ruin your GCX channel's response.

I thought I had this thing figured out before? Um,hmmm. That shows how easily I can fool myself. If this setup is wrong, I hope
somebody will explain why. I have no scope to check this stuff in real world.
I might have the CW to-the-left and CCW to-the-right movement mixed up, I dunno.
Heck, this whole thing might be wrong, but it looks good from where I see it.

[maxbiri]

Hi porkluvr

I decide 4.7 nf by testing with oscope.
Before i set tx at 20 khz and after try different capacitor try finding resonance.
My measure are with oscope so mybe not very exactly, i presume +- 10 %.
I try even reverse tx and rx and device work more oe less in the same way.
On rx coil we need a tank capacitor for decrease noise, rx must not be at 20 khz resonance for best perfomance, i need more test for finding better solution.

Bye Max

[maxbiri]

Hi Simon

I forget said that all my test are done feeding tx stage at 5v so all circuit can be powered at 9 volt, 6 AA or AAA battery or even a single 9v block so i presume a light device.
With new power stage i have a stable simmetrical +/- 4.95 volt, less than 5 mv difference.


Bye Max

. I think that 23.2 is almost beyond the high end range of the RC components on the 555, certainly not mid-range.

If you are relying on simulation results, then you are probably not accounting for cable and coil capacitances to lower the output frequency to within range of the RC timing network.

[porkluvr]

Maxbiri, how did you decide that you needed 4.7nF capacitor for your TX coil?

I think that 4.7nF in the receiver tank will resonate slightly higher than the circuit is intended. I look for 6.2nF to resonate with 10mH at about 20.2kHz - (with the frequency trimmer set approximately at mid point). Why have you chosen 4.7nF?

The Lobo manual indicates an operating frequency of "about 20Khz". Your tank circuit wants to work at 23.2kHz . I think that 23.2 is almost beyond the high end range of the RC components on the 555, certainly not mid-range.

[simonbaker]

Hi Simon

Yes i built newlobo as in last schematic, it work but with same problems.
I'm testing a DD eliptical coil 14*22 cm.
The problems are:
1)Poor deep, i can detect a 50 eurocent at about 20-22 cm in disc.
2) if i set newlobo for best perfomance in all metal i have poor perfomance in disc and viceversa.
3) disc work in strange mode; at minimun disc set i disc big and smal iron target( e.g. hammer head 250 gr.)but i can ear, in clear mode, smal alu foil like this in cigaret box, if i increase disc i lost alu foil but even 50 € cent!
Is not sensibile at smal target e.g. small lead sferic 4.5 mm diameter.
I presume i have phase problem in coil or around lm339 comparator.
I need change small capacitor from ceramic NP0 tipe to polipropilene or polyester type and try a new coil.

Bye Max

Thanks for update. I am interested in Lobo because Tesoro made Lobo super for depth and I've seen results of person who uses it every day. Excellent for both coins and small gold nuggets. I hope we can get a circuit working as well as Tesoro. Only drawback is big heavy box on original.

Cheers,

-SB

[maxbiri]

Hi Simon

Yes i built newlobo as in last schematic, it work but with same problems.
I'm testing a DD eliptical coil 14*22 cm.
The problems are:
1)Poor deep, i can detect a 50 eurocent at about 20-22 cm in disc.
2) if i set newlobo for best perfomance in all metal i have poor perfomance in disc and viceversa.
3) disc work in strange mode; at minimun disc set i disc big and smal iron target( e.g. hammer head 250 gr.)but i can ear, in clear mode, smal alu foil like this in cigaret box, if i increase disc i lost alu foil but even 50 € cent!
Is not sensibile at smal target e.g. small lead sferic 4.5 mm diameter.
I presume i have phase problem in coil or around lm339 comparator.
I need change small capacitor from ceramic NP0 tipe to polipropilene or polyester type and try a new coil.

Bye Max

[simonbaker]

Hi, did you successfully build Lobo? Any test results?

Cheers!

-SB

[maxbiri]

This is last schematic revision.

rx coil is 1 mh + 47 nf
tx coil is 10 mh +4.7 nf
i reversed rx coil and now no more double ring on target.
Disc now seem working, not perfet but working.
coil test not shielded, DD eliptical about 22*14 cm.
rx phase -10,-20° respect tx, tx signal about 9 volt PP rx about 4-5 mv PP.