TESORO GOLDEN SABRE

[Geo]

Works

Ok the GoldenSabre works. I rotate the MPSA13 180degrees and now it is ok. I connected it only in manual mode but i dont like it. I null exactly the coil (with the scope at the out of LF353) but i did not connected any capacitor to the RX coil for fine tuning (there is only one capacitor 15nF on the board).
Results:
Audio........ very low.... it needs a audio amplified (LM380... 386). As i see with the scope there is change in the signal that speeker don't sound, so we loose depth.
Sensitivity........ medium.... 2euro coin at 25 cm , 50Euro cents 25 cm, 2cents at 18cm, aold greek coin 30mm at 28...30 cm.
MAybe today or tomorrow i will try to fine tune the coil and to see the max semsitivity of this detector. If the sensitivity is near at my yesterday test then Ankerr SS-60 is Big Winner.
Max can you tell me at what distance you can detect a 2Euro coin???
Regards

[Qiaozhi]

Yes this will do, in case you are using exact coil on exact device. But if you want versatile coil, better to null it directly attached on scope. Front end can add some noise,jitter,suck some interefernces due some bad component etc... I choose this way due fact that have 8 devices and 16 differnet sized coils...Many of components arent new, taken from scraped electronics...I do have doubts in some of those.
But anyway, you have right. Max done the best method - both. First without front end and after than with. So you can easily see eventyal differences...
Both ways are correct.

For this exercise, the noise generated in the preamp should not be a problem. Don't forget that this is really the signal that you're sampling, not the one from the coil. If you have noise issues here, then there's a real problem. Also remember that the coil will only be properly balanced for the frequency it was designed for. The only way you can keep your coils consistent between detectors is to put the tuning capacitors in the coil, not on the PCB.

Hi
Thanks to all for your answers!.I understand your respective points.
Basically,you are doing so for research purposes, If you want to build ONE (perfect) detector, just using the preamp output is probably the best way.
Ivconic, i work often with radio frequency,so i know what you mean , an "unmached" input can raise the noise figure etc...tsss...perfectionists!
Keep the good work!
Tomorrow i will look for a piece of acrylic glass to make my coil form.

(Im Fred, but i don´t have my password here so i can´t register.)

Best regards!,

Fred.

[ivconic]

Yes this will do, in case you are using exact coil on exact device. But if you want versatile coil, better to null it directly attached on scope. Front end can add some noise,jitter,suck some interefernces due some bad component etc... I choose this way due fact that have 8 devices and 16 differnet sized coils...Many of components arent new, taken from scraped electronics...I do have doubts in some of those.
But anyway, you have right. Max done the best method - both. First without front end and after than with. So you can easily see eventyal differences...
Both ways are correct.

[Qiaozhi]

Hi,
I have been reading your posts with great interest, thanks for sharing your experiences.I have learnt a lot.
About coil nulling, i wonder why you dont measure your voltage at output of 1st opamp instead at the coil istself? this way measurment would be more precise (already amplified), no influences from multimeter or scope probes ,and connecting cable capacitance and resistence already there, as in final use.
Or i am missing something?
regards,
Fred.

No - you've not missed anything. That's the way I do it.

[Max]

Hi,
I have been reading your posts with great interest, thanks for sharing your experiences.I have learnt a lot.
About coil nulling, i wonder why you dont measure your voltage at output of 1st opamp instead at the coil istself? this way measurment would be more precise (already amplified), no influences from multimeter or scope probes ,and connecting cable capacitance and resistence already there, as in final use.
Or i am missing something?
regards,
Fred.

Hi Fred,
you are right... one can also use just 1st preamp to do that... and personally I've made too a number of times... but just prefer having passive components on rx coil when null with scope... cause I actually want measure absolute value without considering other scaling factors like gain of first amplifier.

Now a very easy test I made, after measuring with just scope I connect coil to pcb and read output at first preamp... and of course I read the value measured by scope scaled by gain of first stage... but I look also for distortions due e.g. to bad capacitive effects on preamp... and compare with (perfect) sinus I get from just probe measure...

That way I'm almost sure (if I see again perfect sinus) that 1st preamp doesn't introduce noticeable distortion and similar problems.

But that's just my way of testing things...and others would surely disagree repeating things in the two ways.

Best regards,
Max

[Fred]

Hi,
I have been reading your posts with great interest, thanks for sharing your experiences.I have learnt a lot.
About coil nulling, i wonder why you dont measure your voltage at output of 1st opamp instead at the coil istself? this way measurment would be more precise (already amplified), no influences from multimeter or scope probes ,and connecting cable capacitance and resistence already there, as in final use.
Or i am missing something?
regards,
Fred.

[ivconic]

MPSA14 with EBC ..... Yes i have to reverse it due oposite e and c ...
You solved problem GEO!

[Max]

I think that i found the problem. As i saw at the datasheet for MPSA13 and BC517 there is a mistake at the pcb aboyt the position of MPSA13.The position is right for BC517 but not for MPSA13. I must reverse 180 degrees the mpsa13 to work. What to say. Ivconic or Max what transistor do you use..... mpsa or bc?
When i will go home i will try it

Hi,
yes you have to reverse 180° from BC517 position.
I use japan made C517 (that's the same of BC517) and it rocks.

Best regards,
Max

[Geo]

I think that i found the problem. As i saw at the datasheet for MPSA13 and BC517 there is a mistake at the pcb aboyt the position of MPSA13.The position is right for BC517 but not for MPSA13. I must reverse 180 degrees the mpsa13 to work. What to say. Ivconic or Max what transistor do you use..... mpsa or bc?
When i will go home i will try it

[Geo]

Hi Ivconic.
At the out of LM308 i have good signal (with a coin at 25cm), the same at the out of LM393 and the input of the last LM358. From here begin the problem.For me the problem is around the LM358.... MPSA13 and maybe the BF245C for the battery check, When i will return from work i"ll check or replace it. Finally if i don't find the problem i will connect the pcb with 4N35....555 Lm386 and will be OK.
Who is the difference at sensitivity between "all metal" and discrimination"
Regards

[ivconic]

Did you put compensation on LM308H ? Very important! If you omit those capacitors; either it will not produce audio either it wiil scream as you described. Or maybe you put bad caps there? Also it will scream if you turned LM308H wrong way, pin1. under notch?Like i did! But we already established that? Pin 8. is under notch! Remember!
It has something to do with LM308 or arround those...Check all over again...
Also you might be turned fets reverse? Check again cose i noticed wrong designating on some schematics here...
Holding thumbs for you Geo!

[Geo]

GEO,
Use BF245A, or BF245B. According to spec sheet, the "C" version might not turn off until -7.2V is applied. There is no -7.2 volts in the system, unless you use a different voltage inverter!
Also, could use 2N4393, J107, J113, TIS75, etc.
I think the most critical spec is probably Vgs(off), it should absolutely be no more than whatever the negative rail can supply.
I'm talking about the demodulators. BF245C can work in the transmitter, but I won't say that it will work the way that it was designed.

Hi porkluvr
Fets seem to work fine. But when i will finish the detector i will give a try with 245B. But now i have other problems.
Anyway thanks
Regards

[Geo]

Tomorrow again.
What i found ????? First of all a problem with coil. I connected it right and now the oscillator works at 14.1khz.The 480Hz signal (10Vp-p) was from false connection of scope, but i hear yet it. Until the input of the last LM358(pin5) the detector works fine at all metal mode. With the scope i have very good sensitivity. But what happened and i haven't sound??? I only hear a low continuous tone and nothing else. I changed the LM358 but nothing. I checked the MPSA13 but nothing. Scope shows a small quad pulse (about 0.5V). It seems that battery check works, but voltages are right.
I dont know until tommorow
Any opinion?????
Regards

[porkluvr]

wrong FET

GEO,
Use BF245A, or BF245B. According to spec sheet, the "C" version might not turn off until -7.2V is applied. There is no -7.2 volts in the system, unless you use a different voltage inverter!

Also, could use 2N4393, J107, J113, TIS75, etc.

I think the most critical spec is probably Vgs(off), it should absolutely be no more than whatever the negative rail can supply.

I'm talking about the demodulators. BF245C can work in the transmitter, but I won't say that it will work the way that it was designed.