Frequency checking TGSL question

[billr]

I'm about ready to start making coils. But before I do I wanted to clarify the frequency stage aspect of the build which I dont think is covered in the manual.

Is the frequency for TX and RX coils checked with the USB2 cable attached to the coil via the pcb or is it checked by connecting each coil to the pcb by the short start and finish tails of the coil? I would assume the former is correct.

I ask because it would not surprise me if the USB cable altered the frequency from 14.5 and 13.95 to less or more than required and change the overall performance.

Thanks

[waltr]

I check coil/oscillator resonance without the cable and it was still right on with the cable.

The coils are 6.0 & 6.5mH so the cable does not effect frequency enough to matter.

I'm about ready to start making coils. But before I do I wanted to clarify the frequency stage aspect of the build which I dont think is covered in the manual.

Is the frequency for TX and RX coils checked with the USB2 cable attached to the coil via the pcb or is it checked by connecting each coil to the pcb by the short start and finish tails of the coil? I would assume the former is correct.

I ask because it would not surprise me if the USB cable altered the frequency from 14.5 and 13.95 to less or more than required and change the overall performance.

Thanks

[billr]

Thanks again
Ok so new developments.
Made two coils this morning as a test sample whilst waiting for parts. Did like the pdf stated. Bound on the former without constraints gave me 5.7 and 6.3 mH as predicted by I think ivconic. After constraining with zip ties every 3/4 inch inductance went to 6.0 and 6.5 respectively. However. the frequency for each is quite some way off. Tx = 13.97 and Rx = 13.low something.

If I add more turns then the freq would be correct but the inductance would be off.
So I guess the question is which is most important. Freq or inductance?
Wire used 0.25mm cu.
After several attempts making the TX coil with 99 turns I cant see how it's possible to achieve both numbers.
The zip ties could not have been any tighter and my inductance meter claims + or - 2% at 20mH setting.

One last thing on a different subject.
I seem to be the victim of buying re-labeled LM308's as the lettering wipes off when rubbed with a mild solvent. Also looks like they have been sanded. I do have some TL081 chips. Can these be used instead and does it require any modifications to the pcb or are they plug and play?

[surfdetector]

Thanks again
Ok so new developments.
Made two coils this morning as a test sample whilst waiting for parts. Did like the pdf stated. Bound on the former without constraints gave me 5.7 and 6.3 mH as predicted by I think ivconic. After constraining with zip ties every 3/4 inch inductance went to 6.0 and 6.5 respectively. However. the frequency for each is quite some way off. Tx = 13.97 and Rx = 13.low something.

If I add more turns then the freq would be correct but the inductance would be off.
So I guess the question is which is most important. Freq or inductance?
Wire used 0.25mm cu.
After several attempts making the TX coil with 99 turns I cant see how it's possible to achieve both numbers.
The zip ties could not have been any tighter and my inductance meter claims + or - 2% at 20mH setting.

One last thing on a different subject.
I seem to be the victim of buying re-labeled LM308's as the lettering wipes off when rubbed with a mild solvent. Also looks like they have been sanded. I do have some TL081 chips. Can these be used instead and does it require any modifications to the pcb or are they plug and play?

The TL081 is not pin for pin the same, #8 is NC whereas on the TGSL #8 is connected to the compensating cap.

[billr]

Yep that is why I was asking if the pcb needs modding.

[waltr]

Thanks again
Ok so new developments.
Made two coils this morning as a test sample whilst waiting for parts. Did like the pdf stated. Bound on the former without constraints gave me 5.7 and 6.3 mH as predicted by I think ivconic. After constraining with zip ties every 3/4 inch inductance went to 6.0 and 6.5 respectively. However. the frequency for each is quite some way off. Tx = 13.97 and Rx = 13.low something.

If I add more turns then the freq would be correct but the inductance would be off.
So I guess the question is which is most important. Freq or inductance?
Wire used 0.25mm cu.
After several attempts making the TX coil with 99 turns I cant see how it's possible to achieve both numbers.
The zip ties could not have been any tighter and my inductance meter claims + or - 2% at 20mH setting.

One last thing on a different subject.
I seem to be the victim of buying re-labeled LM308's as the lettering wipes off when rubbed with a mild solvent. Also looks like they have been sanded. I do have some TL081 chips. Can these be used instead and does it require any modifications to the pcb or are they plug and play?

If the coil inductance are correct and frequency is wrong then the resonant Caps are off value.However, the Exact frequency is not that important, it is the difference that is important.

You will know much better when you connect O'scope to TX and RX pre-amp and adjust for NULL. If the coils Null well with a phase shift near 20 degrees then all is good.

I used TL081 for the LM308. NO circuit mod is needed even if you installed the 100pF Comp caps since these are on NC pins.
Just PlugnPlay.

[billr]

Okey dokey. Based on that. I'm going in. Well as soon as my parts arrive I will. So far I am in deep research mode. Reviewing everything. Trying to de glitch the inevitable (which is actually more demanding than building the bugger) Currently have only a few IC sockets installed to date.
Gonna mod the two coils I made to 6 & 6.5 mH regardless of freq and see what transpires. Prolly make a set of coils to match frequency and see which is best.
Building metal detectors is so................um................yes I had pancakes on Friday morning thanks.

[waltr]

That's the way to do it, just get in there and build then test.

If the coil inductance is correct and then oscillate lower in frequency then re-check the Cap in the TX circuit.
C2 (22nF) cap you have installed seems to be too large.
Check the 22nF caps you have with your C meter and pick the one that is closest or less than 22nF.
If lower then easy to add another cap in parallel to increase C.

You may also find the RX cap, 15nF C6, to not be the correct value. If not the correct ratio to C2 then NULLing phase will be off. Again measure this cap before installing.

When Nulling and measuring phase you can add small cap across either the TX, C2, or the RX, C6, this changes the TX oscillating Frequency or the RX's resonant frequency.

Since Scope goes to TX coil and RX pre-amp to NULL you can start looking at this before finishing the remaining circuits.
You will need more pancakes for breakfast to do this step.

[billr]

Have sent wife on mission to buy more pancake mix in readiness for the build. She's been gone hours. Might have left me . Oh well.

Some productivity today whilst waiting for parts. Altered the two coils already made that matched frequency beforehand by adding more winding's to equate to inductance value. Now happy with final figures of each coil. I used aluminum tape as a shield but kept the backing paper on for now as I want to finalize the coils to fit an existing coil shell I have nominated for the job. Until I know how much overlap they need to null out and their correct diameter, I can't commit to making a permanent coil. Not really stoked about making shells. I know the coils aren't pretty but should be enough for the initial setup. Least get me in the ballpark. Passed continuity tests when checked.

waltr.
Really appreciate the info regarding the caps and their placement, and the extra how to do it info. Stuff like that really helps a beginner like myself and I'm sure others. May seem trivial to someone with your knowledge but to me, without it, I'm like. What the Sam Hill did he just say?

Oh yeah nearly forgot.
Have found a few caps at 20-20uF and a few 14uF. Maybe after viewing the attached pics you might say I need to go much lower per cap due to the frequency difference and add to it later.

Oop! Gotta go. Wife's returned and giving me THAT look.

[billr]

Correction to last post missed edit window opportunity to correct. nF cap values not uF.
6'6", 300lb gorilla + small keyboard = typo

[waltr]

Looking good and glad to see you corrected uF to nF.

When you overlap the coils to Null be sure to insulate the shields from each other, else you end up with 'shorted turns'.

What part of the World are you in??
Please add to your profile. This helps us recommend places to get things.

I buy coil shells from Dbowers here in the USA (he is actually only a few hours drive from me).

[billr]

Right now I'm in the UK but not from here. From MA, so almost your neighbor.
Not too many places sell detector parts here. Those that do really charge a bunch for what you actually get. Thought about buying a pre made coil and housing for the project but that would defeat what seems to be the toughest part of the build. I don't enjoy pain but like a challenge, especially when the realization bulb comes on in the empty space between my ears.

Besides. If this goes well I might try the IDX. I hear the coil for that is quite difficult to make. Not ready for the Voodoo yet although I have it.

[Qiaozhi]

Right now I'm in the UK but not from here. From MA, so almost your neighbor.
Not too many places sell detector parts here. Those that do really charge a bunch for what you actually get. Thought about buying a pre made coil and housing for the project but that would defeat what seems to be the toughest part of the build. I don't enjoy pain but like a challenge, especially when the realization bulb comes on in the empty space between my ears.

Besides. If this goes well I might try the IDX. I hear the coil for that is quite difficult to make. Not ready for the Voodoo yet although I have it.

There's one on ebay at the moment ->
https://www.ebay.co.uk/itm/Tesoro-11...wAAOSwL1tfzrTa

[billr]

Wow. That was fast !!
I'll send the seller a message.
Thanks George.

Um
Will I get in trouble for my location? It's kinda off the page by about 67 letters.

[billr]

Alrighty.
Parts arrived and fitted.
Unfortunately I won the Friday afternoon production batch of BF245c's. One tested good, the other not so much. Fitted it anyway and prayed for a miracle. Didn't happen.
No replacements and no substitutes so re ordered.

Went on to nulling.
Can't believe how sensitive this procedure is. Used a small nylon screw and block to finalize the voltage check as indicated in the advanced nulling section. Got it down to 100mV in accuracy.

Checked steps 2 and 4. Images shown.
I left the cursor menu open in case I did a boo boo with the settings. Still getting used to this new scope.

Weird stuff.
When the pcb is turned on, it takes 60 seconds before it will detect a can at 20" sounding a beep. The lower pulse in step 4 can be seen to rise and fall as the can is passed over the coil but no sound until after one minute left on. This only happens once the MD starts to beep on its own accord, at which stage it goes into a constant beep beep beep mode at 2 p/Sec The can is still detected but the beep for it comes on in between the constant string of beeps.
Not sure if this is due to an incomplete circuit TR5 failure or something else.
Also noticed when the GB pot is adjusted whilst running the beeps stop for a short time even if the can is passed over the coil and then restart after 20 seconds. Think I read somewhere the GB pot needs grounding?

No amount of adjusting the sense, disc and GB pots alters this behavior.
The coil is off the ground by some 36" and not near any obvious metal source.
If it wasn't raining I would take it outside and test.