TGSL coil help needed

[slug54]

Hello everyone,
Long time lurker first time poster.
Today I started winding the coils for my TGSL board and I have been going by the instructions in the project document and Ivconic's coil winding document.
For the TX coil I wound 99 turns of 30awg wire on the single sided D form pulled off a few extra feet on the spool and cut the wire stripped the enamel to measure and I get 5.2mh instead of 6mh. I was able to wrap the couple extra feet to bring me to 101 turns and 5.4mh. the wire is currently not constrained but is fairly tight on the form.What is the best way to correct it from here? Can I just solder an extra length of wire on, tape the connection and wind about 5 or 6 more turns? Or should I constrain it and see how close it gets to 6mh? Or start over and wind a lot of extra turns to try to get over 6mh and peel off turns until it's right?
Thanks in advance for any help you can provide.
-Jeff

[dbanner]

Yeh, I wound several coils for tgsl. Never enjoyed doing it.
Look here, just get it close to 6mH as you can. You get the 14.5khz from tweaking with the capacitance of the oscillator.
The critical thing here is 14.5khz after shielding and put in coil shell.
The inductance and resistance are only approximate targets to aim for when winding the dam thing.

[dbanner]

Remember that the capacitance of the colpits oscillator used is effectively about 21nF.
By adding small puffs across the coil effectively increases total capacitance and lowers the tx frequency of the colpits.
So by aiming for higher than 14.5khz with the inductance of coil, then adding capacitance to get you back down to 14.5khz did it for me. After that, I didn't much care if my coil was 5.7 or 5.9 or 5.5 mH. As long as I was in the ballpark.
More than that, I didn't much care if my tx oscillator was 14.5 or 14.7 or 14.3 khz. I would tweak with the Rx resonance to get the spread I wanted. Which was 1.6khz. when balanced, I still got the 20degrees shift.
Hope this helps.

[homefire]

Wait until you have the coil wrapped tight for your end numbers. Things are going to be changing.

[slug54]

Great info thanks,
That makes a lot of sense.
I'm going to constrain the coil and add the tape and shield and work out the frequency with the caps later.
I appreciate the help. -Jeff

[samsearcher]

Finally finished the TGSL coils (for the second time!) they are now 'spot on' frequency (14.5KHz TX & 13.95KHz Rx) when nulling, the coils are in the position shown in the photo, i.e. the RX coil wont fit the shell! I may be able to bend the coil to fit (as I use epoxy, not superglue for the windings) but that will alter the inductance and frequency!.....Help please
The waveform is from pin 5 U103, balance pot at centre, it moves the zero from top to bottom of display (O.K?)

[kt315]

you do not want to understand that we are not recommended to use the foil.
with right shielding you can do a splitted out small winding and BY IT to do balance.
look pic


your problem people you do not want to read the forum. you just do what you want to do. in result you get the garbage.

[samsearcher]

I've been reading this forum for weeks with posts going back YEARS. Using Mylar as recommended here. It's actually working quite well before potting!

[dbanner]

Very good progress. Solution for your problem is easy.
Just bend and shape the Rx coil to fit in the coil shell, nevermind the inductance changing. You can do rough balance by connecting multimeter or scope directly to Rx coil. Do this to make sure you have enough room to shift the coil side to side, for when you are doing final balance.
Once you have done this, just measure the new inductance and use resonance calculator to figure out what value capacitance to use for 16.1khz resonant frequency of Rx. Now you can put new value capacitance in parallel with Rx. Remember there is 15nF cap already on PCB, you will need to add or subtract from its value in order to get the value of what you calculated.
Once that is done, you can go ahead and balance the coils.

[dbanner]

I've been reading this forum for weeks with posts going back YEARS. Using Mylar as recommended here. It's actually working quite well before potting!

Focus on balancing the coil and learning how to accomplish this black art. This is learning process of most difficult part of building a metal detector. Your knowledge and skills will improve. Soon you will be making search coils with eyes closed.
KT315 suggestion for you will overcomplicate your task and may further confuse you down into rabbit hole.
That solution is for people who are master coil builders who knows what they are doing.
This is your first coil. Keep it simple.

[dbanner]

There's better D shape coil( in my opinion) more widely used with commercial coils. That one is much better shape and lends itself more readily for fitting the coil shell and for balancing. Performance is the same. Notice in the pic above how the coil shape is much gentler with softer curves. This is best shape to follow. But keep your sharp d shape. Next coil you make you can try softer shape.
With your Rx coil, you have to shorten the corners so the D becomes deeper in order to have enough overlap of tx to get it to null point. Nevermind the small change in the inductance. Just get it to fit and to be able to null.Follow above.

[dbanner]

Factory made coils are wound on machines under tight qc, the inductance is bang on every time, they fit in coil housing perfectly every time. Shield done on coil shell perfect every time.
They balance perfect every time. Making coil on jig at home is completely different ball game.
There is going to be a lot of tweaking until you get the hang of it.

[samsearcher]

Thanks for that (and apologies to Slug for jumping in on his thread), yes coil winding certainly tests your patience!
The Minipulse was MUCH easier altogether

[samsearcher]

BTW.... I am reading coil frequencies at CLK pin 1 U100.....is that correct, It's higher if I read across the coil?

[dbanner]

Read it from wherever you like, from coil, or from connector, or from oscillator, so long as it says 14.5khz or thereabouts. I f you read 14.6 then set Rx resonant frequency to 16.2, get it? You should maintain the "off resonance" spread between tx and Rx.
It is easier to lock down the tx frequency(fix firmly in coil shell), and then play with the Rx resonance and nulling. You will see what I mean.