New spin on a old friend. 3Dss holllow core wind.

[surfdetector]

With my analog scope and x10 probe I measure SRF to be around 820 kHz without coax, which indicates a capacitance about 105 pF. The self-shielding properties of this design are evident though - on 10 mV / div the noise difference is noticeable on the scope when switching the leads. I have yet to finish the mechanical construction of a detector to test this coil in action. And the construction of the coil too of course - especially the mounting and a protective layer.

What method did you use to ring your coil for measurement? Did you utilize a function generator?

Cheers

[Cazavor]

Won't the Aluminium tape that you taped the pipe ends with act as a target? Regards, Marty

The aluminium tape was only there to keep the sand in the pipe during bending to prevent pipe collapse. I removed the tape and poured out the sand afterwards (well, most of it, I can still hear some grains in there if I shake the coil violently ). I didn't mention this step to keep the text shorter. Afterwards the pipe ends were connected with a small section of 14 mm round wooden stick press fit inside the pipe and further secured with toothpicks through the holes.

What method did you use to ring your coil for measurement? Did you utilize a function generator?

Cheers

I hooked the coil to a 50 MHz analog scope, which admittedly is a bit off calibration, but I programmed an Arduino to generate a 1 MHz PWM signal and used that as a reference to calibrate the scope timebase. I also verified the setting somewhat by using a tone generator on a phone at 10-20 kHz and measuring the output. I'm using a 100 MHz x10 probe. Then I put my PI pinpointer next to the coil to excite it and watched the ringing on the scope. I counted the squares the best I could on the display to get the ringing waveform period, around 1.22 us. Then calculated 1E6 us / 1.22 us = 819.7 kHz SRF.

I don't have a function generator yet.

[surfdetector]

The aluminium tape was only there to keep the sand in the pipe during bending to prevent pipe collapse. I removed the tape and poured out the sand afterwards (well, most of it, I can still hear some grains in there if I shake the coil violently ). I didn't mention this step to keep the text shorter. Afterwards the pipe ends were connected with a small section of 14 mm round wooden stick press fit inside the pipe and further secured with toothpicks through the holes.


I hooked the coil to a 50 MHz analog scope, which admittedly is a bit off calibration, but I programmed an Arduino to generate a 1 MHz PWM signal and used that as a reference to calibrate the scope timebase. I also verified the setting somewhat by using a tone generator on a phone at 10-20 kHz and measuring the output. I'm using a 100 MHz x10 probe. Then I put my PI pinpointer next to the coil to excite it and watched the ringing on the scope. I counted the squares the best I could on the display to get the ringing waveform period, around 1.22 us. Then calculated 1E6 us / 1.22 us = 819.7 kHz SRF.

I don't have a function generator yet.

Very cool, I am fascinated with the process. It keeps the mind working. I used my Barracuda as seen in this video to excite the coil to be measured:
https://youtu.be/fD57KGaGZBM

Cheers

[Cazavor]

Very cool, I am fascinated with the process. It keeps the mind working. I used my Barracuda as seen in this video to excite the coil to be measured:
https://youtu.be/fD57KGaGZBM

Cheers

Informative video about measuring SRF and calculating the damping resistor. The process seems to be more precise with a digital scope. For finding rough damping resistor value this method is definitely good enough.

[MartyJ1963]

Thanks Cazavor,,,The coil looks great, hope it's up to your expectations when you finally connect it to a Detector. The coil you made should be really fast. Regards, Marty

[6666]

Cazavor, interesting geometry I hope it works for you

[godigit1]

Hi Cazavor,
nice job on the coil it looks pretty good. I cant tell if on the winding you skipped back a form or forward a slot as would be the standard 3dss.
On this coil I used both forms all slots and skipped to the other form every 5 wraps.
Im thinking that your capacitance is high due to the form diameter. never made it to SRF testing.
I used a 5/16 form.
Will be curious to see how we vary.
After experimenting with wire spacing Ive come to the conclusion too much is bad lowers inductance with no real capacitance benefit after a one wire space. Same goes for form diameter.
That being said with a smaller diameter form it would take less wraps to get the inductance needed and the less wraps in turn would lower the capacitance.

This is good timing as I had just ordered a new wire to revisit this coil. I found some 26 guage 30 strand tinned silicon coated wire I want to try.
I never got to finish this coil my intent was to remove all tooth picks after foaming the coil but you saw what happened to the litz.
this time that wont happen.
Looking forward to your final implementation to see how it works for you.

[Cazavor]

Hi godigit1, it is hard to explain the transitions but as I completed the 5th turn I skipped 1 slot back and bent the wire around the next toothpick, changing the direction, but still remaining on the same form. As I finished the 10th turn, I just bent the wire around the next toohpick, thereby changing the direction and moving to another form. Then repeat on 15th and 20th turn respectively and so on. Essentially this resulted in 10+10+10+5 turn segments of normal 3DSS where every other segment is offset by 1/74th of the circle, increasing the spacing between parallel wires that would otherwise lie on top of each other with only 1 wire thickness between them. This method should theoretically have at least 3 wire thicknesses between stacked parallel wires at the cost of bringing parallel wires in the neighbouring slots a little closer (because more slots are used).

The form thickness came out to be a little bigger than I would've liked yes. The scale appeared deceptive to me on the pictures shown in the first post. If you used a 5/16 inch as in 8 mm form then mine is quite a bit larger indeed. This definitely explains the lesser inductance. Where did you get such a form?

About capacitance, I am not really sure. I think I increased it a little when I wrapped the coil with tape, compressing the outer layers. But it wasn't dramatically lower before either from what I could tell based on my then uncalibrated oscilloscope measurement. Thinking back, 90 pF maybe.

If I had to do it again I'd use a smaller form and try foam instead of tape. Or a hard shell. Somewhere here I read that foam doesn't slow the coil down noticeably unlike epoxy. Would make sense given its low density and air pockets.

I'll also be looking forward to hearing about your results. For reference the 26 gauge wire I'm using is 7-strand tinned copper.

[surfdetector]

What method will you use to attach the clevis to the coil?

Cheers

[godigit1]

Hi Cazavor,

The form I,m using is a poly tubing the type you would get for a refrigerator water line.
It actually mics out to 6.5 MM.
Its semi transparent so you can just lay your form over the template and mark the drill holes.
I join the tubing by melting and sticking together then shave off the high spots and shrink wrap the joint.
This material can be shaped however you want after you mark the form.

Here is a pic of the form without holes and of some DD coils as a example.


As for capacitance I was hoping this from would reduce it a little over the standard 3dss coils.
Thinking that the two forms even though the wire slots are closer there is less wire crossover.
On the standard form all the wires stack tightly into slots on top of each other.
On the round form the wires curve gracefully around and no what ill call pinch points .

Thinking all in all this form is reducing required wire to get the inductance we want causing the extra benefits of less resistance and capacitance. I hope, I could be wrong .

[Cazavor]

godigit1 thanks for the info, you are a true 3dss fan, even making DD coils from it! I too like how the round forms avoids pinch points.

surfdetector, I'll be using pipe clamps together with a U shaped form I bent from ABS sheet with heat gun. Everything secured by nylon bolts. Doesn't look that good but hopefully will get the job done for now:

[surfdetector]

surfdetector, I'll be using pipe clamps together with a U shaped form I bent from ABS sheet with heat gun. Everything secured by nylon bolts. Doesn't look that good but hopefully will get the job done for now:
[ATTACH]50826[/ATTACH]

I like the idea rather ingenious.
Cheers

[godigit1]

Hi godigit1, it is hard to explain the transitions but as I completed the 5th turn I skipped 1 slot back and bent the wire around the next toothpick, changing the direction, but still remaining on the same form. As I finished the 10th turn, I just bent the wire around the next toohpick, thereby changing the direction and moving to another form. Then repeat on 15th and 20th turn respectively and so on. Essentially this resulted in 10+10+10+5 turn segments of normal 3DSS where every other segment is offset by 1/74th of the circle, increasing the spacing between parallel wires that would otherwise lie on top of each other with only 1 wire thickness between them. This method should theoretically have at least 3 wire thicknesses between stacked parallel wires at the cost of bringing parallel wires in the neighbouring slots a little closer (because more slots are used).

The form thickness came out to be a little bigger than I would've liked yes. The scale appeared deceptive to me on the pictures shown in the first post. If you used a 5/16 inch as in 8 mm form then mine is quite a bit larger indeed. This definitely explains the lesser inductance. Where did you get such a form?

About capacitance, I am not really sure. I think I increased it a little when I wrapped the coil with tape, compressing the outer layers. But it wasn't dramatically lower before either from what I could tell based on my then uncalibrated oscilloscope measurement. Thinking back, 90 pF maybe.

If I had to do it again I'd use a smaller form and try foam instead of tape. Or a hard shell. Somewhere here I read that foam doesn't slow the coil down noticeably unlike epoxy. Would make sense given its low density and air pockets.

I'll also be looking forward to hearing about your results. For reference the 26 gauge wire I'm using is 7-strand tinned copper.

Hi Cazavor,

I finished winding the coil, I'm not totally sure of the inductance, I'm having a problem with my inductance meter the switch is wearing out but it looks like 270uH.

I went 30 wraps, by the calculator I should have had 380uH. I switched the wrap up a little this time. I went 5 switched sides same form 5 then back one slot and kept the same direction for 5 wraps then repeated. you can see there is still plenty wire spacing about 3 to 4 wire widths almost 4mm .
Weird about the inductance's we have yours seems closer to calculated.



Its a 9 inch form 26awg 30 strand silicon insulated wire 600v.
finished diameter 3/4 inch about 18mm.

270uh?
2.3 ohm
Srf tested by two methods 1.563Mhz.
Q At 14 cycles =70
Capacitance=38.4pf

I too like the way you did your clevas, I had to blow it up to see you had folded it.
How did you get the folds to line up so perfectly?
Have you had a chance to fire up your coil?

Be Well

[Sean_Goddard]

If the were me, I'd dip the whole thing in shellac or epoxy to hold everything in place, let it cure fully. Like they did with old fashioned transformers, you know, the ones with the lovely "sweet" electrical smell.

[Cazavor]

Well done with the coil. Looks very good and SRF is also way up there. Silicone wire has a lower dielectric constant than PVC, which may be one of the factors. My wrap ended up 25-27 mm outer diameter.

If you're using a regular coil calculator then I think it makes sense for 3DSS wrap to have a lower inductance for the same amount of turns, as the wires are spaced further apart and are not completely aligned with each other, cancelling some of the magnetic flux. 270 uH is likely correct, I got a similar value when I measured 30 turns of my coil during winding.

For the clevis I clamped the plastic sheet between a piece of wood and a metal bar with two screws and folded it around the edges as necessary. Its not that perfect either, one half is a good few mm higher than the other. It is also easy to overdo it with the heat when working with ABS, I charred one spot slightly. The plastic sheet is cut from a license plate holder.

I haven't got the chance to do a proper test with the coil yet, as I have been having some problems with my Hammerhead II type detector. But I think I pinned down the issue today - intermittent coil connection between the BNC socket and plug. I didn't suspect that at all since those parts were brand new. Soon.

If the were me, I'd dip the whole thing in shellac or epoxy to hold everything in place, let it cure fully. Like they did with old fashioned transformers, you know, the ones with the lovely "sweet" electrical smell.

I find some type of old electronics smell quite pleasant but I'm not sure if potting 3dss windings is such a good idea, as it increases the capacitance and is advised against by some people (see 1st page). Then again I haven't seen any actual test results, so maybe it can be an alright compromise between speed and rigidity.