spiral pi coil

[godigit1]

On SRF and sampling if I may ask?
If you have a SRF wave form that is completly clear of the coil in less than 10us could you sample that coil at ten Us without a damping resistor as the SRF oscillation is gone before first sample?
Here is a wave form of one of the spirals Ive wound. pretty much clear at 5us
This coil is 30 awg
no space 308Uh coil.
5.2 ohms
26 wraps.
33.664 Pf with 30 ince leads.



Looks like I can.
Ive got my test detector up and running finally. I hooked up the 390Uh test coil.
My damping resistor set up goes to 1645 ohms.
For this coil it seems not enough. It acts a little different than my other coils on damping.
I get a nice Knee as you damp the coil but I can keep cranking up the damping without it looking overdamped It just gets faster.

I have to make a set up with more resistance to check this out.

Heres the kicker When I removed the resistor set up to measure damping value I forgot to put it in and the next night I fired up the MPP and started waveing my nickle in front and I was like man its hitting better tonight for some reason.
Then looking at my board i noticed No damping resistor installed.
Ive done this before but usually the detector does not work due to the large amount of ringing at the 0 volt line very unstable.
On this coil the ringing is at shut off and almost gone before the sample window.


Here is a shot of the coil with no damping resistor.


Has anyone else experianced this ???
Thanks,

[godigit1]

[QUOTE=Elliot;260528]Hi godigit1, I had an idea today after reading your post and reading other feedback from my posts. May i suggest using a little bit of copper foil tape in your shield near the cable gland area - say 1" or 2cm long, and then use either nickel shield or carbon paint (or graphite spray) for the rest of the coil shielding - ensure you over-spray the copper area a bit (say by 1cm) to ensure an electrical connection? That way, you can get the best of both worlds - you'll have copper to solder your shield drain wire to, and you'll have the benefits of the remaining shielding material![/QUOTE

Hi Elliot, Thanks for the input I had not thhought as copper tape as a attachment for the shield wire. It may be the way to go.
I was thinking of using a small spray of nickle like a dime in diameter and seeing if I can solder to that then use carbon for the shield.
I also found this product that looks interesting.

https://shop.evilmadscientist.com/productsmenu/831

Its metal impregnated versus a solid sheet. Double sided so you could attach to shield then wire then coat with a little more carbon.??
Just thinking.
Also thinking I can get away with imcomplete coil coverage by maybe using a fractal antenna design painted inside the coil cover for a ground shield just maybe.
Thanks,

[godigit1]

For all who are measuring coil self resonance, keep this mental modal in mind. The oscillations of the coil discharge are based on the coil inductance and the total capacitance as seen by the coil.

Keep this mental model in mind.

If a coil has no capacitance, which is impossible, but good for a mental model, the fly-back pulse would discharge with no oscillations. Then, RX sampling could immediately occur. However, in real life science, the capacitance comes from:
1. Coil turn to turn parasitic capacitance
2. Coax cable from circuit board to coil capacitance
3. Coil to shield capacitance
4. TX circuit capacitance which is mostly MOSFET COSS capacitance that can be reduced by adding a series capacitor between the MOSFET and the coil.

Before the RX circuit can be turned on, the TX oscillations need to drop to near zero to allow the RX circuit to detect the target signal and not the end of the TX signal decay or any residual eddy currents in the coil wire itself.

The value of the damping resistor is directly related to the total capacitance seen by the coil to quickly damp the oscillations. Higher capacitance means lower damping resistor values; and lower capacitance means higher damping resistor values and potentially faster sampling.

Measuring the self resonance is a good way to build the fastest coil possible, meaning the earliest potential sampling.

How to reduce capacitance?

Reduce inter-winding turn to turn capacitance by the wire size, insulation dielectric constant, thickness, and winding technique.
Reduce coli to shield capacitance by using a shield with less area than a solid shield or space farther from the coil or both.
Reduce coax capacitance by using the lowest capacitance shield wire or shortest wire.
Add a capacitor in series with the coil and MOSFET output to reduce COSS.

Faster sampling means detecting smaller and lower Time Constants (TC) targets.


I hope this helps?

Joseph J. Rogowski

Hi Mr Rogowski,
Thanks for your reply. I appreciate the input and love the way you use visualization or the thought process. Most people do not express the value of thought and imagining the results versus just looking at the math. My problem has been that my imagination was uninformed. But Im getting the picture more and more thanks to all the input from members like your self.

Your fast coil article was my first go to when I started to learn hobby electronics. It lead to many experiments and one interesting shielding method.
That was i wound a bundle inside a braided copper shield I literally threaded each wrap through the sheild.

On SRF and sampling if I may ask?
If you have a SRF wave form that is completly clear of the coil in less than 10us could you sample that coil at ten Us without a damping resistor as the SRF oscillation is gone before first sample?
Here is a wave form of one of the spirals Ive wound. pretty much clear at 5us
This coil is 30 awg
no space 308Uh coil.
5.2 ohms
26 wraps.
33.664 Pf with 30 ince leads.



Best Regards.

[godigit1]

are your spiral coils flat or are you using a segmented former ? and what method are you using to space the wires ? very curious.

Hi 6666,

Ive been using double sided tape to make the coils.
Just use a compace to make the inner diameter you want and tape around it with double sided tape. If you hang it over the I.D a 1/4" it helps when you peal the coil off the paper.
I use a bic bright liner highlighter pen as a burnishing tool to press the wire to the double sided tape. You dont even need a form to start just draw the wire around the Id. circle on the paper then stack the next wraps against it.
One the first wrap is on its easy. Ive gotten pretty fast I can make a no space coil in 45 minutes ready for foam. Spaced coils take twice as long.
For the spaced coils Im simple using a couple lenghts of wire each about 3 wraps long and I apply them between wrpas and use them for backing. then you can leap frog the back wire each time. On the two wire space I used 4 leap frog wires.

If you try this use the scotch perminent double sided tape. It has the best adhesive and peals right off the foam afler forming.
Another thing is do it on a warm substrate to keep the adhesive gummy. I use a peice of sheetrock.
I Changed foam from my original locktight gaps and cracks to locktight door and window The door and window stays more true when curing and expands less.

Once you have the coil made I cut windows in the paper and tape it securly to the sheet rock spay it with foam then cut it to size.

Once the foam is cured flip the coil over and compase I.D and O.D to the back of the paper. Use a exacto and cut a full blades length deep avoiding the leads of coarse.
At this point all I do is use a hot wire cutter to slice 3/8" slice through the top and the coil just peels out.
Another slice of foam sandwiched on top and you have a coil you can package.

I tried eyeballing the first few wraps on this coil before i started using a space wire you can see some wobblyness there but the rest stays pretty consistant.
The black spacer is silicon flex tape I use to insulate the - wire where it passes over the coil . It has a suposed 8000 volt rating.

[Elliot]

Thanks for your response I will bare the wires close to the coils and recheck before I start target testing.

I intend to try nickel shield and carbon paint as shielding. Nickle will only be used sparingly where Im connecting the shielding and coil halves.

Hi godigit1, I had an idea today after reading your post and reading other feedback from my posts. May i suggest using a little bit of copper foil tape in your shield near the cable gland area - say 1" or 2cm long, and then use either nickel shield or carbon paint (or graphite spray) for the rest of the coil shielding - ensure you over-spray the copper area a bit (say by 1cm) to ensure an electrical connection? That way, you can get the best of both worlds - you'll have copper to solder your shield drain wire to, and you'll have the benefits of the remaining shielding material!

[Elliot]

I've been using (Q=pi*cycles to decay to 37% of start amplitude). Tried your formula(Q=5*cycles to decay to 50% of start amplitude)with my spice simulation. Looks like multiplier should be closer to 4.3*cycles to 50% of start amplitude. I'm rounding .368 to .37. Are you rounding or am I doing something wrong? Don't have a signal generator to try your method to compare with mine. Wondering if you could try my method to compare with yours. Clip scope probe to ground lead making a loop. Lay loop on test coil and move test coil near a operating PI coil to excite resonance. Measure decay frequency with scope.

As far as the shorted scope probe idea goes (i.e. connecting the probe clip to the earth lead), my scope lead wont reach to the isolated (metal free zone) where I test my coils. It may work, but I like to use a square wave input signal, as a square wave contains all frequencies, not just the single PI pulse TX frequency, so I'm guaranteed to hit the primary resonance frequency that way.

[Elliot]

Where did you obtain such thin copper tape ?

Search Coil Faraday Shield: (15mmx50m) Copper Foil Tape – Green centre spool

https://fr.aliexpress.com/item/25M-double-sided-Conductive-EMI-Shielding-Copper-Foil-Tape5MM-6MM-25MM-Adhesive-Barrier/32802248980.html

[Elliot]

I've been using (Q=pi*cycles to decay to 37% of start amplitude). Tried your formula(Q=5*cycles to decay to 50% of start amplitude)with my spice simulation. Looks like multiplier should be closer to 4.3*cycles to 50% of start amplitude. I'm rounding .368 to .37. Are you rounding or am I doing something wrong? Don't have a signal generator to try your method to compare with mine. Wondering if you could try my method to compare with yours. Clip scope probe to ground lead making a loop. Lay loop on test coil and move test coil near a operating PI coil to excite resonance. Measure decay frequency with scope.

Yes, I rounded to 5. As Altra points out, a more precise multiplier is 4.53, but for this work, that degree of precision is not necessary in my view.

[6666]

Where did you obtain such thin copper tape ?

[Elliot]

foil shield is killing all sensitivity. i see nothing of excellent. just he splitted the coil from foil thats right and did not fill by epoxy also right.

I get your point, but the article was far more than just about shielding. On this subject, however, if the copper foil shield killed the sensitivity, the coil wouldn't work at all. I agree that any shield will attenuate the sensitivity (by definition, as eddy currents will always exist, and some shields are "better" than others - agreed), but I have to say that the copper foil I used in my example for beginners (noting that its not the be-all and end-all of shielding types) actually works pretty well. The copper tape I used is just 4/100ths of a mm thick (0.0015748 of an inch, which is hardly "fat") - and that includes the thickness of the protective top layer and adhesive backing! It detects small gold and small sized samples of single leaf/layer aluminium foil, so it's not all bad. The other issue that doesn't get discussed too often, is that the longevity of the connection between the shield drain wire and the shield is also very important. Soldering a wire onto a copper shield will last a very long time (I always clean that joint of flux). Conversely, many coils that use sprayed graphite, die due to loss of connectivity to the drain wire, which is often just simply taped onto the shield. Wrapping a wire around the graphite shield may last longer than tape, but it also adds more metal near the coil. When I build my coils, I want them to last the distance, and not prematurely fail. I guess there's pros and cons for all options. This is what makes our hobby so much fun!

[bbsailor]

For all who are measuring coil self resonance, keep this mental modal in mind. The oscillations of the coil discharge are based on the coil inductance and the total capacitance as seen by the coil.

Keep this mental model in mind.

If a coil has no capacitance, which is impossible, but good for a mental model, the fly-back pulse would discharge with no oscillations. Then, RX sampling could immediately occur. However, in real life science, the capacitance comes from:
1. Coil turn to turn parasitic capacitance
2. Coax cable from circuit board to coil capacitance
3. Coil to shield capacitance
4. TX circuit capacitance which is mostly MOSFET COSS capacitance that can be reduced by adding a series capacitor between the MOSFET and the coil.

Before the RX circuit can be turned on, the TX oscillations need to drop to near zero to allow the RX circuit to detect the target signal and not the end of the TX signal decay or any residual eddy currents in the coil wire itself.

The value of the damping resistor is directly related to the total capacitance seen by the coil to quickly damp the oscillations. Higher capacitance means lower damping resistor values; and lower capacitance means higher damping resistor values and potentially faster sampling.

Measuring the self resonance is a good way to build the fastest coil possible, meaning the earliest potential sampling.

How to reduce capacitance?

Reduce inter-winding turn to turn capacitance by the wire size, insulation dielectric constant, thickness, and winding technique.
Reduce coli to shield capacitance by using a shield with less area than a solid shield or space farther from the coil or both.
Reduce coax capacitance by using the lowest capacitance shield wire or shortest wire.
Add a capacitor in series with the coil and MOSFET output to reduce COSS.

Faster sampling means detecting smaller and lower Time Constants (TC) targets.


I hope this helps?

Joseph J. Rogowski

[6666]

godigit1

are your spiral coils flat or are you using a segmented former ? and what method are you using to space the wires ? very curious.

[godigit1]

Think your coils have lead wires explains the difference. My coil lead has more capacitance than my spiral coils. Think yours might also.

Thanks for your response I will bare the wires close to the coils and recheck before I start target testing.

The coil that tests the best I plan on packaging up to play with shielding and cable capacitance.
If I use the coil wire as leads I plan to make a custom cable by pulling apart a copper shielded microphone cord and re-stuffing it with my leads.
I picked up some ptfe tubing in 14 to 26 awg I.D. and some braided sleaving. I think I can make a assymbly to slide back into the shielded cord.
that being said the finished coil with added shield capacitance should finish up very low.
If I use Elliots figure for a copper shield its 15Pf.
If i Add 5 pf to Elliots shield just for example the worst case out of the 5 coils will be around 58Pf when finished.
I dont intend to twist the leads as they will be inside a copper shielded cable so no added capacitance there.
I intend to try nickel shield and carbon paint as shielding. Nickle will only be used sparringly where Im connecting the shielding and coil halves.


Been reading and rereading about Q and bandwidth And circut impediance. I see how to figure proper Q now but wondering why im not able to get the calculators to come up with the same #s. I played with a bandwidth calculator and was able to get resonant matched to my coil but I had to add a 100meg impediance to get the #s to match.
I was able to find this article breaks things down into a little more understandble language for the electrically challenged like me. LOL
It also has some spice circuts for simulation if anyone is interested
.https://www.ibiblio.org/kuphaldt/ele...s/AC/AC_6.html

Best Regards

[green]

Hi Green,
All the test coils Ive made have leads.
I think we are calcing capacitance the same way. Ive double checked the calculator Im using to specs from some of your coils. But maybe my SRF is wrong?
This is the calculator Ive been using.

http://www.1728.org/resfreq.htm

Im wondering if the less capacitance of your coils is due to larger wire diameter? I have a oval spiral I did some time back with 26 awg I dont think I brought it in my on the road box. Ill have to check. I never tested it for Srf or capacitance. Curious Now!
Thanks,

Think your coils have lead wires explains the difference. My coil lead has more capacitance than my spiral coils. Think yours might also.

[godigit1]

On a normal PI, pulsing frequency or "PPS" has nothing to do with the Q or SRF. The srf and Q are about how fast a coil can release the stored energy (current). The higher the SRF the narrower the flyback period the sooner you can sample.

So you are not going to find a pulsing frequency for a certain Q coil

Hi Altra,
I was thinking kinda in terms of a antenna or signal reception.
Meaning if a coil had a q of 500 what frequencys would it be most useful for.
I guess what im trying to find is what q is best for a particular op amp bandwidth Or if it even matters to try to optimize that.

Thanks