Parallel windings....Good idea or not?

[Satindas]

Hi,
I'm kinda new to this but I feel I'm learning a lot through my attempts at making decent coils. I've tried a few mono coil builds for an MPP, regular PVC, then enamelled and lastly Teflon which was a bit weak at 186uH (not sure why cos I followed the fast coil design specs to the dot!) and I've just completed an 8" basket coil for a Surf PI 1.2 which is bang on at 248uH,1.8ohms 1.02MHz SRF.

I'd like to try an experiment in winding a parallel wound mono (2 or 3 parallel windings of AWG 26/28 ) for the MPP. I get the feeling that this would reduce overall inter-winding capacitance (like a kind of improvised litz wire) but before I waste time, has anyone tried this or have any info on the possible outcome.

Many thanks in advance

Simon

[Davor]

Not a waste of time. It works.

[reptooyep]

Hi,
Could you make a draw or scheme of what you want to do ?

[Teleno]

Reducing winding capacitance is a noble endeavour but mostly futile for a bunch of reasons.

1. The usual PI designs (Surfmaster, MiniPulse etc.) use slow amplifiers/samplers that can't take advantage of the faster decays.
2. The coil will require a higher damping resistor which increases noise and reduces gain.
3. Most of the coil discharge will occur thorugh the MOSFET's avalanche diode, leaving you with a slower decay than expected, unless you upgrade to a higher voltage MOSFET.
4. If you manage to sample earlier then changes in ground capacitance will cause false signals.
5. If you shield to prevent ground from changing your coil's capacitance then the coil capacitance increases and you're back to square 1.

[Qiaozhi]

1. The usual PI designs (Surfmaster, MiniPulse etc.) use slow amplifiers/samplers that can't take advantage of the faster decays.

Although this statement is true for the Surf PI and the original MiniPulse, it is not true for the MiniPulse Plus. This detector has a 2-stage pre-amp that allows for earlier sampling. One of the real problems with using fast coils on any of these machines though, is that of a ground balance circuit (or lack of).

[Teleno]

Although this statement is true for the Surf PI and the original MiniPulse, it is not true for the MiniPulse Plus. This detector has a 2-stage pre-amp that allows for earlier sampling. One of the real problems with using fast coils on any of these machines though, is that of a ground balance circuit (or lack of).

MiniPulse Plus uses the NE5532 that has a unity gain bandwidth of 10Mhz just like the NE5534 of the Surf PI. What's more, it has two stages rather than one, which slows things furher a bit. I don't see how it can be faster at all.

[green]

Hi,
I'm kinda new to this but I feel I'm learning a lot through my attempts at making decent coils. I've tried a few mono coil builds for an MPP, regular PVC, then enamelled and lastly Teflon which was a bit weak at 186uH (not sure why cos I followed the fast coil design specs to the dot!) and I've just completed an 8" basket coil for a Surf PI 1.2 which is bang on at 248uH,1.8ohms 1.02MHz SRF.

I'd like to try an experiment in winding a parallel wound mono (2 or 3 parallel windings of AWG 26/28 ) for the MPP. I get the feeling that this would reduce overall inter-winding capacitance (like a kind of improvised litz wire) but before I waste time, has anyone tried this or have any info on the possible outcome.

Many thanks in advance

Simon

http://www.geotech1.com/forums/showt...770#post213770

An example in link above. Coils A and B. Twisting two wires lowers the resistance allowing a higher peak current or the same peak current with a lower power supply voltage. Had little effect on SRF with the spider web(flat basket) coil. Don't know what effect it would have for a bundle wound coil.

[Satindas]

Although this statement is true for the Surf PI and the original MiniPulse, it is not true for the MiniPulse Plus. This detector has a 2-stage pre-amp that allows for earlier sampling. One of the real problems with using fast coils on any of these machines though, is that of a ground balance circuit (or lack of).

Thanks Qiaozhi,
If the lack of ground balance is such a problem, is there really any point in making a fast coil for the MPP unless you only plan to use it on white sandy beaches ? ( HA! If I could afford white sandy beaches I probably wouldn't be out treasure hunting !)

Reducing winding capacitance is a noble endeavour but mostly futile for a bunch of reasons.

1. The usual PI designs (Surfmaster, MiniPulse etc.) use slow amplifiers/samplers that can't take advantage of the faster decays.
2. The coil will require a higher damping resistor which increases noise and reduces gain.
3. Most of the coil discharge will occur thorugh the MOSFET's avalanche diode, leaving you with a slower decay than expected, unless you upgrade to a higher voltage MOSFET.
4. If you manage to sample earlier then changes in ground capacitance will cause false signals.
5. If you shield to prevent ground from changing your coil's capacitance then the coil capacitance increases and you're back to square 1.

Thanks Telono for the useful info. Regarding point 5 though, if a shield is to be applied does it not make sense to get the coil IW capacitance as low as possible first?

One of the main reasons I want to try this is that PTFE insulated wire is quite expensive to play around with. I have access to a large number of old 20m VGA cables (PVC) which I could use for this... and it wouldn't cost a penny. I've been pondering over these cables but I know the individual cores are around AWG26 or 28 ( maybe even 30) and 20m isn't gonna be long enough so I thought about parallel windings. I guess even if the result isn't particularly fast I still get to build a coil or two for nowt

[green]

Thanks Qiaozhi,
If the lack of ground balance is such a problem, is there really any point in making a fast coil for the MPP unless you only plan to use it on white sandy beaches ? ( HA! If I could afford white sandy beaches I probably wouldn't be out treasure hunting !)



Thanks Telono for the useful info. Regarding point 5 though, if a shield is to be applied does it not make sense to get the coil IW capacitance as low as possible first?

One of the main reasons I want to try this is that PTFE insulated wire is quite expensive to play around with. I have access to a large number of old 20m VGA cables (PVC) which I could use for this... and it wouldn't cost a penny. I've been pondering over these cables but I know the individual cores are around AWG26 or 28 ( maybe even 30) and 20m isn't gonna be long enough so I thought about parallel windings. I guess even if the result isn't particularly fast I still get to build a coil or two for nowt

Posted reply #7 the same time you posted reply#8. 20meters is long enough for a 200mm diameter spider wire coil.

[green]

MiniPulse Plus uses the NE5532 that has a unity gain bandwidth of 10Mhz just like the NE5534 of the Surf PI. What's more, it has two stages rather than one, which slows things furher a bit. I don't see how it can be faster at all.

Approximate, closed loop bandwidth equals unity gain bandwidth/closed loop gain. One stage with gain of 100, 10 MHZ/100 equals 100 kHz. Two stage with gain of 100, each stage 10MHz/10 equals 1MHz.

[Teleno]

Approximate, closed loop bandwidth equals unity gain bandwidth/closed loop gain. One stage with gain of 100, 10 MHZ/100 equals 100 kHz. Two stage with gain of 100, each stage 10MHz/10 equals 1MHz.

OK I get it. Thanks for the insight!

Noise reduction takes the opposite approach: a stage of gain 5,000 followed by a 0.5 divider is less noisy that one/two stages with total gain of 1,000.

[Davor]

Yes, but negligibly so.

[Qiaozhi]

MiniPulse Plus uses the NE5532 that has a unity gain bandwidth of 10Mhz just like the NE5534 of the Surf PI. What's more, it has two stages rather than one, which slows things furher a bit. I don't see how it can be faster at all.

Approximate, closed loop bandwidth equals unity gain bandwidth/closed loop gain. One stage with gain of 100, 10 MHZ/100 equals 100 kHz. Two stage with gain of 100, each stage 10MHz/10 equals 1MHz.

Also, the 2-stage preamp spends less time railed out than an equivalent single stage. i.e. it's recovery time is better due to increased closed loop bandwidth (as mentioned by green).

[Funfinder]

Parallel or even triple winding could be interesting when using completely different EM-fields directly and always at the same time.
But of course this will goes on costs of inductivity and weight. But some ULF plus 100kHz IB-circuit could be created
with it which works better than those meanwhile existing multi frequency designs.

Of course its also possible sending different frequencies directly at once through one and the same wire.
But not completly different or even diametral changing EM-field structures because it may shortcut the whole transmitter.

[6666]

I have access to a large number of old 20m VGA cables (PVC)

If you just want to mess around with coils, just strip out single wires and solder them in series .