PI Coils Resistance // Inductance

[Juno]

So, if I've well understand...
For PI coil the first more crucial is resistance. Inductance plays a second roll...
But, for one resistance, is there any inductance-limit to not go over/under ?
Is there any diameter of insulated copper wire you like better to use ?

Thanks in advance.

[simeon [email protected]]

Re: PI Coils Resistance // Inductance

a way to vary the coil resistance is to use a pnp transisotor on each + rail
and buffer it with a 100 or 200R accross c e this way you can "turn up & down" the resistance with a ramp of v+

usng a cap this can be changed to a nice level by 4 pots or a graphic eq type arrangemement using a switched cap filter chip {dsp design by microchip using filter cad free app } and feedback correction applied for a nice set of coils that will multi plex

[simeon [email protected]]

Re: PI Coils Resistance // Inductance

i see your points on stuff also


"
So, if I've well understand...
For PI coil the first more crucial is resistance. Inductance plays a second roll...
But, for one resistance"


yes i agree
but....



, is there any


inductance-limit to not go over/under ?
the drive is important any coil of wire will do with some resistance


the composition and metal is very relevent to sensitivities and directionality

an iron anval would work if you donk it hard enough

Is there any diameter of insulated copper wire

{try to use enamled for smaller coils and for realy big squares use 1 mm or 3 core and connect over cores brown blue, blue earth

spare ... blue and earth then is a coil }

that you can fit inside plumbers tubing for sinks to make fold up large coils

[Eric Foster]

Re: PI Coils Resistance // Inductance

Hi Juno,

From my experience the inductance is the more critical parameter. The received signal is proportional to T squared, and so is the inductance. The coil resistance just changes the final current and hence signal, in proportion. There are several interelating factors. The coil and its associated resistances have a time constant, which is important both for switch on and switch off of the TX current. Some manufacturers use a very low resistance coil giving a long time constant. The TX current is switched off well before it has reached its final value in its inductive growth. This gives a sawtooth type of current waveform. I often use a high resistance coil and drive circuit so that the current rapidly reaches its final value and then stays constant until the switch off. Both systems work in spite of a 10:1 difference in TX/Coil resistance. However the coil inductances are much the same.

Higher inductances mean larger back emf's and longer switch off times, so it depends what you are looking for to some degree. You can beneficially use large inductances and high currents if you are looking for large targets with long time constants. Small, low conductivity items need low inductances and smaller currents to cut the field off and sample quickly.

It's always best to use stranded wire, with Litz being the ultimate, to avoid eddy currents being generated in the wire cross section. These eddy currents will limit the small object sensitivity as they have the effect of additional damping. In some applications I use solid wire, but only 0.4mm diameter maximum.

Eric.

[Nicko]

Re: PI Coils Resistance // Inductance

Eric,

I've been experimenting with a critically dampled 120uH, 1.3 ohm coil (amongst others), a 12V high current source and an assortment of FETs, both high & low-side switched, and both P&N type.

As you say, if the coil does not saturate, then you will never get the maximum energy out of it. However, by using low impedence FET gate drivers you can turn on and off the FETs *very* quickly. I am seeing 550V+ back EMF on my test rig at the moment, with the first usable signals about 3.5uS after the turn-off (i.e. when the clipping diodes stop conducting). With this setup, the coil does not saturate until its been energised for more than 80uS, but knowing that the current growth is inverse exponential, you can integrate under the curve to get very precise control over the stored energy, either in absolute terms or as a function of the maximum possible for the coil/damping/FET RDSon that is in place. This can be either done by standard PWM, or by monitoring the current, using a fast integrator then using a fast comparitor to trigger the turn-off. By looking at the first differential of the current, you will now when the coil is saturating anyway (so you don't need to know too much about the actual coil in place). I'm playing with this exact setup at the moment (its 2:30 in the morning and I'll get into trouble if I don't go to bed soon ;-).

Next step is to sweep the transmit power and tie that in with the time constant, distance & size of a detected target...

bed time,

Nicko

[simeon webber]

Re: PI Coils Resistance // Inductance

i had thought of getting away from the whole coil thing and using instead a low 1.4 coil to drive a loop of microbore tube

the results were good using just a loop { like a sausage} to the side like a sieth situlation and in practice
shaped like a sieth this loop had a dc shift of of 1.77 ohms after i added also a little coil for runner of the pinger in series it was 102 ohms so the rx was the loop... resonant at a vhf freq except the pinger as a 100 ohm load... in series
{small 5cm coil on a bobin1" thick can have metal corealso shaped core}
this gave a 2 meter depth using the detector like a sieth action { but was unpradictable only!!!! showing marked discrimination qualitys though between aluminum and copper and the pinger 100T coil as tx ... every other pulse..25% mark space... and sample 100us good for gold finding !like sailing up the field!!!!


this work'd giving 50% more depth than just a round coil {like the loop at a fair {unfair } ground}

[Eric Foster]

Re: PI Coils Resistance // Inductance

Hi Nicko,

Interesting experiments. Let us know how you get on. Are you letting the Fet avalanche, or do you have a snubbing arrangement?

Eric.

[Nicko]

Re: PI Coils Resistance // Inductance

Hi Eric,

I'm using snubbers - a 1.5KE400CA + 1.5KE47CA, giving approx 450V cutoff - total Vbrmax about 470V, Vclamp about 610V - yes, I know thats a bit high !

As I'm rather pro surface mount designs, I'm looking at a number of D2-PAK N HEXFETs, such as the IRFS9N60A (600V. 9A, 0.75ohm) - cheap, and with low RDSon, so no extra heat sink required - ordered some from Farnell today... I can't find any decent high voltage SMD TVSs, so am open to ideas for replacements for the 1.5KExxCAs...

D2-PAK is a nice package - it allows the larger die size, and therefore lower RDSon i.e. higher current with no heat sink...

I'm aiming to make this unit very small & light...

Nicko

[Juno]

Re : Thanks for the answers (N/T)

Thanks for the answers