Simplest PI detector possible?

[Harp]

In article of Eddy71 Russian PI http://www.geotech1.com/forums/attac...5&d=1358762518
is it possible directly on the output 555 use only one power transistor that drives the coil, with only the necessary resistors around it?
If so, can anyone draw a modified scheme - I try on the testplate all possible combinations with strong NPN and PNP transistors, but is always overloaded currents.
If not, please explanation to learn something
Thanks!

[Davor]

Pin 7 of 555 is an open collector, and with a proper diode to protect it during flyback ... sure, you could drive a coil with a bare 555. It has up to 100mA loading capability, and it is quite fast.

[6666]

Harp
have a look here http://www.geotech1.com/forums/showt...and-no-ferrous
for simple idea.


In the attachment there should be a small 10-22 ohm R to the gate

[porkluvr]

Here is where you see a description for how it works.

According to LTspice, You should swap the two resistors R2 and R3. To get closer to the original PPS and PW, use 1k3 instead of 1k6

[Harp]

thank you, now works!
I have N type MOSFET, 10 ohms on the GATE, inverted resistant R2 and R3.
I can not get detection 2 Euros on the 40cm, not even approximately. Start "crunch" at about 10cm distance, and at a 4cm appears full sound.
I will still try to tune a little bit device with a damping resistance.
The coil of wire diameter 0.4mm, inner diameter of 17cm, inductance 260uH, flat spiral - 2 layers.

[porkluvr]

What coil resistance?? My favorite coil calculator does not have an applet that covers a dual layer spiral coil, so help me out here.

[Harp]

The resistance of the coil is around 2.5-3 ohms, can not accurately be measured with my digital voltmeter so little resistance.
Length is about 13.5m
http://electronbunker.ca/InductanceCalcML.html

[porkluvr]

The resistance of the coil is around 2.5-3 ohms, can not accurately be measured with my digital voltmeter so little resistance.
Length is about 13.5m
http://electronbunker.ca/InductanceCalcML.html

You could try increasing pulse width (PW) by increasing the 1.6k (or 1.3k, whatever you have at R2) to about 3.3k.

That should allow coil current to reach very near its peak attainable value (Vbat divided by the sum of the various series resistances) which I have to assume is around 3.8A.

There are some down-sides to simply increasing PW and I have some other things to suggest, but for now this is all I have time for.

One more thing: I think my final "solution" will be using a 2.4kΩ resistor at R2, and adding a 1Ω (1/2 watt, maybe) series resistance at the coil. I'll be back later with an explanation.

[porkluvr]

An easy way to increase target illumination would be to increase PW. Too short of a tx pulse (for a given power supply voltage and series resistances including the coil) will leave the coil only partially "charged" at the moment that the tx switch is opened. From what I understand, it is best to have PW somewhat "tailored" to supply voltage and coil characteristics, arranged so that coil current has time to -almost- settle to zero di/dt before end of tx pulse (beginning of flyback interval).

If you have the tx pulse too short, a portion of the flyback current will be cancelled by remnants of the charging current. Caveat: as you increase PW, your flyback time will also take more time to settle. Also, wasted power and overheating if you go too far with it.

A two pronged approach for fixing things may be to increase the PW a bit, but also add series resistance in the coil circuit. Series R serves to reduces the maximum value that coil current can rise to. Bare in mind that as you add resistance in the coil circuit intending to lower that maximum attainable coil current, the added series resistance will cause the whole process to take somewhat longer.

I read where Eric Foster suggested that if your coil current reaches about 95% of the coil's max possible value (before switch-off), that is good... but I saw no other hint regarding what an acceptable range of percentage might be. But, understand that at "switch-on", di/dt starts off high and then decreases exponentially with time. Do not shoot yourself in the foot by allowing too-short of a transmit pulse.

I believe (according to LTspice, using an approximation) that your maximum attainable coil current is something around 3.9A. Switch-off (start of flyback) is occurring at about 2.8A, well below maximum. That implies your coil current would not be well settled (instead still changing).

If your coil works as well as I think it may, you might be able to shave a few microseconds off the flyback pulse interval by increasing the damping resistance to 360Ω, but that is something (more) that you would need to determine for yourself.

[porkluvr]

We could talk about time constants and how that relates to fully charging the coil, but in my humble view, the R component of the time constant formula will probably be underestimated, making any results of using the formula less exact than we might like.

There is skin effect resistance at the beginning of the tx pulse, and throughout the duration of the switch-on interval, whatever type switch used will have resistance that very likely does not remain constant.

edit: The way I determined "maximum" coil current in the previous post was to roughly triple the pulse-width, then measure the value of the flattened top of the current waveform. If the waveform had not reached a point of flattening, well then make the pulse wider. No math involved.

And on the subject of math, when I mentioned di/dt, any of you non-math type persons (which unfortunately includes me) should simply substitute the words "rate of change of current". No sweat.

[Harp]

PCB for those who want to try:

[WM6]

PCB for those who want to try:

Very nice PCB Harp, thanks.

What final schematic suit to this PCB?

[Harp]

Thank you!
From the original scheme I just threw out transistor, and resistors were replaced as indicated in a few posts ago, the diode is added, and instead 100n to 555 was placed 1n.
Damping resistor in the original is 220 ohms and 330 ohms here.

[bernte_one]

Hello Harp ,

i build the unit from you layout

but i does NOT work


if you are able please load a video of your unit or some pictures of pcb
something is rally wrong here

it was my second try to get this think working
seems i go stupid ant cant do simplest detectors

[bernte_one]

or would you measure the voltage at some points ic pins bc547 audio stage .....
thx