Another P.I. link

Dont know if this has been posted before .....

You are not allowed to view links.
Please Register or Login

Linkback:

You are not allowed to view links.
Please Register or Login

http://www.thunting.com/smf/index.php/topic,3797.msg25955.html#msg25955

Hi, I am interested in the PI, but any further info on the Coil?

Linkback:

You are not allowed to view links.
Please Register or Login

http://www.thunting.com/smf/index.php/topic,3797.msg176924.html#msg176924

thanks a lot 

Linkback:

You are not allowed to view links.
Please Register or Login

http://www.thunting.com/smf/index.php/topic,3797.msg177370.html#msg177370

hello , regarding the previously mentioned PI detector designed by mike starcher that appeared in old radio builder,

poss the most straight forward design ive seen so far , so decided to build this detector 2 weeks ago , but with a few mod's, and had a play with it.

clear detection of ali drinks can at 80cm
50p coin at 40cm

over the first weekend of use found £2.40 in change , a 1950 2 shilling piece , lots of ring pulls,tent pegs,cruched cans of varying shape and size , oh , what a wonderfull hobby !

coil used was threaded into 15mm "poly plumb" plastic heating pipe , 30cm dia , 25 turns of maplins solid pvc covered 0.6mm hook up wire , then wrapped tight , no shield , ran out of space , no worries , no interferance , little reaction to ground , salt water sand , or my hand , so , i was lucky there.

right then , main mods , 1st op amp changed to a TL071 and fed straight from R9 , done away with c10 coupling , no point , only creates instability. and no offset req.

U1 , 4536 , very good idea , i like but at £3 per chip , ditched it for a simple 555 with the pin 7 / 6/2 join diode to get duty down to useable level.

4066 output to 2 caps , simplified to one cap sample and hold.

U5A and U5B , replaced with a pair of trusty 741's , yes the legand is still a usefull chip.

reduced the gain on U5B , R13 now 47K done away with R14 , no point , doesnt increace sensitivity , only changes the detection from progresive (nice) to all or nothing.

M1 and circuitry removed , just my choice , sound is all i wanted , didn't want to stare at a meter , rather look where i was going.

OC1 , gone just feed output from U5B into base of an npn , collector to + emitter to 555 pin 7 via a 4.7K

output direct to phones via a 10k pot.

VR1 now an lm317t set to 16V.

mosfet an IRF840.

U2 4538b now a cd4098 , only as they were cheaper , and i had lots , they work well.

at first i used an lm741 as the first op-amp , amazed that it was fast enough , later changed to a TL071 as it recovered better from rail out so i could sample sooner for more conductive metal.

i know ive moded the design a bit , but every respect to mike starcher , brilliant , simple design , works well .

just thought i'd share the experiance.







Linkback:

You are not allowed to view links.
Please Register or Login

http://www.thunting.com/smf/index.php/topic,3797.msg190439.html#msg190439

hello 6666 ,

yes 555 with the diode in to lower the duty is a fine replacement.

i kept the single voltage supply ,
although raised to 16V , more power to mosfet , bigger "window" for the signal to play in ,
but a little close to the absolute limit of a 555 and the 4066 of 18Vmax ,

quite a alot op amp's can amplify the few mV's above the top rail , some upto 0.5V above top rail ,
and were only interested in the last 10 mV or so ,
so using the cap for a simple voltage shift is not required ,
and dual voltages arn't required either.

first tried a 741 , which worked fine , i was absolutley amazed as it's only 1V/us slew ,
but slow to recover from rail out at the begining so sample time had to be later ,
the TL071 works fine , recovers sooner , 13V/us slew , so can sample earlier.

just wanted to make a good basic design even simpler ,
can't have gone far wrong as performance (distance) is on par with other designs,

it doesn't have the same features as some other designs , it's only a single sample machine , but ,
just goes to show , more wires and more chips don't equal more depth.

my next project i'm working on is a only a dual sample machine , but with FE discrimination / indication ,
designing it to be "user simple" , no threshold to keep adjusting , automatic and a manual zero , sort of getting there
using a circuit simulator , then breadboard it , followed by build.

might work , may not , who knows ,have to try and see.

Linkback:

You are not allowed to view links.
Please Register or Login

http://www.thunting.com/smf/index.php/topic,3797.msg190522.html#msg190522