Eric, PI -With unstable threshold

[Anonymous]

About 10 monhts ago I purchased a new CS -PI detector from a dealer in coloradio, he ordered the CS from a dealer in california, who mailed it to me in florida US.When it arrived here batterys were dead, it was turned on.after replaceing batterys, I checked it out. CS had a very bad pulseing problem,a unstable threshold,poor dept. would only detect a nickel at about 8". I have been useing a detector since 1962, over 40 yr.it was easy to see that CS was some-ones reject and that the dealer in california had resold it to me. Why dident I send it back.No warranty form came with it.the manufacture has a poor reputation of honor-ing its warranty here.if I sent it back to caifornia,I might have lost my money and the detector.CS might be resold again.
With very good help I desided to try to repair it.We replased 709 IC with a 5534 and lowered delay.now CS will detect a nickel at 11"+ air test. But threshold is un stable.
Reason for this post. last week May 16, you posted a old PI front end . I missed part from IC 5534 to threshold.will you repost it from 5534 to includeing all threshold circuit. Thank you.
Frank.

 

[Anonymous]

It would be nice if Eric posted more circuits but he has only ever posted the front end circuit that he recently re-posted and the integrator circuit that comes after it. He hasn't posted what comes after that in the 4 years that I've been visiting his 2 forums. You could have a look at the PI-1 circuit on Carl's Geotechnology website. It has a typical threshold circuit there that will probably be helpful.
good luck,
FJ

 

[Anonymous]

See if this sit helps you: http://www.thunting.com/geotech/....click on projects...also at this sit is a guy in Canada...Gary@chemelec who is a retired electronic wiz and his hobby is this type of problem and helping people...I got this from this forum...scroll back to 03, April and you can read the discussion

 

[Anonymous]

I have just completed building a PI detector that is a composite of a couple different designs. It also exhibits a problem with Threshold stability. In my version I use a 555 as a master oscillator which runs at from 1 to 3 Khz at a 50% duty cycle. I use it to do two things, One is to develop the +5 supply using the familiar diode capacitor circuit. This gives me quite a bit more current capacity on the +5 supply. The output also triggers another 555 which supplies the 100usec pulse for the coil switch as well as the monostables for the sampling pulses. I chose to do it this way because in earlier versions where I did not sync the 555 for the power supply to the transmitter timing I had a big problem with switching transients within the two sampling windows. What I have seen with this circuit is that there is a some instability in the timing pulses. I am not quite sure why this happens. One would expect that the sample pulse timing from the monostables would track any variation in the 555 output but this does not seem to be totally true. So what happens is a small variation in the pulse width of the second 555 results in a shift in the sample delay as well as a shift in the sample pulse width. I am using the two steering diodes to make both sample pulses the same width, but when the shift occurs one gets wider and the other shorter which totally throws off the differential integrator and the subsequent threshold as well. I suspect that the values for the timing components that I have selected may have the 555 or the monostables operating too close to the operational limits for the desired pulse width. My next step will be to change the timing components to try to get better stability. I realize that in most commercial circuits that the timing is controlled by a pic or micro of some sort which is probably the best way to go, and which I probably will change to after brushing up on my software skills some. I just thought I would throw my 2 cents worth into the discussion in case some of the other builders out there may have experienced similar problems and might find it interesting.
Thanks
Russ_NY

 

[Anonymous]

Hi Russ,
I had a problem once with 555's generating the +5V rail with a diode pump. When you get a signal, the receiver and audio circuits pull a bit more current from the +5V. This caused a momentary drop in frequency of the 555, which then caused a change in the sampling widths in the following pulse generator circuits. This was with a circuit that is often used to give a 50% duty cycle, i.e. the timing resistor taken back from pin 3. The cure was to use the other method of timing with two resistors; one from the supply to pin 7, and one between pin 7 and pin 2/6. You can't get an exact 50/50, but this should not matter.
However, I'm not sure if this would happen with the circuit you are using, if you are edge triggering the following monostables.
Eric.