White's PI3000 re-build

[Octopulse]

Have you checked Q4? Maybe this is faulty too.

Diode test for a PNP transistor.....about 0.45v and 0.55v between emitter and collector.

Q4 might be a problem. I'm not sure what has lead to the R16 burnout as I am keeping to the exact stock specifications except for the voltage pump IC but that shouldn't cause a problem.

Thanks.

[Octopulse]

you have not a pulse signal in TP6 (test point 6). Q3 is constantly open. Q4 is too. R16 burn.

I don't have any test equipment to check that....just my humble MM.

Thanks.

[Lucar]

Hi,

I'm Luca , the author of the posts You resumed from Findmall if I recall; I don't remember anything about that, it was a long time ago, but I still have the PI 3000 somewhere in my house and one thing for sure is that the 950 coil is not the original one. As far as I remember the original coil was not shielded and that was a damned noisy detector. Anyway adding a pic16F did allow for flexible timings programming.

Cheers
Luca

[Carl-NC]

R16 has nothing to do with the charge pump. 3 possibilities:

1. The TX pulse is too wide or the TX timing is not working
2. There is a dead short in the search coil
3. R16 power is rated too low

From the photo R16 does not look like a 1W resistor, more like a 1/2W or maybe even 1/4W. I would need to know the TX pulse timing to figure out what it needs to be. If you replace it, try running the circuit without the search coil attached to see if the TX pulse is working correctly.

[dbanner]

It might be the coil.

The clock is 30kHz. It looks like the PRF is 3kHz? With a pulse width of 30uS. Time to first sample is 30uS after end of Tx pulse. Time to second sample is 270uS after end of Tx pulse. All pulse widths are 30uS in duration.

In the notes on the first post, the original coil was opened and found to have 36 turns @ 11 inch diameter which works out to app. 1200uH.

The 950 coil is around 300uH????.

Put in a higher power rated resistor for R16.

[dbanner]

I see clock inhibit(pin 13) is tied to Vss (set to low) on schematic. This pin must be set to low to allow clock to count normally. If there is a problem with the power supply, you need to check the voltages on the power rails of the PCB as well.

[dbanner]

In the sim I use plus and minus 5V. The actual circuit uses VDD +5 and VSS -3.5

[kt315]

It might be the coil.

The clock is 30kHz. It looks like the PRF is 3kHz? With a pulse width of 30uS. Time to first sample is 30uS after end of Tx pulse. Time to second sample is 270uS after end of Tx pulse. All pulse widths are 30uS in duration.

In the notes on the first post, the original coil was opened and found to have 36 turns @ 11 inch diameter which works out to app. 1200uH.

The 950 coil is around 300uH????.

Put in a higher power rated resistor for R16.

it is maybe second concept of Eric. late concept he did. MAY BE, i do not sure know here fo the 3000. yes, then, 1mH coil, high TX frequency, a resistor in circuit of coil is 30+Ohm. we say '1mH coil PI detector'.

[dbanner]

the original coil is orange and looks different : https://www.youtube.com/watch?v=q951...RlY3RvciB0ZXN0

Yet the schematic for PI1000 shows notes giving 11inches dia. 18 turns for the coil which works out to around 300uH (0.4mm wire)

[kt315]

the original coil is orange and looks different : https://www.youtube.com/watch?v=q951...RlY3RvciB0ZXN0

Eric used orange coil enclosures and orange cable.

[GeoMax]

Hi folks,

the yellow Pulse Diver Coil 950 is an original coil by Whites for the PI3000. The first issues of the PI3000 came with a 3-spoke yellow coil which looked the same as the orange coil from the PI1000 & PI2000 detector. The later versions of the PI3000 came with the Pulse Diver Coil 950. Whites offered also this coil as an upgrade to the 3-spoke yellow coil. The coil specification was the same. The orange coil from the video belongs to the Whites PI1000.

[Octopulse]

R16 has nothing to do with the charge pump. 3 possibilities:

1. The TX pulse is too wide or the TX timing is not working
2. There is a dead short in the search coil
3. R16 power is rated too low

From the photo R16 does not look like a 1W resistor, more like a 1/2W or maybe even 1/4W. I would need to know the TX pulse timing to figure out what it needs to be. If you replace it, try running the circuit without the search coil attached to see if the TX pulse is working correctly.

Thanks Carl.....I got the R16 values from the schematic.Might be wrong on the wattage though. I think maybe Point 2 you mentioned is worth revisiting. I re-wired exactly the same as the original with the shield and centre conductor going to the correct pins on the board. As mentioned earlier, nothing has changed from the original specs. I will have to check my coil connection again and re-solder.
Unfortunately I don't have an oscilloscope.

[Octopulse]

Hi,

I'm Luca , the author of the posts You resumed from Findmall if I recall; I don't remember anything about that, it was a long time ago, but I still have the PI 3000 somewhere in my house and one thing for sure is that the 950 coil is not the original one. As far as I remember the original coil was not shielded and that was a damned noisy detector. Anyway adding a pic16F did allow for flexible timings programming.

Cheers
Luca

The yellow 950 coil is original from White's and is most definitely shielded. Thanks, Tony.

[Octopulse]

It might be the coil.

The clock is 30kHz. It looks like the PRF is 3kHz? With a pulse width of 30uS. Time to first sample is 30uS after end of Tx pulse. Time to second sample is 270uS after end of Tx pulse. All pulse widths are 30uS in duration.

In the notes on the first post, the original coil was opened and found to have 36 turns @ 11 inch diameter which works out to app. 1200uH.

The 950 coil is around 300uH????.

Put in a higher power rated resistor for R16.

What value is ideal.....wattage and resistance?
Although I am still unsure why the original spec R16 is too small....assuming no other problems with the coil connection.

[kt315]

to define the value is easely. remove paint with a scalpel, do accurately. then you must see a point of burn on metallization strip. use zoom glass.
measure one part and second part of intact strip - from a pin to the point edge. add both values and you get whole value.