PI5 from ITMD

[geoscash1]

Has anyone built the PI5?

I made a go of it but didn't do so well but I always build on perf or punch board.. Does someone have a pcb drawn? not sure why they didn't add one in the book
like the first 4. For whatever reason I cannot seem to get it drawn right and I SUCK at trying to use any of the software to do it..

If anyone has built it, how well does it work and what coil specs did you use. I guessed at using a 350'ish uH coil and have to assume that would be fine
for what it is.

Did you use the 5532 or swap to the 5534?

Power consumption, what do you get on a set of batts (8 AA's)?

[Qiaozhi]

Has anyone built the PI5?

I designed the original circuit many years ago (2006) when I was experimenting with the idea of removing the sampling gates. The basic concept is to "mask out" the flyback pulse using a combination of NAND gates and a monostable. The TX runs at a pulse rate that is in the audible range, and the signal breaks through whenever a metal target gets near the coil. It was only ever intended for detecting coin-sized targets.

I made a go of it but didn't do so well but I always build on perf or punch board.. Does someone have a pcb drawn? not sure why they didn't add one in the book
like the first 4. For whatever reason I cannot seem to get it drawn right and I SUCK at trying to use any of the software to do it..

The original circuit was also built on stripboard (Veroboard). In fact (after some searching) I've just rediscovered the detector, which I built into an old Whites Beachcomber enclosure. If I can also rediscover what happened to the search coil and the battery pack, I'll fire it up and see what happens.

If anyone has built it, how well does it work and what coil specs did you use. I guessed at using a 350'ish uH coil and have to assume that would be fine
for what it is.

The design shown in ITMD is a version that was modified by Carl to make it more consistent with the other four PI designs. In the original unit, Q1 = TIP31C, Q2 = BC461, and IC1a = CA3140.
IC1b did not exist, as there were two 12V batteries (one for the positive supply, and the other for the negative).

There are also a number of other modifications made by Carl to maintain consistency, but I'm confident that performance was not degraded, otherwise it would never had made it into the final print.

According to the original circuit diagram, the coil was 2.9mH (very high for a PI) and the damping resistor was 1k. Clearly any ringing from the flyback was less important using this concept. I can post the original circuit later, for comparison, if anyone is interested.

[Chet]

Hi Qiaozhi
" The basic concept is to "mask out" the flyback pulse using a combination of NAND gates and a monostable. "
I am very interested in this, please post the original circuit when you have the time.
Thank you,

[Qiaozhi]

Here's the original circuit:

Unfortunately I've not had time to rediscover the coil and power pack yet.

[6666]

Thanks Q

[Chet]

Qiaozhi
Thank you for the original circuit. It looks as if Carl was true and consistent with your basic design. It would probably take more precision to have it operate with a fast 300 uh coil.

[geoscash1]

Thanks Qiaozhi, your reply is much appreciated... I like the circuit and it looks like it can be something to build/expand on, just the idea (" The basic concept is to "mask out" the flyback pulse using a combination of NAND gates and a monostable. ") was interesting to me.

In your first reply you mentioned "It was only ever intended for detecting coin-sized targets" and that is something I think would be good to expand on, if it could be made to detect only coin size targets, could it be modified to only detect silver/gold objects or non-ferrous/ferrous of similar size.. I don't know, just thinkin outloud... I want to build it correctly and mess with it some more but am having a time getting it laid out properly...lol..

Qiaozhi
Thank you for the original circuit. It looks as if Carl was true and consistent with your basic design. It would probably take more precision to have it operate with a fast 300 uh coil.

I had wondered the same when I first started messing with it and thought that was what the issue may be when I couldn't get my original working. I would think it wouldn't be to tough to get a fast coil working with it but don't know that for sure.

[Qiaozhi]

OK guys ... after 8 years of languishing in a box of parts, the original PI-5 circuit has been connected to a +/-24V supply ... and it worked first time.

Considering that this was basically an experiment which was not developed further, it's actually not that bad. As you would expect with a 2.9mH coil, zero amplification after the pre-amp, and only one channel, there are some limitations. The 6" mono coil is the original one from Whites. It gets a Victorian penny at 4", and a Cola can at 9". Both foil and pull tabs are undetectable because of the large inductance.

This would be an interesting concept to take further. If anyone is interested, I would suggest using the circuit presented in ITMD, rather than the original one I posted earlier. There is also the need for a method of eliminating the Earth's magnetic field. Perhaps this design would be more suited to a PI-based pinpointer.

[Chet]

Qiaozhi
Good work. I might try something based on this design in the future. For the present I am having fun experimenting on the bench with the Minipulse Plus PI. For reference do you know what US coin is similar to a Victorian? Thank you for the update.

[Qiaozhi]

Qiaozhi
Good work. I might try something based on this design in the future. For the present I am having fun experimenting on the bench with the Minipulse Plus PI. For reference do you know what US coin is similar to a Victorian? Thank you for the update.

Not sure, but I'm think someone said a Silver Dollar (not to be confused with a Geotech member of that name ) had a similar response to the Victorian penny. I'll do the comparison sometime tomorrow ... need to get some sleep now as it's getting late here.

[geoscash1]

Thanks for all the info Qiaozhi, much appreciated..

[geoscash1]

CD4011 pinout

One thing I did notice on the old (original not in book) pdf drawing Qiaozhi posted is the pinout on the CD4011, according to my datasheet, the numbering on the original is not correct for IC3 (uc) a,b,c, and d. Shouldn't it be 1&2 in, 3 out, 5&6 in, 4 out, 8&9 in, 10 out and 12&13 in and 11 out? I may be wrong but that's what my sheet is showing for pinout..
CD4011.pdf

[Qiaozhi]

One thing I did notice on the old (original not in book) pdf drawing Qiaozhi posted is the pinout on the CD4011, according to my datasheet, the numbering on the original is not correct for IC3 (uc) a,b,c, and d. Shouldn't it be 1&2 in, 3 out, 5&6 in, 4 out, 8&9 in, 10 out and 12&13 in and 11 out? I may be wrong but that's what my sheet is showing for pinout..
[ATTACH]31207[/ATTACH]

It appears that you are correct. However, the original circuit does work, so I can only assume that I spotted this issue (back in 2006) but never updated the schematic. Damn ... and I've only just put the whole thing back in its enclosure.

[geoscash1]

556 instead of 555's

It appears that you are correct. However, the original circuit does work, so I can only assume that I spotted this issue (back in 2006) but never updated the schematic. Damn ... and I've only just put the whole thing back in its enclosure.

LOL.... Hey if it works then its all good, it must be connected right or I don't believe it would work with the pinout the other way.....


I'm switching to an LM556 dual timer to eliminate the two 555's... and really thinking may end up using a pic to replace that, I think a 12f629,12F675 or a 683 will do the trick.
Maybe a 16F628 to replace the 555/556 clock and have the ability to add the GB/Disc channels with other options as well... This could be a very scalable little project...

Also, normally near the coil, C1 is usually a 1000uF (showing in original schematic), but this circuit in the ITMD is using a 100uF, is that correct or is it maybe a typo or would it really matter which is used?

[Qiaozhi]

Also, normally near the coil, C1 is usually a 1000uF (showing in original schematic), but this circuit in the ITMD is using a 100uF, is that correct or is it maybe a typo or would it really matter which is used?

It probably wouldn't hurt to increase the capacitance of C1 to 1000uF. In fact there's quite a few improvements that could be made, such as the use of linear regulators to create more stable power supplies.