complete project

[Tinkerer]

I been thinking Tinkeror
You mention starting modular approach, im all for that, infact ive done that many times in the past with different projects .
So if we are going down that road which is a good idea what we need is a open jig, ie plinth with front and rear upright panels made as light as possible that finaly can have a cover fit over it to do various outside tests.
The large one off heat sink can be mounted on one of the retaining panels or free standing depending on choice of design.
My idea is make the jig as explained using styrene prepared with shielding which will be explain to all later, its so easy to cut and glue and of course once all built its easy to mount the top cover etc, you end up with a custom housing to suite requirements which is very cheap and light as a feather.
So after saying that maybe the jig and power supply comes first.
Have you any idea of dimentions that going to be required?

Regards

OK, let's see what we want on the front panel.

ON - OFF switch
Audio volume control pot.
The LCD of the UNIPI
The encoder push button of the UNIPI
The other 2 push buttons of the UNIPI
SAT control pot

Rear panel

Heat sinks. Actually we might make the rear panel THE heat sink.

Cable outlets

There are 3 cables, twisted, going to the TX coil, maybe we should shield them too.
There are 3 cables, twisted and shielded, going to the RX coil.
Bulkhead connectors? Cable connectors?

Specific cables for each?

The Battery
We could simplify things for a start, using direct 12V battery output, unregulated. Of course, this means that the performance of the detector will vary as the battery voltage changes from full to low.
It will also cut down the original goal of depth rating to about 30%.

The battery could be a simple lead-gel cell. A bit heavy, but we need weight for counterbalancing the large coil.

Tinkerer

[satdaveuk]

Using a jig

I been thinking Tinkeror
You mention starting modular approach, im all for that, infact ive done that many times in the past with different projects .
So if we are going down that road which is a good idea what we need is a open jig, ie plinth with front and rear upright panels made as light as possible that finaly can have a cover fit over it to do various outside tests.
The large one off heat sink can be mounted on one of the retaining panels or free standing depending on choice of design.
My idea is make the jig as explained using styrene prepared with shielding which will be explain to all later, its so easy to cut and glue and of course once all built its easy to mount the top cover etc, you end up with a custom housing to suite requirements which is very cheap and light as a feather.
So after saying that maybe the jig and power supply comes first.
Have you any idea of dimentions that going to be required?

Regards

[satdaveuk]

Audo stage

This is a good audio amp low current that can be cranked up to 2 watts into a 8 ohm speaker with a good frequency responce used in the Ivonics TGSL, not to be confused with earlier and later circuits its based round the common 1 off BD140 and 2 off BC157.
This circuit as whole works better when using hand picked and tested components of the exact specification as listed like many projects ive built and tested, so where possible thats the way to go here for us to get optium performance.

Regards

[Tinkerer]

yes i agree, and if some want something a lttle different im sure it will be able to be adapted in someway, and as you said earlier to someone on here this project will never be finished as in final which is what Geotech is all about.
Maybe a good idea to use version numbers for those who get involved later on.
Like your test setup tinker you have been busy.
So now where do we go from here?

Regards

Maybe a good way to proceed, would be to tackle the project at the different block simultaneously.
We could look at the different options, discuss and suggest and come to a consensus of the version we want to execute. This will depend more on yourself than on me, because we are designing the detector for your specific purpose, namely, to walk the fields in the UK.

Things are very different here and I have to work with the parts at hand. It is very costly for me to get parts, so I buy only once or twice per year.

So, when I suggest a way to do things, it is a guess from far away, from your angle things may look different.

To get started:

The coil is one of the most important parts. It would be good to get going on that. I have built quite a few coils and suggest a very specific coil for the project, based on my experience.
This coil is slightly different from other coils I have built, with the intention of improving on it's predecessor that worked fine.
If this is OK with you, I will send you exact dimensions and data tomorrow.

Another part to get busy with, is the PSU. A good, efficient and noiseless powersupply can make all the difference. The PSU I used on the Tinkerers IB-PI_V1, way synchronized with the TX, but wasted 50% of the power.
I prefer to apply the power on the TX instead of wasting it.
Like all things electronic, new and better parts become available every day, but it is getting more and more difficult to find through hole parts.
So, lets see what we can find in parts and design the PSU around that.

Still another design block to look at, is the audio output. The UNIPI gives a clicking output. I like the clickers, it was my preferred audio, using a "bone conductor" for the underwater detector. However, on land, it is different. OK maybe with a headphone, but, how do we amplify a clicker so it can be heard a few feet away?

These are some subjects for suggestions.

I am all ears.

Tinkerer

[satdaveuk]

yes i agree, and if some want something a lttle different im sure it will be able to be adapted in someway, and as you said earlier to someone on here this project will never be finished as in final which is what Geotech is all about.
Maybe a good idea to use version numbers for those who get involved later on.
Like your test setup tinker you have been busy.
So now where do we go from here?

Regards

[Tinkerer]

Here is a peep at the TX board and the UNIPI board. I think we could make a single board with the PSU, the UNIPI and the TX board.

The 4 Mosfets must be tied to the same heat-sink. As it is, it gets barely warm.

Tinkerer

[Tinkerer]

The old pre47 sixpence is 50% silver pre1920 92.5% silver after that there was no silver so anywhere around that time should be similar to a hammered silver coin.
Far as where to start, the coil is as good as place as anywhere although would be nice to see a full circuit diagram with pcb to have a peep at to see where we are going.
Regards

The size, or surface area presented to the coil plays an important role in the detectability of the targets.
To search deep, we want a large coil. Large coils are inherently less sensitive to small targets.
To get the best compromise, we need to balance the minimum target size with the maximum depth.

A circuit diagram with PCB.

I design my PCB's for bench testing. This is not exactly the same as putting the design into a box, to be used in the field. For example, the pots on a box are not the same as the trimpots on the PCB. Changes need to be made.

The same happens with SMD opamps and such. Many modern advanced technology opamps are simply not manufactured in through hole shape.
For example, the ADG5434, Analog switch, that I am using at present, only comes in 20 lead TSSOP package. Very tiny and a pain to solder and delicate to make the thin traces on the PCB. The same goes for many excellent opamps. I use the AD8626 in SOT8, does the job real well. Is it available in DIP8? I don't know. Need to look it up.

Is this the only way? No, there are good parts available in DIP packages. We can very well design the boards to use only through hole parts. But the board design is different.

The parts need to be searched and researched. A lot of reading and comparing of data-sheets.

Then comes the box.

Above, we talk about a maximum power consumption of 30W. 30W in Kilo calories that need to be dissipated. No problem in a cool climate. But what about that hot day in the sun?

I once bought a top class magnetometer from a Canadian company. 30,000 US$. Supposedly good to be used anywhere in the world. When I tried to run it in a tropical country, it run 10 minutes and then flipped out for overheating.
The advice from the manufacturer was to use a fan to blow on it or air conditioning. Yeah, do that in an open boat out on the ocean.

So, yes, we need to design the detector and the box, so that they fit together, not only in size, also in power dissipation.

Ah, about size. There are some fair size capacitors. The TX is running off them. We recycle the TX power, running the current from the capacitor bank through the coil and back again. The battery only supplies the resistive and switching losses.
Capacitors come in different shapes. With no space restrictions, we can use any kind. Inside a box, we need to consider the space. Once we decide on the shape of the capacitor, we can see how we place it on the PCB and how large the box needs to be for the PCB.

Should I go on? No, I think we have all understood that if we want to see a finished design, we all have to help to get it done.

Of course, "we all" is still another problem. "We all" have to agree on each aspect of the full design.

Can we agree on that?

Tinkerer

[Tinkerer]

Maybe it would be a good idea to assign your designs some version numbers so that people can hang on to something.

Good idea.

So this new design goes under the name TINKERERS_TEM_IB-PI.
It is a detector designed for searching the open fields, for relics and coins that lie deeper than other detectors can locate.

Tinkerer

[satdaveuk]

What are the dimensions of an old six penny? At present, PI detectors do not indicate any specific metal. Only ferrous or non ferrous.

As far a sensitivity, the 1 meter square coil has no difficulties at all, of locating a coca-cola can at 1.5m depth.
However, I think you are more interested in small hammered silver coins, so I suggest we make the first coil only 1m x 0.5m, rectangular. This will be for medium depth.

Depth comes with a price. Power and size of the coil.
With this size coil, we need to look at the ergonomics right from the start.

So why don't we start with building the coil?

Tinkerer

The old pre47 sixpence is 50% silver pre1920 92.5% silver after that there was no silver so anywhere around that time should be similar to a hammered silver coin.
Far as where to start, the coil is as good as place as anywhere although would be nice to see a full circuit diagram with pcb to have a peep at to see where we are going.
Regards

[Tinkerer]

Id say they are very good perameters to go for, you say one meter depth that will discriminate i presume that would be something the size of a jerry can if buried in typical soil with medium iron content, would be good if we can hit a old pre 1947 silver sixpenny bit at around 12" which would show as silver, or a medium size gold ring at anything above 9" would be nice.

Regards

What are the dimensions of an old six penny? At present, PI detectors do not indicate any specific metal. Only ferrous or non ferrous.

As far a sensitivity, the 1 meter square coil has no difficulties at all, of locating a coca-cola can at 1.5m depth.
However, I think you are more interested in small hammered silver coins, so I suggest we make the first coil only 1m x 0.5m, rectangular. This will be for medium depth.

Depth comes with a price. Power and size of the coil.
With this size coil, we need to look at the ergonomics right from the start.

So why don't we start with building the coil?

Tinkerer

[satdaveuk]

Id say they are very good perameters to go for, you say one meter depth that will discriminate i presume that would be something the size of a jerry can if buried in typical soil with medium iron content, would be good if we can hit a old pre 1947 silver sixpenny bit at around 12" which would show as silver, or a medium size gold ring at anything above 9" would be nice.

Regards

[Tinkerer]

TINKERERS_TEM_IB-PI

Hi Tinkerer
Firstly would like to say a big thank you to yourself, Moodz and others that have gave input to this project both in research and design, not a easy task by far as im sure everyone here and within the manufacturing industrie would agree.
The goals we are after is the same as they have been since metal detecting began, which is deeper with more reliable discrimination.
Air testing is one thing, which is it proves a detector is working and indeed discriminating in the air, but thats really where it ends.
Another thread that was started here by yourself, correct me if im wrong, was headed the perfect ground balance, include a good ground filtering circuit, if we Crack that one then everything else should quickly come into place because thats where the problem lies as you all well know.
Ive been following your threads since they began and cant wait to construct your latest designes to try in the real world with the soils we have here in the uk.
Theres not such a problem with beach detecting quoting a good example would be the Whites DFX although working well on pasture land compared to many other makes and models, soon as you hit beach wether it be wet sand or dry it works like a totaly different machine far as depth and discrimination, its near perfect as ive seen, which of course goes back to ground filtering which inturn effects discrimination and depth.
In my opinion with the typical designes of PI/IB machines theres little hope for improvement the way they stand and thats why im so interested in your new design which could end up been the way to go for many years to come, however it remains to be seen, but of course we need to think positive
You asked what im after in a metal detector, well if I want to greedy the answer is through hole construction which makes it easy for those of us who construct are own PCBs, parts that are not to exspensive but take into account we are not cutting corners which will effect the performance, hex files that can be flashed to the chips for people like me with little knowledge of programing, nothing too heavy on the battery life.

My main detecting time is spent on plough and pasture relic hunting, normally only disciminate out iron, the rest I dig.

I think after what you said Tinkerer that after tieing up a few loose ends your design is ready and certainly worth constructing.
Is this equipment going to be made modular or on one main PCB board, or are you still thinking about it?
All the best and thanks again

Thanks for the feedback.

Let's see if we can agree on some of the parameters.
I think we can achieve the following:

Maximum power consumption..................................... 30 Watt
Maximum design detection depth................................. 1 meter
Maximum pulse repetition rate per second, PPS...............10,000
Maximum coil diameter.......................................... ......1 meter
FE recognition at full depth
Capable of working in adverse ground conditions
Motion type detector
Fast response

Now you add the minimums and other parameters that are important to you.

FOR EVERYBODY INTERESTED IN THE PROJECT:

This is an OPEN SOURCE PROJECT you are free to build and experiment with it. If you want to use the information for commercial use, please contact me, I will be glad to come to an easy arrangement.

We are starting this project now.
Give your input now, it will be difficult to make changes once many people are already involved.

Tinkerer

[Davor]

Maybe it would be a good idea to assign your designs some version numbers so that people can hang on to something.

[Tinkerer]

When it is final? And then, just a theory and practical design no. Search for nothing ... Just read on ...

Smity,

you ask when the design is going to be final.
Well, here is the honest answer: NEVER
I will always try to improve it, make it more perfect. There is no end to that. EVER.

Tinkerer

[satdaveuk]

Satdaveuk,

I have no use for a finished metal detector myself, but I enjoy tinkering with the design of metal detectors, always trying to improve the performance.

The Tinkerers_V1 is about complete, but it's parts are scattered over several threads, we would have to look for the last version of each part and join them together. It worked fine in the lab, but I did not make any real field tests.

I learned a lot during the design of the Tinkerers_V1. Now I am applying this knowledge in a more recent and advanced design.
In the meantime, MOODZ has also released the UNIPI. The UNIPI is an advanced, PI, programmed dspPIC, I think MOODZ is also giving away for free, the programming code.

Anyway, MOODZ was so kind of sending me an UNIPI and that is what I use now for the timing.

So, please tell me about your design goals:

What do you want to use the detector for? The design will need to be different for relic hunting than for nugget hunting.

There seems to be a continuous interest in a very deep reaching detector. To reach really deep, the coil needs to be large. Such detectors often use coils of 1m x 1m square. I prefer round myself, but the difference is not important.

With traditional PI detectors the large diameter coils have little sensitivity to small targets, like a hammered silver coil of 2cm diameter.

We could put together a detector that can overcome this limit. This might be quite a popular success.

A simple, yet advanced design.

Through hole, easy to find non expensive parts.

What else would be on your wish list?

Should we make a poll to find out what the popular wishes might be?

Tinkerer

Hi Tinkerer
Firstly would like to say a big thank you to yourself, Moodz and others that have gave input to this project both in research and design, not a easy task by far as im sure everyone here and within the manufacturing industrie would agree.
The goals we are after is the same as they have been since metal detecting began, which is deeper with more reliable discrimination.
Air testing is one thing, which is it proves a detector is working and indeed discriminating in the air, but thats really where it ends.
Another thread that was started here by yourself, correct me if im wrong, was headed the perfect ground balance, include a good ground filtering circuit, if we Crack that one then everything else should quickly come into place because thats where the problem lies as you all well know.
Ive been following your threads since they began and cant wait to construct your latest designes to try in the real world with the soils we have here in the uk.
Theres not such a problem with beach detecting quoting a good example would be the Whites DFX although working well on pasture land compared to many other makes and models, soon as you hit beach wether it be wet sand or dry it works like a totaly different machine far as depth and discrimination, its near perfect as ive seen, which of course goes back to ground filtering which inturn effects discrimination and depth.
In my opinion with the typical designes of PI/IB machines theres little hope for improvement the way they stand and thats why im so interested in your new design which could end up been the way to go for many years to come, however it remains to be seen, but of course we need to think positive
You asked what im after in a metal detector, well if I want to greedy the answer is through hole construction which makes it easy for those of us who construct are own PCBs, parts that are not to exspensive but take into account we are not cutting corners which will effect the performance, hex files that can be flashed to the chips for people like me with little knowledge of programing, nothing too heavy on the battery life.

My main detecting time is spent on plough and pasture relic hunting, normally only disciminate out iron, the rest I dig.

I think after what you said Tinkerer that after tieing up a few loose ends your design is ready and certainly worth constructing.
Is this equipment going to be made modular or on one main PCB board, or are you still thinking about it?
All the best and thanks again