Getting ready to build the MPP

[Chet]

Hi kellogs
Silverdog sold the Chance PI boards sometime ago. I don't know if they are available now.
Chet

[Achillion]

--------------------------------------
What is your self resonant frequency with 36" of feed line? Depending on what feed line is used I would expect the SRF to drop to 1.050 mHz or so which is right about where all of my 300 uh to 335 uh spider coils end up with feeds. However once they are in circuit on my CHANCE PI they really end up at about an SRF of about 850 to 950 kHz.

Regarding the Infineon IPD50R800 it looks like the device has a COSS of about 400pf at 12 volts (DSS) but they show it has 23pf @ 100 volts Drain to Source. Is that better than the IRF 740?

Regards,

Dan

hi Dan,
The COSS of the IRF740 is about 1400pf @ 10V so specs are definitely better for the IPD50R800. Also it allows me to build smaller as I will attempt to build the front end inside the coil housing to further eliminate capacitance of the feedline. Will give it a go and see what comes out.

[baum7154]

hi Dan,
The COSS of the IRF740 is about 1400pf @ 10V so specs are definitely better for the IPD50R800. Also it allows me to build smaller as I will attempt to build the front end inside the coil housing to further eliminate capacitance of the feedline. Will give it a go and see what comes out.

----------------------------------------

After looking closely at the IRF740 and the IPD50R800 datasheets the COSS of the IRF740 at 10v is about 350pf while the COSS of the IPD50R800 is about 400pf at 10 volts. At 0 volts the IRF740 is about 1400 pf and the IPD50R800 is about 800 pf. The Vishay diode you mention has half the capacitance of an HER208 and it will help.

I often see comments about 'In the Coil Electronics' but I don't under stand how these can be implemented without the metals they contain becoming targets themselves. If the detector will be used to search for small gold it will need to sample below 10us and it seems the copper, solder, tin in the coil mounted electronics will impede its ability to do so.

If the objective is to keep capacitance low I do it by using a continuation of the 600 volt PTFE coil wire to create a 34" long twisted pair. It is very difficult to find coax that has lower capacitance than this configuration and in addition coax will require solder joints in the coil and this should be avoided in my opinion.

Just a thought.

Regards,

Dan

[green]

----------------------------------------

After looking closely at the IRF740 and the IPD50R800 datasheets the COSS of the IRF740 at 10v is about 350pf while the COSS of the IPD50R800 is about 400pf at 10 volts. At 0 volts the IRF740 is about 1400 pf and the IPD50R800 is about 800 pf. The Vishay diode you mention has half the capacitance of an HER208 and it will help.

I often see comments about 'In the Coil Electronics' but I don't under stand how these can be implemented without the metals they contain becoming targets themselves. If the detector will be used to search for small gold it will need to sample below 10us and it seems the copper, solder, tin in the coil mounted electronics will impede its ability to do so.

If the objective is to keep capacitance low I do it by using a continuation of the 600 volt PTFE coil wire to create a 34" long twisted pair. It is very difficult to find coax that has lower capacitance than this configuration and in addition coax will require solder joints in the coil and this should be avoided in my opinion.

Just a thought.

Regards,

Dan

Hi Dan
I posted some solder targets awhile back. The solder targets have a short TC. I don't think a solder connection would effect detecting small nuggets. Try soldering two short pieces of your coil wire together and see if you can detect it. If you can tape it to your coil and see if it effects the detection distance of your gold targets. I'm with you, if you put enough stuff near the coil it will.

[Tinkerer]

Mr. Green,
your tests are very interesting and accurate. Yours are the the most precise information about target TC that are publicly available.

There is one other subject about which it is difficult to find accurate information: The relation between target surface area and TX coil surface area.
A general rule of thumb says that a target with a surface area of 1% of the surface area of the TX coil, is about the smallest target that can be detected at the center of the coil.
Did you do any tests on this subject?

[green]

Mr. Green,
your tests are very interesting and accurate. Yours are the the most precise information about target TC that are publicly available.

There is one other subject about which it is difficult to find accurate information: The relation between target surface area and TX coil surface area.
A general rule of thumb says that a target with a surface area of 1% of the surface area of the TX coil, is about the smallest target that can be detected at the center of the coil.
Did you do any tests on this subject?

Not yet. I'm redoing my test rig right now. Looks like it might be an interesting test.

[Chet]

I think that as long as the solder blob or other object is stationary and not very large there should be little effect. A small distortion in the detection pattern near the coil may occur depending on the size of the object. Even small ferrous objects should be ok. The eddy current loss should be very small. See attached diagrams. The iron object would be similar to a nail near the coil.

Chet

[baum7154]

Hi Dan
I posted some solder targets awhile back. The solder targets have a short TC. I don't think a solder connection would effect detecting small nuggets. Try soldering two short pieces of your coil wire together and see if you can detect it. If you can tape it to your coil and see if it effects the detection distance of your gold targets. I'm with you, if you put enough stuff near the coil it will.

-----------------------------------------------

Hi Green,

I just ran a test using two 3/4" pieces of scrap 24 AWG 600 volt PTFE coil wire soldered together using 1/4" of their length for the clean and minimal solder joint. This target is in the shape of a 'T' with the 1/4" solder junction as the midpoint or vertical part of the 'T' and is not a closed loop. My modified CHANCE PI detector set at 8us delay sees this sample repeatedly
at 1 1/4" for first detection and at 7/8" for repeatable full tone lockup. This is using my 4" X 12.5" 3:1 fast spider wound oval coil. This target response/distance is significantly stronger than the response I recently posted for some placer flakes (largest one 0.95 grain) I documented in the CHANCE PI COIL thread. Granted this is just an air test and results may be much reduced over actual soil. As far as taping the sample target to the coil it was not done because the coil would not operate at the 8us delay setting if the sample was any closer than 1 1/4".

Minelabs has a patent I have seen posted in the Forums for a technique that places ferrite around the solder joints in some/many of it's coils to reduce/eliminate coil connection solder responses to the detector. Though this is probably a minor problem overall I think it is still good practice to eliminate all solder possible from the coil field.

Regards,

Dan

[6666]

Interesting results Dan , but in normal MD coil construction how many T connections do you have ?
what is the test result with just a normal series solder joint ?

[baum7154]

Interesting results Dan , but in normal MD coil construction how many T connections do you have ?
what is the test result with just a normal series solder joint ?

------------------------------------------------------------------

Hi 6666,

There is a misunderstanding here as there is no 'T' connection in what I did in this test. The 'T' I described is not formed by a wire intersecting another wire but was formed after soldering 1/4" of two 3/4" long parallel wires together and bending the remaining 1/2" of the two wires soldered together 90 degrees in opposite directions leaving only the soldered stub perpendicular. The 90 degree bends were done to minimize inter wire capacitance.

If you re-read post 153 I questioned how 'in the coil electronics' can be implemented with out negatively affecting coil performance. So in the context of placing electronics in the coil I believe that there is much more opportunity for true 'T' connections to be present in the coil field from the circuit board and it's connections, even though that is not what I was simulating. In the context of simply soldering a coax to a coil you are correct that a true end to end series solder joint would be best and I will do that test too.

I did also test a 1" long scrap piece of the same coil wire (effectively eliminating the solder joint of the first test) and could only get very sporadic detection in very close proximity to the windings and never did get full tone lockup.

Series solder joint test results to follow.

Cheers,

Dan

[baum7154]

Just ran a new target made of 2 each 3/4" coil wire scraps with an in line solder joint of 1/4" of each scrap. The original 'T' shaped sample and the new series joint sample were matched for weight at 2.2 grains each to insure they both had close to the same solder content. The series sample did show less detection distance with 1st repeatable detection at 7/8" and full tone lockup at 3/8". The original target was confirmed at it's prior detection distance with the 1/4" stub kept in a plane parallel with the coil plane. So it appears that the 1/4' solder stub of the original sample makes it more detectable than a smooth in line solder joint for a simple coil to feed connection. While the in line joint is less detectable it is still detectable to the point of full tone lockup.

I am trying to understand how 'in the coil electronics' can be implemented without degrading coil performance and... as simple as it is to eliminate solder connections in a coil field, why do it? Any help on this is appreciated.

Thanks,

Dan

[baum7154]

Mr. Green,
your tests are very interesting and accurate. Yours are the the most precise information about target TC that are publicly available.

There is one other subject about which it is difficult to find accurate information: The relation between target surface area and TX coil surface area.
A general rule of thumb says that a target with a surface area of 1% of the surface area of the TX coil, is about the smallest target that can be detected at the center of the coil.
Did you do any tests on this subject?

-------------------------------------------------------
Hi Tinkerer,
After reading the 'Rule of Thumb' above I did a bit of calculating on the smallest target detectable with my round 8" fast spider coil. The actual diameter to the centerline of the coil is 7.7" and this gives an included area of 46.56 square inches within the coil perimeter.

Using a 1/4" X 1/4" (1/16 of a square inch) square target cut from a 0.004" thick aluminum soda can this target is repeatably detected at the coil center with solid tone lockup at a distance of 2.75" in air presented flat to the coil. The detector used in this test is a modified Chance PI detector with a 2 stage amp in the front end.

The 1/4" X 1/4" target is 46.56/ 1/16 =745 or 1/745th (.134%) of the coil area.

1% of the 46.56 area would be 0.4656 sq. in. and the most similar target I have tested is a .5" X 1" in piece of 0.004 soda can aluminum. This target is easily detected at coil center with full tone lockup at 7" in air.

Target thickness definitely affects detection too so perhaps the rule of thumb only applies to extremely thin sub .001" thick foil targets.

Regards,

Dan

[Tinkerer]

-------------------------------------------------------
Hi Tinkerer,
After reading the 'Rule of Thumb' above I did a bit of calculating on the smallest target detectable with my round 8" fast spider coil. The actual diameter to the centerline of the coil is 7.7" and this gives an included area of 46.56 square inches within the coil perimeter.

Using a 1/4" X 1/4" (1/16 of a square inch) square target cut from a 0.004" thick aluminum soda can this target is repeatably detected at the coil center with solid tone lockup at a distance of 2.75" in air presented flat to the coil. The detector used in this test is a modified Chance PI detector with a 2 stage amp in the front end.

The 1/4" X 1/4" target is 46.56/ 1/16 =745 or 1/745th (.134%) of the coil area.

1% of the 46.56 area would be 0.4656 sq. in. and the most similar target I have tested is a .5" X 1" in piece of 0.004 soda can aluminum. This target is easily detected at coil center with full tone lockup at 7" in air.

Target thickness definitely affects detection too so perhaps the rule of thumb only applies to extremely thin sub .001" thick foil targets.

Regards,

Dan

Dan, thank you for the feedback.
The rule of thumb mentioned above, is at least 10 years old, but seems to be the only information publicly available. This shows that detectors have evolved a lot and that your modified Chance PI is a very good detector.
It would be interesting to do a bit more calculations.
For example: with a 1/4" square target, at 2.75" from the coil, what is the field or flux density inciding on the target? What I mean to say, you have 46.56 square inches of coil area.
What is the peak coil current?
How many turns?
...Amps x ...turns is your field strength.
Divide this field strength by the TX coil area 46.56, is your field density per square inch, divide by 16 is the field incidence on the 1/4" square target.

At the distance of about half coil diameter, the field strength is about 15 less, so the target should have about 15 times more surface area for the same response.

These are just "ball-park" figures off the top of my head, of course there are proper formulas to calculate all this.

Thickness of the target: There is something of a corner response where the target thickness is up to one skin depth, governed by the Skin effect. Thinner targets behave one way, thicker targets behave an other way. This could be the subject of another interesting discussion.

All the best

Tinkerer

[baum7154]

Tinkerer,

Fascinating stuff and it gives me something more to figure out. I believe the coil turns are 32 and resistance is 2.2 ohms. IRF 9640 has 0.5 ohm max Rds, and HER208 series diode has about 0.7 voltage drop. So with coil voltage at 12.0 volts (assuming 12.7 volts to the mosfet) and 2.7 ohms resistance = 4.44 peak amps coil current X 32 turns = 142.22 field strength divided by 46.56 = 3.05 field density per Square Inch divided by 16 = .1909 = field incidence on the 1/4" X 1/4' target…If I did this all correctly!

Thanks,

Dan

[green]

Mr. Green,
your tests are very interesting and accurate. Yours are the the most precise information about target TC that are publicly available.

There is one other subject about which it is difficult to find accurate information: The relation between target surface area and TX coil surface area.
A general rule of thumb says that a target with a surface area of 1% of the surface area of the TX coil, is about the smallest target that can be detected at the center of the coil.
Did you do any tests on this subject?

Was looking at a chart I posted awhile back. The .5in x .5in target has an area of about 1.25 percent of the coil area. Maybe the general rule is for the center of the coil at a distance of the coil diameter.