Fast PI coil by Joseph J. Rogowski a good choice for MiniPulse Plus?

[alexdavey]

Is the fast coil tutorial by Joseph J. Rogowski (from the Geotech website) a good match for the Minipulse Plus? I am planning on building it exactly as described in the article. If not do you have a tutorial you would recommend?

[Qiaozhi]

Is the fast coil tutorial by Joseph J. Rogowski (from the Geotech website) a good match for the Minipulse Plus? I am planning on building it exactly as described in the article. If not do you have a tutorial you would recommend?

Yes - Joe's coil spec is an ideal match for the MPP.

[alexdavey]

Great, thank you

[alexdavey]

At the end of the article it talks about adjusting the damping resistance for the hammerhead by tuning the coil. How does this work for the MPP as it only has one input?

[Qiaozhi]

At the end of the article it talks about adjusting the damping resistance for the hammerhead by tuning the coil. How does this work for the MPP as it only has one input?

Are you referring to the part where it says : "... if you plan to use multiple coils on your PI machine ..."?
If so, this does not infer that the Hammerhead has multiple inputs. It does not. This sentence is asking whether you intend to connect different coils to your machine by swapping them over. This is not asking whether you plan to connect more than one coil at the same time.
The MPP has the same coil connection facility as the HH, in that it can use either a mono or an induction balance coil.

[bbsailor]

Yes - Joe's coil spec is an ideal match for the MPP.

George,

The purpose of my article was to provide a practical tutorial about the various things that influence the ability to sample fast... below 10 microseconds (10uS). The article starts out by saying that there are a "few things" the coil builder can do that collectively allow "fast sampling". These few things are all related to basic electronic theory that the Hammerhead demonstrates in most common mono coil PI metal detectors of that era.

Unfortunately, these "few thing" sometimes go beyond the coil making techniques and include the TX pulse width, TX frequency and the frequency adjustment range to allow similar PI machines working close together to not interfere with each other or be operating at a harmonic of the local power line frequency.

The DC resistance of the coil, any resistor in series with the coil to limit current and the coil driver MOSFET R-on resistance all make up the total TX circuit resistance that will define the TX coil time constant Coil inductance divided by the coil TX circuit resistance). If a 300 uH coil has a total TX circuit resistance of 6 ohms, then the coil time constant is 50 uS. The Miniplus Plus has a TX pulse of 58 uS, so the coil should be optimized for this pulse width. Let's assume that the total TX circuit resistance (including MOSFET R-on resistance) is 10 ohms then the time constant (TC) is now 30 us. This means that in one TC the maximum current will rise to about 63 percent of maximum current (E or voltage divided by total TX circuit resistance) and that a longer TX pulse (60 uS total) of another 30 us will rise to about 85 percent of maximum current. The amplitude of the flyback pulse and energy that needs to be damped will get higher with longer TX pulses. This means that the optimum damping resistance is for a TX pulse range that keeps the flyback pulse critically damped for the most early sampling desired. This is why Eric Fosters TX pulse rates and Pulse widths were adjustable to keep about the same energy level in the TX pulse so damping could be optimized over the full range of frequency adjustment. This translates into keeping the Minipulse TX frequency and pulse width optimized for the target type mostly sought with the damping resistance optimized for this setting.

One unfortunate side effect of the IRF740 is the COSS or MOSFET output capacitance that tends to add more capacitance to the TX circuit that supplies additional energy to the flyback circuit the needs to be damped. If you could use a lower COSS MOSFET or put in a series high voltage diode and bleed resistor you could make the coil damp faster with a higher resistance damping resistor to make the PI coil more sensitive to smaller sought after targets with a shorter eddy current decay times (like small gold nuggets or small chain link jewelry). Unfortunately, one size coil, one TX pulse length and one TX frequency range is not good for all targets. These characteristics should be optimized for the targets sought and then the coil parameters (as outlined in my article about fast coils) with TX and RX adjustment parameters optimized for these targets. The easy to understand rule of thumb is to minimize:
1. the coil capacitance turn-to-turn
2. coil-to-shield capacitance
3. MOSFET COSS or use a series diode with bleeder resistor
4.And after all the above: optimize TX frequency, TX pulse, RX sample width for the coil TX circuit parameters with the optimum damping resistor value set for this range of TX PI control settings and TX energy needing to be damped.

Eric Foster recommends that the PI sample delay be about five times faster than the total decay time in the primary targets sought. Otherwise, the eddy currents decay in the small targets before the RX circuit turns on and has the potential to see the target.

It is my hope for the forum PI builders to see how all these things interrelate and that coil winding techniques are but one issue related to the bigger whole. The simple measurement is to measure the coil self resonant frequency at the end of the cable or coax that feeds to the PI coil connector. Make any alteration to make the final coil result have a higher self resonant frequency as this means that there is less capacitance in the circuit and can be potentially damped faster. However, this faster coil only works well on a PI machine that has the potential to use this faster coil. The best fast PI circuit potential comes from using a dual stage (lower gain per stage) op amp that will come out of saturation faster than a high gain single stage op amp.

I hope this helps?

Joseph J. Rogowski

[mschmahl]

Thanks a lot for this explanation and your "fast coil tutorial".
One question I have: What is a bleeder resistor? If I use a MUR460 Diode (see datasheet) do I have to connect it with a extra or special resistor?

[baum7154]

I had this discussion some time ago with DAVOR and the bleeder resistor is meant to help bleed off the COSS capacitive charge of the mosfet. In the case of an IRF9640 fet the COSS is about 375pf and the bleeder resistor for this is in the neighborhood of 1000 ohms to give a 5 Tau value of about 1.8us if my calculation was correct. This bleeder resistor would be placed at the connection between the series diode and the mosfet with the other end of the resistor to ground.

Regards,

Dan

[Qiaozhi]

Hi Joe,

Thanks for your detailed explanation.

As you correctly stated, the MPP has a fixed 58us TX pulse width, but this can be changed to suit anyone's particular detecting requirements. In fact, some constructors have already done that. I don't get the impression that alexdavey is particularly looking for a fast coil, or even intending to search for small gold nuggets, so I think your coil spec will work sufficiently well for initial testing, even if the default MPP settings are not ideal. The original purpose of the MPP was to provide a simple to build project that had a greater chance of producing a working detector than either the Surf-PI or the Baracuda (both of which seem to be fraught with problems for the beginner). After that I was hoping (and that appears to have been the case) that others would use the MPP as an experimental platform, to better understand the PI parameters you mentioned, and to make further tweaks to improve performance. In many ways your fast coil article is quite applicable to the MPP, as it has a two-stage preamp that provides for sub-10us sampling.

[mschmahl]

Thank´s a lot Dan, it becomes clearer now. How can we calculate the bleeding resistance? Or is there a way to see it with the o-scope?

[crane]

Thank´s a lot Dan, it becomes clearer now. How can we calculate the bleeding resistance? Or is there a way to see it with the o-scope?

A bleeder resistor would defeat the diodes purpose because the diode's capacitance decreases with increased voltage on the cathode.

Look at the graph for the MUR 440 and 460 in the pdf file. The capacitance drops to around 7pf at 50v on the cathode.

You definitely don't want a leaky diode in that circuit.

[alexdavey]

Are you referring to the part where it says : "... if you plan to use multiple coils on your PI machine ..."?
If so, this does not infer that the Hammerhead has multiple inputs. It does not. This sentence is asking whether you intend to connect different coils to your machine by swapping them over. This is not asking whether you plan to connect more than one coil at the same time.
The MPP has the same coil connection facility as the HH, in that it can use either a mono or an induction balance coil.

Sorry, i'm referring to this:

The damping resistance should beadjusted experimentally while observingthe output of the first amplifier(Hammerhead IC6) with an oscilloscope.Place a 1200 ohm resistor in parallelwith a 5K ohm pot that has a 1200 ohmlimit resistor in series with the pot. Thisallows you to adjust the damping resistorvalue in the range of 600 to 1K ohms,(R11 in the Hammerhead) to optimizethe coil’s performance on your PImachine.

[baum7154]

A bleeder resistor would defeat the diodes purpose because the diode's capacitance decreases with increased voltage on the cathode.

Look at the graph for the MUR 440 and 460 in the pdf file. The capacitance drops to around 7pf at 50v on the cathode.

You definitely don't want a leaky diode in that circuit.

----------------------------------------------------------------
The diodes capacitance DECREASES with increased REVERSE VOLTAGE on the diode. The stored 12vdc in the mosfet COSS reduces that reverse flyback voltage by 12vdc and once flyback is done it continues to forward bias the diode if it still persists. It is this undesired 12vdc that the bleeder resistor is helping to quickly drain.

Regards,

Dan

[baum7154]

Thank´s a lot Dan, it becomes clearer now. How can we calculate the bleeding resistance? Or is there a way to see it with the o-scope?

-----------------------------------------
How fast do you want to bleed off the mosfet COSS charge? I chose to have it bled off in under 2us. The time constant or Tau of an RC circuit is the time it takes to charge to about 63% of the charge voltage. It is generally considered that the charge voltage will be completed in 5 Tau. If we have a COSS of 375pf multiplied by 1000 ohms we get 0.000000375 second or 1 Tau. Multiply that times 5 Tau intervals and you have 0.000001875 second or 1.875us.

Regards,

Dan

[Qiaozhi]

Sorry, i'm referring to this:

I was commenting on your question: "How does this work for the MPP as it only has one input?"
I'm not sure what you meant by this statement. Whatever is written in the Hammerhead article, regarding selection of the damping resistor value, also applies to the MPP.