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

[green]

I played with a bleed resistor with LT spice awhile back. Haven't been using a bleed resistor. Used a scope to record amplifier and coil volts with and without the bleed resistor.

[baum7154]

By the way the 'penalty' for the 1000 ohm bleed resistor in the circuit at the mosfet output is 12 ma of drain at 12 volts and is fairly insignificant especially with the short duty cycle of the TX. A 1/2 watt resistor should be adequate.

Dan

[baum7154]

I played with a bleed resistor with LT spice awhile back. Haven't been using a bleed resistor. Used a scope to record amplifier and coil volts with and without the bleed resistor.

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

Not good, can you elaborate? What do you attribute the 140v loss of coil flyback voltage to? Was the resistor placed at the junction of the mosfet out and the series diode? What was the DC voltage at the mosfet out? So the scope shots are actual results and not LT Spice? Also looks like critical damping was negatively affected did you reset the damping?

Dan

[green]

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

Not good, can you elaborate? What do you attribute the 140v loss of coil flyback voltage to? Was the resistor placed at the junction of the mosfet out and the series diode? What was the DC voltage at the mosfet out? So the scope shots are actual results and not LT Spice?

Dan

The resistor is at the diode fet junction and common. The volts is across the coil. They are scope results I took this morning. The bleeder resistor is in parallel with Rd until the coil volts decays to .6 volts. 1k parallel with 1k, Rd=500 ohms above .6 volts. Decay sees fet capacitance I think. Missed your question on volts at mosfet out , didn't record but I can.

[baum7154]

The resistor is at the diode fet junction and common. The volts is across the coil. They are scope results I took this morning. The bleeder resistor is in parallel with Rd until the coil volts decays to .6 volts. 1k parallel with 1k, Rd=500 ohms above .6 volts. Decay sees fet capacitance I think. Missed your question on volts at mosfet out , didn't record but I can.

--------------------------------------------------------------
But in flyback the diode should be reverse biased, isolating the 1K I think. What would happen if the Bleeder was increased to 2K or 3k?


After drawing this out on paper it is apparent that the Bleeder creates a path to ground behind the diode and allows the diode to be forward biased in flyback too. Never mind.


Dan

[mschmahl]

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

Thank´s a lot Dan! I think all under 5 us is good. Will test some parts.

[green]

Volts across coil and mosfet diode junction. The mosfet is a STP11NK40Z not a IRF740. I have two coil circuits and wasn't paying attention, probably doesn't matter.

[baum7154]

--------------------------------------------------------------
But in flyback the diode should be reverse biased, isolating the 1K I think. What would happen if the Bleeder was increased to 2K or 3k?


After drawing this out on paper it is apparent that the Bleeder creates a path to ground behind the diode and allows the diode to be forward biased in flyback too. Never mind.


Dan

---------------------
On CHANCE PI
Perhaps a diode in series with the Bleeder with the cathode end at ground.

[crane]

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

Not sure where you get the stored 12v DC from.

The mosfet is turned on during the pulse and therefore has no "stored 12vdc in the mosfet COSS". The mosfet then tuns off and the voltage on the drain rises rapidly to the clamped flyback voltage, which for example, is approx 180v for a Minelab PI. The diode capacitance also falls rapidly as the charge on the drain and the voltage on the cathode rises and it reaches it's lowest point in this case at around 50v. The energy stored in the diode's capacitance then begins to discharge back into the system.

The mosfet coss is isolated by the diode, you can use a capacitor substitution box to place a relatively high capacitance (say 8000 pf) across the mosfet drain/source (with the diode in series and no bleeder) and still see no significant change in the decay curve.

You would be better off building and perfecting the thing before introducing stuff that will cause problems you end up blindly chasing. The only time a bleeder would work is if you introduced it immediately as the flyback begins to rise and it would need a very clever and fast feedback loop to control this and to get it to actually work.... but then this is the subject of a patent recently discussed on this forum!!!!

[moodz]

Not sure where you get the stored 12v DC from.

The mosfet is turned on during the pulse and therefore has no "stored 12vdc in the mosfet COSS". The mosfet then tuns off and the voltage on the drain rises rapidly to the clamped flyback voltage, which for example, is approx 180v for a Minelab PI. The diode capacitance also falls rapidly as the charge on the drain and the voltage on the cathode rises and it reaches it's lowest point in this case at around 50v. The energy stored in the diode's capacitance then begins to discharge back into the system.

The mosfet coss is isolated by the diode, you can use a capacitor substitution box to place a relatively high capacitance (say 8000 pf) across the mosfet drain/source (with the diode in series and no bleeder) and still see no significant change in the decay curve.

You would be better off building and perfecting the thing before introducing stuff that will cause problems you end up blindly chasing. The only time a bleeder would work is if you introduced it immediately as the flyback begins to rise and it would need a very clever and fast feedback loop to control this and to get it to actually work.... but then this is the subject of a patent recently discussed on this forum!!!!

...stay away from bleeder resistors ... the COSS is lower at high VDS vs low VDS ...

Just use a diode or a silicon carbide switch :-)

~[ATTACH]33264[/ATTACH

[baum7154]

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

After drawing this out on paper it is apparent that the Bleeder creates a path to ground behind the diode and allows the diode to be forward biased in flyback too. Never mind.


Dan

-----------------------------------------------
Moodz--...stay away from bleeder resistors ... the COSS is lower at high VDS vs low VDS

Thanks, I got that in my post response to Green above, hence the 'never mind' statement. This was a discussion of the comment of 'bleeder resistance' from Joe's posting and an earlier discussion I had with Davor on the topic and were kicking around how this could be done and wasn't intended to be a proven mod to any detector. Thanks for your inputs on this topic.

Regards

Dan

[bbsailor]

-----------------------------------------------
Moodz--...stay away from bleeder resistors ... the COSS is lower at high VDS vs low VDS

Thanks, I got that in my post response to Green above, hence the 'never mind' statement. This was a discussion of the comment of 'bleeder resistance' from Joe's posting and an earlier discussion I had with Davor on the topic and were kicking around how this could be done and wasn't intended to be a proven mod to any detector. Thanks for your inputs on this topic.

Regards

Dan

It was not my intent to create a debate about bleeder resistors. I recall that Eric Foster who made very fast PPS (about 3,000 PPS) PI detectors would add a series diode to attempt to reduce the MOSFET effective capacitance by placing another capacitance device, the diode in series with the MOSFET, thereby reducing the total capacitance. However, he found that at the junction of the MOSFET and the diode the voltage would raise to near the peak flyback voltage and he choose to bleed off that voltage above the supply voltage in an attempt to speed things up. Eric was also a fan of integrating the output in an attempt to get more sensitivity on the RX side rather than pumping more power into the TX side with the side effects of increasing the flyback voltage, MOSFETS and passive resistors heating up and inefficient use of battery power. See this link for new information on using digital boxcar integration for metal detectors something that Eric did years ago in an analog way. http://www.anadigm.com/_apps/AN231013-U308.pdf

Anything you can do to reduce the effect of capacitance in the TX circuit will allow you to sample faster but only if the PI circuit has the potential to use this faster coil response. There are may ways to skin this cat. Today's digital technology will eventually allow us to drive a PI metal detector external hardware module and coil from our Ipod/Touch/Ipad or Android device, have the processing power in the digital app and view the target signal response on the display screen. By adding some intelligent used-based inputs, desired target types, searching soil conditions, coil types and sizes and search speeds can be optimized to aid in selecting the best hardware and searching parameters for the given situation. This forum is one of the places where technical discussions are happening that will allow this to happen.

Joseph J. Rogowski

[moodz]

It was not my intent to create a debate about bleeder resistors. I recall that Eric Foster who made very fast PPS (about 3,000 PPS) PI detectors would add a series diode to attempt to reduce the MOSFET effective capacitance by placing another capacitance device, the diode in series with the MOSFET, thereby reducing the total capacitance. However, he found that at the junction of the MOSFET and the diode the voltage would raise to near the peak flyback voltage and he choose to bleed off that voltage above the supply voltage in an attempt to speed things up. Eric was also a fan of integrating the output in an attempt to get more sensitivity on the RX side rather than pumping more power into the TX side with the side effects of increasing the flyback voltage, MOSFETS and passive resistors heating up and inefficient use of battery power. See this link for new information on using digital boxcar integration for metal detectors something that Eric did years ago in an analog way. http://www.anadigm.com/_apps/AN231013-U308.pdf

Anything you can do to reduce the effect of capacitance in the TX circuit will allow you to sample faster but only if the PI circuit has the potential to use this faster coil response. There are may ways to skin this cat. Today's digital technology will eventually allow us to drive a PI metal detector external hardware module and coil from our Ipod/Touch/Ipad or Android device, have the processing power in the digital app and view the target signal response on the display screen. By adding some intelligent used-based inputs, desired target types, searching soil conditions, coil types and sizes and search speeds can be optimized to aid in selecting the best hardware and searching parameters for the given situation. This forum is one of the places where technical discussions are happening that will allow this to happen.

Joseph J. Rogowski

Appreciate your comments BB ... I dont think the discussion on bleeder resistors was very clear and any discussion should be based on facts. The references to Eric's circuits should have been backed with a diagram and a basic google search of COSS and its relationship to Vdrain would have immediately revealed that a bleeder resistor may not be the best option. I am always upset where less informed folks on this forum are misled by other folks who dont do the basic research before going to print. Thats my opinion only. :-)

[bbsailor]

Moodz,

Anyone looking at the MOSFET COSS charts will see that the effective MOSFET capacitance tends to decrease with increased voltage. But, since the high voltage diode also has capacitance, and it is in series with the MOSFET COSS, this combination represents a total lower capacitance in the TX circuit and thus has the potential to sample at lower delays with a higher damping resistor value. If the bleeder resistor between the diode and MOSFET is not the best option, is there better option to allow a faster TX spike turn-off with a faster RX turn-on to sample at lower delays? This is a forum to help PI builders learn about what works and why. This is the approach I took in writing my fast coil article.

Thanks for the feedback.

Joseph J. Rogowski

[crane]

But the diode is a diode first and a capacitor second and it conducts while the mosfet
capacitance is charging so the in-series capacitance rule can't help us in this case.