Tweet
In case anyone is interested, I have come across some GREAT university lecture series. Topics include physics, math, electronics, etc. I especially like Walter Lewin physics series from MIT... http://www.youtube.com/profile_play_list?user=MIT http://www.youtube.com/profile_play_list?user=nptelhrd http://freevideolectures.com/physics.html
-
-
Hi Rube,Originally posted by rube the noob View Post
thanks very much for the links! I like the mad professor, that teaches in MIT electro dynamics. Really very interesting lessons. This will help to check and improve my coil software.
Regards,
Aziz
-
There is an interesting topic in lecture 16, starting at 34:50 minutes, in the MIT E/M series. It is about induced EMF due to a solenoid and non-intuitive consequences which result from non-conservative fields. If anyone has 20 minutes to spare, this is a good way to spend it. Maybe it has subtle implications to metal detector coils and circuity.... http://www.youtube.com/watch?v=G3eI4...FD494&index=16Comment
-
On lecture 20 (http://ocw.mit.edu/OcwWeb/Physics/8-...ures/index.htm ), there is also very interesting topic: switch LR-circuit on and off.
Particularly, after switch off (t:= 0), dI/dt has its maximum (through coil and dumping resistor) and also the produced B-Field too. So this tells PI-people, that they should start sampling as quick as possible after switching off. This is the reason, why some newer PI detectors began start sampling much earlier than former to get more sensitivity. Especially for gold.
Very charming professor Walter Lewin.
During some lessons last night, he remembered me the frequency domain behavior of coils again and I could improve my software.
AzizComment
-
Things to Consider
AZIZ
Go back to your previous post where you drew an illustration of a coil and MOSFET switch discharging through its own resistance. That inaccurately causes a discharge time that is equal to the charge time and is wrong in terms of a PI circuit. In the coil discharge path, place the value of a damping resistor and you will notice that the discharge time is much shorter than the charge time.
A 300 uH coil with a total charge path resistance of 10 ohms (coil resistance plus the MOSFET on-resistance plus cable resistance) has a time constant (TC) of 300/10 or 30 uS. According to the video, after one TC the current will rise to 63% of max. After 2 TCs the current will rise to 86% of max and after 3 TC the current will rise to 94% of max and after 4 TCs the current will rise to 97% of max and after 5 TCs the current will rise to over 99% of max, normally called the flat line current. The discharge time is 300uH divided by approximately 500 ohms and is about .6us. See below for more details about how this is calculated. You must put real circuit component values in your calculation to be realistic.
Once the MOSFET turns off the current will discharge in three stages.
1. The MOSFET clamping stage if the flyback voltage exceeds the MOSFET voltage and the MOSFET clamps the flyback voltage until it falls below the MOSFET clamping voltage, typically about a microsecond or two.
2. The conductance voltage of the clamping diodes between .06 and .07 Volts causes the input resistor (Rin) of a mono coil first amplifier circuit op amp to be effectivly in parallel with the damping resistor (Rd) until the flyback voltage oscillations fall below .06 to .07 V. This means that if Rd is 1000 ohms and Rin is 1000 ohms then Rd is effectivly 500 ohms for most the discharge time. use this value in calculating the discharge TC of the coilk until the voltage reaches .06 to .07V depending on the diodes used for clamping.
3. Below .06 to .07V the value of RD becomes the new value for calculating the TC voltage decay until the voltage reaches the saturation voltage of the op amp. Here the voltage powering the op amp and the gain determined where saturation occurs. Op amps take some time to come out of saturation and this thie determines the earliest sampling time.
The value of the damping resistor is directly related to the inductance value of the coil and it's resonant frequency. Lower resonant frequency means higher winding capacitance and that a lower value Rd must be used to dampen the flyback oscillations.
Anything that can be used to make the coil resonant frequency higher means that a higher damping resistor can be used and that the area under the discharge curve will be less. This includes all things in the TX circuit which tend to add capacitance such as:
1. Coil capacitance and insulation dielectric
2. Shield-to-coil capacitance
3. Coax cable capacitance
4. MOSFET capacitance (COSS)
Increasing the value of any of the above things tends to slow a PI coil down as it is reflected in a lower Rd value to attempt to dampin the flyback oscillations. As the areas under the decay curve increases the potential speed of the coil decreases and vica versa.
You should consider redrawing your Coil-MOSFET-Switch graphic and add Rd to the discharge path to properly analyze and evaluate coil dynamics.
Once you do this you will see how a PI circuit relates to the lecture that you commented on below.
bbsailorComment
-
Hi bbsailer,Originally posted by bbsailor View Post
Excellent explanation - but did you mean to say "between 0.6 and 0.7 volts"?Comment
-
Originally posted by Qiaozhi View Post
Yes,
This was a slip of my mind and unforgiving fingers once I hit the send button. Thanks for catching this error on my part.
The actual value should be .6 V to .7 V as diode voltage. I hope Aziz will forgive me for my error.
bbsailorComment
-
Hi bbsailor,
this is a very interesting point of view. May be this is a good task for simulating the RLC circuit with apropriate simulation software (PSPICE?). I am not familiar with PSPICE. But I could implement some similar code, that handles such complex networks. So the hole circuit with its complex resistors has to be taken into account. Is there any known software for calculating the parallel-capacitance of arbitrary coil? Otherwise, this have to be measured. All other parameters are quite easy to get.
At the moment, I am busy with bug-fixing the coil software. I will pipeline this interesting issue.
Regards,
AzizComment
-
Aziz,
Go to the web link below.
http://www.miscel.dk/MiscEl/miscelAirCoil.html
There is a graphic component that depicts the inductor charge and discharge curves. You must enter the coil inductance and discharge current and discharge resistance.
You must do this in two phases. First above .7V with Rd and Rin in parellel and then again below .7V when Rd is the only discharge path.
bbsailorComment
-
There is a brief and not-too-enlightening section on metal detectors in lecture 25, starting at 43:36. I thought I would mention it anyway in case anyone is interested. http://ocw.mit.edu/OcwWeb/Physics/8-...ures/index.htmComment
Comment