Shop OBEX P1 Docs P2 Docs Learn Events
I'm sorry magic smoke, ur just not real.. — Parallax Forums

I'm sorry magic smoke, ur just not real..

RinksCustomsRinksCustoms Posts: 531
edited 2007-07-29 23:17 in Propeller 1
It bothers me to see some that get led astray by "mostly acurate" information. In any case, I am going to put the "magic smoke" theory to rest. THERE IS NO SUCH THING AS MAGIC SMOKE! That smoke you see is the unconsumed remains of the internals of the device which is in a thermal runaway state. When·Electronic Engineers design a circuit, they have been taught to select components for the circuit in question using the 2x rule. Basically you want to select a component that can tollerate twice the power the circuit was designed for. When you have voltage and current and present resistance to it within a circuit, the resistance drops the voltage across the resistive load and disipates the excess energy in the form of heat. Resistors for example, have wattage ratings ranging from 1/16W to 200W+. You may apply as much voltage as you like through the resistor, as long as you do not exeed the calculated current for the given wattage of said resistor.

Tutorial Ex. on Ohm's Law:

If we have a circuit with 12V and a 10 Ohm resistor rated for 10W, would the current flowing through the resistor exeed the rated 20 watts sending the resistor into a thermal runaway state?

It's simple to answer with a little calculation. Any math formula needs at least two parts to find an unknown answer. In this case we need to determine maximum wattage the resistor must dissipate with the voltage and current presented to it. What's that? I didn't say how much current is going through it? Your right, I didn't, because I don't know... yet. Using Ohms law, presented in the pyramid graphic below, all need be done to find our missing current value is cover up with a finger what we're looking for (in this case I), and the formula to solve presents itself exactly as shown.

OhmsLawPyramid.gif

E = Electromotive force : base·- Volt
I = Current : base - Ampere
R = Resistance : base - Ohm

Covering up "I" (current), you'll see that the formula is E over R. This means you divide E by R to get I, so, plugging the numbers in and:

· E···/·· R····· =··· I
12v /·10 ohm = 1.2A

Huh, now we both know exactly how much current is passing through the resistor. That's great but we still don't know how much wattage the resistor has to try and dissipate. To now solve for wattage or power as it's also referred to you can use this pyramid:
PowerPyramid.gif

P· = Power "Wattage" : base - Watt
I·· = Current : base - Ampere
E = Electromotive force: base Volt

Here we want to solve for power so we cover P up and we have the right formula for finding the wattage the resistor has to dissipate in the form of thermal energy.
· I··· x·· E·· =·· P
1.2A x 12V = 14.4W

This calculation shows that the resistor must dissipate almost 150% the energy it was designed for. So to answer the question of the resistor going into a thermal runaway state is obvious. To be able to short a car battery with a 10 Ohm resistor and not have it fail we should use the 2x rule. We know that this circuit uses 14.4W, so 2 x 14.4 = 28.8W. Since nobody makes a 28.8W resistor, err higher and choose a 30W 10 Ohm resistor. Yes 15W versions exist and it may be able to dissipate the load, but it would be very hot and could damage nearby componets with infrared heat radiation. The 30W resistor should only get warm to luke warm and will stablize at that temperature continuously as long as the voltage to this circuit is stable.

The Wiki entry on magic smoke (you should note), scroll down to the bottom of the page and note the catgory it's under, Electronics| In-joke | Computer humor. I posted this thread in hopes that people reading the forums with lesser expierience in electronics do not take this bad joke seriously, a.k.a. as a seroius terminology for electronics.· Often i see posts regarding "magic smoke" as a seroius term and I just want to set the record straight. In case any are in doubt, i have read the wiki definition of thermal runaway·and it is accurate, but I would not swear by the site as i found a few inconsistencies already.


▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
E3 = Thought

http://folding.stanford.edu/·- Donating some CPU/GPU downtime just might lead to a cure for cancer! My team stats.
«1

Comments

  • GadgetmanGadgetman Posts: 2,436
    edited 2007-07-24 21:45
    Anyone taking the 'magic smoke' comments seriously probably also believes that the Earth is flat, that the Easterbunny is Kevin(from the www.kevinandkell.com comic) and that santa has a factory on the North pole...
    (There is nothing but sea and ice on the North Pole. Atomic subs have sailed UNDER the ice there... )

    Mostly the 'magic smoke' comments are to lighten the mood after someone have torched an expensive chip...

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Don't visit my new website...
  • deSilvadeSilva Posts: 2,967
    edited 2007-07-24 22:05
    I assume the pyramid sketches are meant as an memory hook, but I cannot see why someone who cannot memorize Ohm's Law should be able to memorize where the hell E, I, or R has to be placed inside the pyramid.

    I always found it very convincing to just use common sense:

    When you increase the voltage at the ends of a resistor more current will flow, which can straigtforward be expressed as:
    I prop. U
    On the other side, when you increase the resistor's value , less current will flow, mathematically
    I prop. 1/R
    Putting it together: I = U/R

    Similar argument holds for
    P = U*I (More Volts, more Watts; more Amperes , more Watts)
  • RinksCustomsRinksCustoms Posts: 531
    edited 2007-07-24 22:22
    ever think someone might have a photographic memory? True you don't need the pyramid, but it's reinforcing for someone with a photographic memory. Representing U for E doesn't make a whole lotta sense to me, but meh, to each his own.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    E3 = Thought

    http://folding.stanford.edu/·- Donating some CPU/GPU downtime just might lead to a cure for cancer! My team stats.
  • Mike GreenMike Green Posts: 23,101
    edited 2007-07-24 22:50
    RinksCustoms,
    Just for the record, what you're describing may or may not be thermal runaway. The term refers to a positive feedback situation where increasing heat causes a worsening of the problem (too much power being dissipated or too little ability to dissipate it). I'd call your description that of thermal overload. A runaway situation might be one where the resistance of a material decreases with increasing temperature. As the temperature rises, the resistance drops, current through the device increases and more energy gets converted into heat. Another runaway possibility might be that the electrical resistance is stable, but the thermal resistance is increasing (like the radiator in a car boiling over ... eventually there's loss of coolant and the temperature markedly increases). It might be that the resistance of molten material is the same as the solid material, but mechanical failure occurs due to things melting ... that's not thermal runaway ... it's material failure due to extreme temperatures.
  • deSilvadeSilva Posts: 2,967
    edited 2007-07-24 22:52
    RinksCustoms said...
    Representing U for E doesn't make a whole lotta sense to me.
    My fault, again! Americans use "V" rather than "U", as in rest of the world smile.gif
  • CJCJ Posts: 470
    edited 2007-07-24 23:08
    Rinkscustoms,

    from your topic title, I thought you had conducted some sort of torture test for the propeller smile.gif

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Parallax Forums - If you're ready to learn, we're ready to help.
  • parts-man73parts-man73 Posts: 830
    edited 2007-07-24 23:11
    Gadgetman said...
    Anyone taking the 'magic smoke' comments seriously probably also believes that the Earth is flat, that the Easterbunny is Kevin(from the www.kevinandkell.com comic) and that santa has a factory on the North pole...
    (There is nothing but sea and ice on the North Pole. Atomic subs have sailed UNDER the ice there... )

    Mostly the 'magic smoke' comments are to lighten the mood after someone have torched an expensive chip...

    Don't visit my new website...

    ...and people that believe that, may also believe that you have a new website smilewinkgrin.gif

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Brian

    uController.com - home of SpinStudio
  • BeanBean Posts: 8,129
    edited 2007-07-24 23:12
    Yeah magic smoke. I wonder where you can buy a can of that ? I'd like to try getting some back into some chips I toasted.

    While I'm at it, I need some "bit lube" too. I don't have any problem with the "O" bits, but the "1" bits keep getting stuck if the wire is too small. At least I think that is what's happening...

    Bean

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Teacher: What is the difference between ignorance and apathy ?
    Student: I don't know and I don't care
    Teacher: Correct !
    - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
    www.hittconsulting.com
    ·
  • mirrormirror Posts: 322
    edited 2007-07-24 23:38
    While we're at it.

    It seems the device most of us know the most and the least about is the widget! If it had a spec it would run to millions of pages to incorporate all the functionality that's been assigned to it.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
  • Fred HawkinsFred Hawkins Posts: 997
    edited 2007-07-25 05:24
    RinksCustoms said...
    ever think someone might have a photographic memory? True you don't need the pyramid, but it's reinforcing for someone with a photographic memory. Representing U for E doesn't make a whole lotta sense to me, but meh, to each his own.

    (Raises hand)·Speaking for the ·algebra-challenged, I like your·motifs. They even work for the divisions.

    Of course, now I have to remember which one goes into the·pyramid's top.

    Did you ever notice that before magic smoke you get burns?

    ·
  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2007-07-25 05:39
    RC,

    The best jokes are the ones that are told and taken seriously so it is always with a serious expression on my face that I will state that I know what the problem is .... "the magic smoke has escaped". Magic smoke?? what?? Then I explain....

    Or else you explain how electronics works, it's not with silicon but with smoke, and it must be magic because look, when it escapes the chip stops working, see!

    So you see the term "magic smoke" is a very serious term indeed.

    Also, we know when we explain things that we will perhaps use the wrong term and somebody will correct us. I'm sure you know the difference between thermal runaway vs dissipation but Mike clarified this for the benefit of some readers. What you did not clarify yourself was the fact that thermal dissipation relies upon the environment, heatsinking, convection, thermal resistance, radiant characteristics of the material, duty cycle etc. I could use a 10R 1W resistor across 12V if I limited the duty cycle and maximum on time, or if it had an effective heatsink etc.

    Too often somebody will apply +12V to a regulator that is normally running warm at +8V or draw 300ma from a 500ma regulator when it normally runs at 100ma but is not equipped with any heatsink and then we see magic smoke! Seriously smilewinkgrin.gif

    *Peter*
  • GadgetmanGadgetman Posts: 2,436
    edited 2007-07-25 06:54
    parts-man73 said...

    ...and people that believe that, may also believe that you have a new website smilewinkgrin.gif

    It WAS a new website back when I wrote that.
    And one of these days, I'll even finish it...
    (Yeah, and Santa is stuck in my chimney.)

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Don't visit my new website...
  • SkogsgurraSkogsgurra Posts: 231
    edited 2007-07-25 07:19
    Bean (Hitt Consulting) said...
    Yeah magic smoke. I wonder where you can buy a can of that ? I'd like to try getting some back into some chips I toasted.

    While I'm at it, I need some "bit lube" too. I don't have any problem with the "O" bits, but the "1" bits keep getting stuck if the wire is too small. At least I think that is what's happening...

    Bean

    Dear Bean,

    This is a well known phenomenon that is more and more common these days when structures are going nanometres and 32 bit variables are common. It was noticed in some critical applications that narrow information channels not only make bits stick - they also get worn and fuzzy from friction against narrow communication channel walls.

    Since bits are valuable pieces of·information and virgin bits are hard to find these days, the practice of bit recycling is gaining popularity. I have used products from this company with great success: http://cd.textfiles.com/hotgames/PROGRAMS/RECYCLER/RECYCLER.TXT·But it seems that they are out of recyclers for the moment. Nonetheless, they will probably be well stocked before Christmas. Why not put one on the wish-list and hope that Sandyclaw gets out of the chimney till then?

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
  • mirrormirror Posts: 322
    edited 2007-07-25 08:07
    Bits that are worn and fuzzy should be laid to rest.

    See http://academics.vmi.edu/ee_js/Research/IC_Datasheets/digital_cmos/Write%20Only%20Memory.pdf·for the original solution and·http://en.wikipedia.org/wiki/Write_Only_Memory·for more information.
  • LarryLarry Posts: 212
    edited 2007-07-25 08:52
    There are the concepts that become true if enough people believe in them.

    eg. DC current flows from + to -

    If somebody writes up magic smoke with enough formulas and graphs, it will become accepted, too

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
  • SkogsgurraSkogsgurra Posts: 231
    edited 2007-07-25 09:35
    mirror!

    I am shocked - nay FLABBERGASTED - by your obvious lack of concern for the environment.

    Please bear in mind that the Signeticts WOM was conceived, designed and marketed during a period where natural resources were abundant and one couldn't care less about the future of the Earth.

    What served well at those dark ages does NOT serve us well today. Please, PLEASE! Do not erode the good work carried out by Greenpeace, Mr Gore and all good men and women trying to make the Earth a good planet for us, our children, grandchildren and all coming generations to live on.

    Therefore: Rethink! Do not just throw old bits away. RECYCLE!

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
  • hippyhippy Posts: 1,981
    edited 2007-07-25 11:51
    And while we are at it - Could someone please stop the Propeller cogs executing opcodes ? They've done nothing wrong and do not even get the benefit of a fair trial before execution.
  • inserviinservi Posts: 113
    edited 2007-07-25 12:48
    Hello RinksCustoms,

    I am sorry but I have friends who know somebody who work in a factory where this 'magic smoke' is produced. The Blue one !

    dro.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    in medio virtus
  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2007-07-25 13:06
    One thing about WOMs is that they were highly considered for aerospace applications due to the absolute guarantee from the manufacturer that no memory device would suffer from read errors, a problem that aerospace engineers had even with rad hardened devices. This single assurance almost got them a foot in the door until they found out about the six foot fan because they only had room for a four foot one. Seriously.

    *Peter*
  • SkogsgurraSkogsgurra Posts: 231
    edited 2007-07-25 13:57
    I can see your point, hippy.

    But, I must say that I have tried them opcodes a fair number of times. If that isn't a fair trial...

    Some of them are real stubborn and can't be executed at all.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
  • MightorMightor Posts: 338
    edited 2007-07-25 15:23
    Well, the NOPs just sit around and do nothing anyway.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    | What the world needs is more geniuses with humility, there are so few of us left.
  • inserviinservi Posts: 113
    edited 2007-07-25 19:42
    Hello Bean,

    We will have soon a solution to this problem. A team is currently working over algorithm which makes it possible to compact informations on 1 bit. This work is practically completed. I propose to help them for the second phase which consists in developing the routine of loosening.

    dro.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    in medio virtus
  • Peter JakackiPeter Jakacki Posts: 10,193
    edited 2007-07-25 23:18
    Compacting has already been accomplished for many years now. How many times have we seem Hollywood enhance enhance enhance a picture so that the reflection of the reflection in the mirror reveals the culprits face clearly! Amazing! I've done my calculations and figure that with this Hollywood technology we can forget about increasing megapixels on cameras and do it all with a single pixel. Come to think of it with this technology I can map the surface of Mars from my bedroom window, I just got to figure out a way to automate the repeated pressing of the enhance button.

    *Peter*
  • SkogsgurraSkogsgurra Posts: 231
    edited 2007-07-26 03:57
    Nano imprint technology is where the action is!

    Nano imprint and even more so, sub-nano imprint, allows scientists in Molvinia to imprint information on all six sides of a bit. Thus reaching information densities more than two orders of magnitude higher than possible with yesterday's outdated information technology, a Molvinia Central Committee for the Advancement of Science spokesman says.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
  • mirrormirror Posts: 322
    edited 2007-07-26 04:12
    Skogsgurra said...
    ..., sub-nano imprint, allows scientists in Molvinia to imprint information on all six sides of a bit.
    Kind of like a Rubik's bit.
  • Graham StablerGraham Stabler Posts: 2,510
    edited 2007-07-26 09:09
    Peter,

    I was watching an episode of CSI the other evening, they had CCTV footage of a guy in an airport and a jewelry store but only from the back. There was a stain on his jacket so they zoomed in improbably (typical CCTV will be 640X480) and compared them, at the end of the process which involved much fancy animation the computer came up with "100% seam match found".

    So not only can they zoom beyond the camera's capability they had software ready specifically to compare seams of jackets. I can only imagine what their start bar looks like [noparse]:)[/noparse]

    I'd guess that 90% of CSI is complete BS bust still they do have nice glass boards.

    Graham

    p.s. How do they get things to stick to the glass boards seemingly without adhesive?
  • RinksCustomsRinksCustoms Posts: 531
    edited 2007-07-29 20:27
    inservi said...
    Hello RinksCustoms,

    I am sorry but I have friends who know somebody who work in a factory where this 'magic smoke' is produced. The Blue one !

    dro.

    You are refering to Aerogel i do hope. Mike Green, I stand corrected, i was refering to thermal overload, whereas thermal runaway is a·phenomena mainly concerned with transistors.

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    E3 = Thought

    http://folding.stanford.edu/·- Donating some CPU/GPU downtime just might lead to a cure for cancer! My team stats.
  • RinksCustomsRinksCustoms Posts: 531
    edited 2007-07-29 21:44
    Peter Jakacki said...
    RC,

    The best jokes are the ones that are told and taken seriously so it is always with a serious expression on my face that I will state that I know what the problem is .... "the magic smoke has escaped". Magic smoke?? what?? Then I explain....

    Or else you explain how electronics works, it's not with silicon but with smoke, and it must be magic because look, when it escapes the chip stops working, see!

    So you see the term "magic smoke" is a very serious term indeed.

    Also, we know when we explain things that we will perhaps use the wrong term and somebody will correct us. I'm sure you know the difference between thermal runaway vs dissipation but Mike clarified this for the benefit of some readers. What you did not clarify yourself was the fact that thermal dissipation relies upon the environment, heatsinking, convection, thermal resistance, radiant characteristics of the material, duty cycle etc. I could use a 10R 1W resistor across 12V if I limited the duty cycle and maximum on time, or if it had an effective heatsink etc.

    Too often somebody will apply +12V to a regulator that is normally running warm at +8V or draw 300ma from a 500ma regulator when it normally runs at 100ma but is not equipped with any heatsink and then we see magic smoke! Seriously smilewinkgrin.gif

    *Peter*
    Perhaps i confused thermal runaway with thermal overload, and I am aware of the factors of thermal conductivity, duty cycles and so fourth. I did not include an explination of all factors because i'd still be writing the post and they would have more or less canceled out the purpose of the post.


    "Or else you explain how electronics works, it's not with silicon but with smoke, and it must be magic because look, when it escapes the chip stops working, see!"

    Ugh... I won't explain how electronics work, it'd take too much of my time. http://computer.howstuffworks.com/diode.htm·<--- if you'd like to know precicely how electronics work.
    Basically, when you overload a PN junction with more Voltage or current that it was designed for, it genrates heat at the PN junctions. Thermal conductivity data included with any datasheet expresses how good the case material is at dissipating the junction heat. This is typically expressed as C/W (how much the case temperature will increase in celcius per watt disipated with no heat sink and at a 70F room temp).



    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    E3 = Thought

    http://folding.stanford.edu/·- Donating some CPU/GPU downtime just might lead to a cure for cancer! My team stats.
  • RinksCustomsRinksCustoms Posts: 531
    edited 2007-07-29 22:00
    Skogsgurra said...
    Nano imprint technology is where the action is!

    Nano imprint and even more so, sub-nano imprint, allows scientists in Molvinia to imprint information on all six sides of a bit. Thus reaching information densities more than two orders of magnitude higher than possible with yesterday's outdated information technology, a Molvinia Central Committee for the Advancement of Science spokesman says.

    That's a pretty cool concept, i wonder what the integrity is of that memory tech?

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    E3 = Thought

    http://folding.stanford.edu/·- Donating some CPU/GPU downtime just might lead to a cure for cancer! My team stats.
  • RinksCustomsRinksCustoms Posts: 531
    edited 2007-07-29 22:25
    I stand corected!!! AfterMuch testing with my $5 Petaillion cosmic ray crossvectoring sub-particle nanoscope (which is capable of travelling between silicon atoms) on the propeller chip, guided by advanced CPS technology (Chip Positioning System) THAT MAGIC SMOKE REALLY DOES EXIST!!!!! It is chemically similar to the same smoke that when applied to a specific region, makes the subject feel more confident and secure about shady subject matter - see a used car salesman for more details on that particular application. In fact it is also this same magic smoke Cheech & Chong used in all their movies!!

    And this discovery also just made, A Polock from poland has infact found the corner of the wall in a round room!! This achievement is only superseeded by his miracle inventions: the Screen Door device for sumarines, and the household item, The solar Powered Flashlight, which requires absolutely no batteries as it runs soley on solar power!!

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    E3 = Thought

    http://folding.stanford.edu/·- Donating some CPU/GPU downtime just might lead to a cure for cancer! My team stats.
Sign In or Register to comment.