Charge Conservation - Hint of the Day
in [Design]
|
From: George Herold on 21 Jul 2010 21:41 On Jul 21, 12:52 pm, John Larkin <jjlar...(a)highNOTlandTHIStechnologyPART.com> wrote: > On Wed, 21 Jul 2010 07:28:10 -0700 (PDT), George Herold > > > > > > <gher...(a)teachspin.com> wrote: > >On Jul 20, 8:32 pm, Tim Wescott <t...(a)seemywebsite.com> wrote: > >> On 07/20/2010 09:32 AM, Jim Thompson wrote: > > >> > On Tue, 20 Jul 2010 08:53:22 -0700, Tim Wescott<t...(a)seemywebsite.com> > >> > wrote: > > >> >> On 07/20/2010 08:24 AM, Jim Thompson wrote: > >> >>> Charge Conservation - Hint of the Day: > > >> >>> How many Coulombs can a 1mH inductor charged to 1A deliver? > > >> >> That's insufficient information, and I rather expect that you know it. > > >> > No. It's provided to cause young bucks to do some thinking. Looks > >> > like it didn't work with you :-( > > >> > (Except that it did annoy Larkin, yet again... so a partial success > >> > :-) > > >> Since you didn't answer I have to assume that you couldn't. > > >> Either this is a trick question, and the answer is "however many excess > >> electrons it has sitting on it when I hand it to you", or the answer is > >> "that depends on the coil resistance". > > >> A 1mH superconducting inductor with 1A will deliver (or flow, if you > >> want to quibble about the common EE definition of "deliver") an infinite > >> charge to a dead short, assuming all conductors are also zero resistance. > > >> Otherwise a 1mH inductor that sees R ohms of total circuit resistance in > >> the inductor and the load (charge target?) will see it's current decay > >> as (1A)*e^-(R/L)*t; this will integrate to (1A) * (L/R). So for 1 ohm > >> total resistance that'd be 1mC, for a 10 ohm total resistance that'd be > >> 100uC, for a 0.1 ohm total resistance it'd be 10mC, etc. > > >> Answers involving loads that aren't purely resistive are more > >> complicated, but still obvious if you can understand the above. > > >> But to answer how much charge that 1mH inductor _can possibly_ deliver > >> when it has 1A flowing through it depends on the particular inductor's > >> winding resistance and possibly also on whether it's really a 1mH > >> inductor when it has 1A flowing through it. > > >> You may want to pop over to the closest ASU campus that presents EEE 202 > >> and see if you can audit the course. This problem is no great mystery > >> for someone who's gotten through sophomore electronics engineering. > > >> -- > > >> Tim Wescott > >> Wescott Design Serviceshttp://www.wescottdesign.com > > >> Do you need to implement control loops in software? > >> "Applied Control Theory for Embedded Systems" was written for you. > >> See details athttp://www.wescottdesign.com/actfes/actfes.html > > >Hi Tim, I agree with your calculation. But not the interpretation. > >Sure you integrate current over time and you get charge. But this is > >not the charge delivered to a resistor, it is how much charge flowed > >through it. (Oh unless that's what is meant by delivered.) > > >George H. > > That's a common way of stating it. When we put parts on boards, their > electrostatic potential is not often of concern. So we say that we put > charge into a capacitor or a battery, literally we say "charge a > battery" or "charge a capacitor" as opposed to "run charge through a > capacitor", and we measure how much in ampere-seconds, namely > coulombs. > > Since both a cap and a battery save the ampere-seconds and can > return/deliver them later, it's reasonable to think that they stored > charge. > > The numbers work. Engineering is about what works. > > John- Hide quoted text - > > - Show quoted text - Yeah sure, I use the same words. But isn't this the cause of the current .... 'confusion'? NPI It's kinda like the water in a hose analogy of current in circuits. It mostly works... but you can't spray charge out the end of a circuit, you've gotta have a conductor attached. (let's assume no high voltages.) We really only measure voltages and currents, we all know there are charges moving around, but as soon as you try and get them to hold still they disappear. (Well you can use a Faraday bucket.) I've been looking at flux leakage with our Keithley 601B electrometer. At the 10^11 ohms scale (with the multiplier set at X10 or X30) I can't walk near the circuit with out upsetting things. George H.
From: George Herold on 21 Jul 2010 21:59 On Jul 21, 1:27 pm, Tim Wescott <t...(a)seemywebsite.com> wrote: > On 07/21/2010 07:28 AM, George Herold wrote: > > > > > > > On Jul 20, 8:32 pm, Tim Wescott<t...(a)seemywebsite.com> wrote: > >> On 07/20/2010 09:32 AM, Jim Thompson wrote: > > >>> On Tue, 20 Jul 2010 08:53:22 -0700, Tim Wescott<t...(a)seemywebsite.com> > >>> wrote: > > >>>> On 07/20/2010 08:24 AM, Jim Thompson wrote: > >>>>> Charge Conservation - Hint of the Day: > > >>>>> How many Coulombs can a 1mH inductor charged to 1A deliver? > > >>>> That's insufficient information, and I rather expect that you know it. > > >>> No. It's provided to cause young bucks to do some thinking. Looks > >>> like it didn't work with you :-( > > >>> (Except that it did annoy Larkin, yet again... so a partial success > >>> :-) > > >> Since you didn't answer I have to assume that you couldn't. > > >> Either this is a trick question, and the answer is "however many excess > >> electrons it has sitting on it when I hand it to you", or the answer is > >> "that depends on the coil resistance". > > >> A 1mH superconducting inductor with 1A will deliver (or flow, if you > >> want to quibble about the common EE definition of "deliver") an infinite > >> charge to a dead short, assuming all conductors are also zero resistance. > > >> Otherwise a 1mH inductor that sees R ohms of total circuit resistance in > >> the inductor and the load (charge target?) will see it's current decay > >> as (1A)*e^-(R/L)*t; this will integrate to (1A) * (L/R). So for 1 ohm > >> total resistance that'd be 1mC, for a 10 ohm total resistance that'd be > >> 100uC, for a 0.1 ohm total resistance it'd be 10mC, etc. > > >> Answers involving loads that aren't purely resistive are more > >> complicated, but still obvious if you can understand the above. > > >> But to answer how much charge that 1mH inductor _can possibly_ deliver > >> when it has 1A flowing through it depends on the particular inductor's > >> winding resistance and possibly also on whether it's really a 1mH > >> inductor when it has 1A flowing through it. > > >> You may want to pop over to the closest ASU campus that presents EEE 202 > >> and see if you can audit the course. This problem is no great mystery > >> for someone who's gotten through sophomore electronics engineering. > > >> -- > > >> Tim Wescott > >> Wescott Design Serviceshttp://www.wescottdesign.com > > >> Do you need to implement control loops in software? > >> "Applied Control Theory for Embedded Systems" was written for you. > >> See details athttp://www.wescottdesign.com/actfes/actfes.html > > > Hi Tim, I agree with your calculation. But not the interpretation.. > > Sure you integrate current over time and you get charge. But this is > > not the charge delivered to a resistor, it is how much charge flowed > > through it. (Oh unless that's what is meant by delivered.) > > See my preface. The notion of "delivered charge" crops up pretty > consistently in electrical engineering, and it almost always means > "charge delivered to (some lead of) some component (while ignoring > charge coming out of one or more other leads)". Yes thanks, I don't have any real problem with the notion of delivered charge. I use the same words/ideas all the time. Electrons buzz all about in our circuits. But charge conservation on a capacitor is not ussually important. (Isn't this JT's notion?) George H. > If you're willing to do more math, and to posit a switch that magically > opens when current drops to zero, without itself having any voltage > drop, then you can get exactly the same results (i.e. -- the total > delivered charge is limited by your desire or the total resistance in > the circuit) by having the inductor charging up a capacitor (which will > have no more net charge when the process is done than when it started). > > Or replace the switch by a real-world diode, and find that the total > charge that will flow through the inductor is limited by the diode's > characteristics as well as the rest of the circuit. > > -- > > Tim Wescott > Wescott Design Serviceshttp://www.wescottdesign.com > > Do you need to implement control loops in software? > "Applied Control Theory for Embedded Systems" was written for you. > See details athttp://www.wescottdesign.com/actfes/actfes.html- Hide quoted text - > > - Show quoted text -
From: George Herold on 21 Jul 2010 22:04 On Jul 21, 9:59 pm, George Herold <gher...(a)teachspin.com> wrote: > On Jul 21, 1:27 pm, Tim Wescott <t...(a)seemywebsite.com> wrote: > > > > > > > On 07/21/2010 07:28 AM, George Herold wrote: > > > > On Jul 20, 8:32 pm, Tim Wescott<t...(a)seemywebsite.com> wrote: > > >> On 07/20/2010 09:32 AM, Jim Thompson wrote: > > > >>> On Tue, 20 Jul 2010 08:53:22 -0700, Tim Wescott<t...(a)seemywebsite.com> > > >>> wrote: > > > >>>> On 07/20/2010 08:24 AM, Jim Thompson wrote: > > >>>>> Charge Conservation - Hint of the Day: > > > >>>>> How many Coulombs can a 1mH inductor charged to 1A deliver? > > > >>>> That's insufficient information, and I rather expect that you know it. > > > >>> No. It's provided to cause young bucks to do some thinking. Looks > > >>> like it didn't work with you :-( > > > >>> (Except that it did annoy Larkin, yet again... so a partial success > > >>> :-) > > > >> Since you didn't answer I have to assume that you couldn't. > > > >> Either this is a trick question, and the answer is "however many excess > > >> electrons it has sitting on it when I hand it to you", or the answer is > > >> "that depends on the coil resistance". > > > >> A 1mH superconducting inductor with 1A will deliver (or flow, if you > > >> want to quibble about the common EE definition of "deliver") an infinite > > >> charge to a dead short, assuming all conductors are also zero resistance. > > > >> Otherwise a 1mH inductor that sees R ohms of total circuit resistance in > > >> the inductor and the load (charge target?) will see it's current decay > > >> as (1A)*e^-(R/L)*t; this will integrate to (1A) * (L/R). So for 1 ohm > > >> total resistance that'd be 1mC, for a 10 ohm total resistance that'd be > > >> 100uC, for a 0.1 ohm total resistance it'd be 10mC, etc. > > > >> Answers involving loads that aren't purely resistive are more > > >> complicated, but still obvious if you can understand the above. > > > >> But to answer how much charge that 1mH inductor _can possibly_ deliver > > >> when it has 1A flowing through it depends on the particular inductor's > > >> winding resistance and possibly also on whether it's really a 1mH > > >> inductor when it has 1A flowing through it. > > > >> You may want to pop over to the closest ASU campus that presents EEE 202 > > >> and see if you can audit the course. This problem is no great mystery > > >> for someone who's gotten through sophomore electronics engineering. > > > >> -- > > > >> Tim Wescott > > >> Wescott Design Serviceshttp://www.wescottdesign.com > > > >> Do you need to implement control loops in software? > > >> "Applied Control Theory for Embedded Systems" was written for you. > > >> See details athttp://www.wescottdesign.com/actfes/actfes.html > > > > Hi Tim, I agree with your calculation. But not the interpretation. > > > Sure you integrate current over time and you get charge. But this is > > > not the charge delivered to a resistor, it is how much charge flowed > > > through it. (Oh unless that's what is meant by delivered.) > > > See my preface. The notion of "delivered charge" crops up pretty > > consistently in electrical engineering, and it almost always means > > "charge delivered to (some lead of) some component (while ignoring > > charge coming out of one or more other leads)". > > Yes thanks, I don't have any real problem with the notion of delivered > charge. > I use the same words/ideas all the time. Electrons buzz all about in > our circuits. > But charge conservation on a capacitor is not ussually important. > (Isn't this JT's notion?) > > George H. opps usually. (Have I told you I can't spell?) > > > > > If you're willing to do more math, and to posit a switch that magically > > opens when current drops to zero, without itself having any voltage > > drop, then you can get exactly the same results (i.e. -- the total > > delivered charge is limited by your desire or the total resistance in > > the circuit) by having the inductor charging up a capacitor (which will > > have no more net charge when the process is done than when it started). > > > Or replace the switch by a real-world diode, and find that the total > > charge that will flow through the inductor is limited by the diode's > > characteristics as well as the rest of the circuit. > > > -- > > > Tim Wescott > > Wescott Design Serviceshttp://www.wescottdesign.com > > > Do you need to implement control loops in software? > > "Applied Control Theory for Embedded Systems" was written for you. > > See details athttp://www.wescottdesign.com/actfes/actfes.html-Hide quoted text - > > > - Show quoted text -- Hide quoted text - > > - Show quoted text -- Hide quoted text - > > - Show quoted text -
From: John Larkin on 21 Jul 2010 23:12 On Wed, 21 Jul 2010 18:41:50 -0700 (PDT), George Herold <gherold(a)teachspin.com> wrote: >On Jul 21, 12:52�pm, John Larkin ><jjlar...(a)highNOTlandTHIStechnologyPART.com> wrote: >> On Wed, 21 Jul 2010 07:28:10 -0700 (PDT), George Herold >> >> >> >> >> >> <gher...(a)teachspin.com> wrote: >> >On Jul 20, 8:32�pm, Tim Wescott <t...(a)seemywebsite.com> wrote: >> >> On 07/20/2010 09:32 AM, Jim Thompson wrote: >> >> >> > On Tue, 20 Jul 2010 08:53:22 -0700, Tim Wescott<t...(a)seemywebsite.com> >> >> > wrote: >> >> >> >> On 07/20/2010 08:24 AM, Jim Thompson wrote: >> >> >>> Charge Conservation - Hint of the Day: >> >> >> >>> How many Coulombs can a 1mH inductor charged to 1A deliver? >> >> >> >> That's insufficient information, and I rather expect that you know it. >> >> >> > No. �It's provided to cause young bucks to do some thinking. �Looks >> >> > like it didn't work with you :-( >> >> >> > (Except that it did annoy Larkin, yet again... so a partial success >> >> > :-) >> >> >> Since you didn't answer I have to assume that you couldn't. >> >> >> Either this is a trick question, and the answer is "however many excess >> >> electrons it has sitting on it when I hand it to you", or the answer is >> >> "that depends on the coil resistance". >> >> >> A 1mH superconducting inductor with 1A will deliver (or flow, if you >> >> want to quibble about the common EE definition of "deliver") an infinite >> >> charge to a dead short, assuming all conductors are also zero resistance. >> >> >> Otherwise a 1mH inductor that sees R ohms of total circuit resistance in >> >> the inductor and the load (charge target?) will see it's current decay >> >> as (1A)*e^-(R/L)*t; this will integrate to (1A) * (L/R). �So for 1 ohm >> >> total resistance that'd be 1mC, for a 10 ohm total resistance that'd be >> >> 100uC, for a 0.1 ohm total resistance it'd be 10mC, etc. >> >> >> Answers involving loads that aren't purely resistive are more >> >> complicated, but still obvious if you can understand the above. >> >> >> But to answer how much charge that 1mH inductor _can possibly_ deliver >> >> when it has 1A flowing through it depends on the particular inductor's >> >> winding resistance and possibly also on whether it's really a 1mH >> >> inductor when it has 1A flowing through it. >> >> >> You may want to pop over to the closest ASU campus that presents EEE 202 >> >> and see if you can audit the course. �This problem is no great mystery >> >> for someone who's gotten through sophomore electronics engineering. >> >> >> -- >> >> >> Tim Wescott >> >> Wescott Design Serviceshttp://www.wescottdesign.com >> >> >> Do you need to implement control loops in software? >> >> "Applied Control Theory for Embedded Systems" was written for you. >> >> See details athttp://www.wescottdesign.com/actfes/actfes.html >> >> >Hi Tim, �I agree with your calculation. �But not the interpretation. >> >Sure you integrate current over time and you get charge. �But this is >> >not the charge delivered to a resistor, it is how much charge flowed >> >through it. (Oh unless that's what is meant by delivered.) >> >> >George H. >> >> That's a common way of stating it. When we put parts on boards, their >> electrostatic potential is not often of concern. So we say that we put >> charge into a capacitor or a battery, literally we say "charge a >> battery" or "charge a capacitor" as opposed to "run charge through a >> capacitor", and we measure how much in ampere-seconds, namely >> coulombs. >> >> Since both a cap and a battery save the ampere-seconds and can >> return/deliver them later, it's reasonable to think that they stored >> charge. >> >> The numbers work. Engineering is about what works. >> >> John- Hide quoted text - >> >> - Show quoted text - > >Yeah sure, I use the same words. But isn't this the cause of the >current .... 'confusion'? > >NPI > >It's kinda like the water in a hose analogy of current in circuits. >It mostly works... but you can't spray charge out the end of a >circuit, you've gotta have a conductor attached. (let's assume no >high voltages.) > >We really only measure voltages and currents, we all know there are >charges moving around, but as soon as you try and get them to hold >still they disappear. (Well you can use a Faraday bucket.) > >I've been looking at flux leakage with our Keithley 601B >electrometer. At the 10^11 ohms scale (with the multiplier set at X10 >or X30) I can't walk near the circuit with out upsetting things. > >George H. I have a 610C ftp://jjlarkin.lmi.net/Keithley_1gig.JPG On the 1e-14 amp range, with just an open Pomona plug as an antenna, I can shuffle my feet on the carpet 10 feet away and pin the meter. Beautiful gadget. John
From: George Herold on 22 Jul 2010 10:02
On Jul 21, 11:12 pm, John Larkin <jjlar...(a)highNOTlandTHIStechnologyPART.com> wrote: > On Wed, 21 Jul 2010 18:41:50 -0700 (PDT), George Herold > > > > > > <gher...(a)teachspin.com> wrote: > >On Jul 21, 12:52 pm, John Larkin > ><jjlar...(a)highNOTlandTHIStechnologyPART.com> wrote: > >> On Wed, 21 Jul 2010 07:28:10 -0700 (PDT), George Herold > > >> <gher...(a)teachspin.com> wrote: > >> >On Jul 20, 8:32 pm, Tim Wescott <t...(a)seemywebsite.com> wrote: > >> >> On 07/20/2010 09:32 AM, Jim Thompson wrote: > > >> >> > On Tue, 20 Jul 2010 08:53:22 -0700, Tim Wescott<t...(a)seemywebsite..com> > >> >> > wrote: > > >> >> >> On 07/20/2010 08:24 AM, Jim Thompson wrote: > >> >> >>> Charge Conservation - Hint of the Day: > > >> >> >>> How many Coulombs can a 1mH inductor charged to 1A deliver? > > >> >> >> That's insufficient information, and I rather expect that you know it. > > >> >> > No. It's provided to cause young bucks to do some thinking. Looks > >> >> > like it didn't work with you :-( > > >> >> > (Except that it did annoy Larkin, yet again... so a partial success > >> >> > :-) > > >> >> Since you didn't answer I have to assume that you couldn't. > > >> >> Either this is a trick question, and the answer is "however many excess > >> >> electrons it has sitting on it when I hand it to you", or the answer is > >> >> "that depends on the coil resistance". > > >> >> A 1mH superconducting inductor with 1A will deliver (or flow, if you > >> >> want to quibble about the common EE definition of "deliver") an infinite > >> >> charge to a dead short, assuming all conductors are also zero resistance. > > >> >> Otherwise a 1mH inductor that sees R ohms of total circuit resistance in > >> >> the inductor and the load (charge target?) will see it's current decay > >> >> as (1A)*e^-(R/L)*t; this will integrate to (1A) * (L/R). So for 1 ohm > >> >> total resistance that'd be 1mC, for a 10 ohm total resistance that'd be > >> >> 100uC, for a 0.1 ohm total resistance it'd be 10mC, etc. > > >> >> Answers involving loads that aren't purely resistive are more > >> >> complicated, but still obvious if you can understand the above. > > >> >> But to answer how much charge that 1mH inductor _can possibly_ deliver > >> >> when it has 1A flowing through it depends on the particular inductor's > >> >> winding resistance and possibly also on whether it's really a 1mH > >> >> inductor when it has 1A flowing through it. > > >> >> You may want to pop over to the closest ASU campus that presents EEE 202 > >> >> and see if you can audit the course. This problem is no great mystery > >> >> for someone who's gotten through sophomore electronics engineering. > > >> >> -- > > >> >> Tim Wescott > >> >> Wescott Design Serviceshttp://www.wescottdesign.com > > >> >> Do you need to implement control loops in software? > >> >> "Applied Control Theory for Embedded Systems" was written for you. > >> >> See details athttp://www.wescottdesign.com/actfes/actfes.html > > >> >Hi Tim, I agree with your calculation. But not the interpretation. > >> >Sure you integrate current over time and you get charge. But this is > >> >not the charge delivered to a resistor, it is how much charge flowed > >> >through it. (Oh unless that's what is meant by delivered.) > > >> >George H. > > >> That's a common way of stating it. When we put parts on boards, their > >> electrostatic potential is not often of concern. So we say that we put > >> charge into a capacitor or a battery, literally we say "charge a > >> battery" or "charge a capacitor" as opposed to "run charge through a > >> capacitor", and we measure how much in ampere-seconds, namely > >> coulombs. > > >> Since both a cap and a battery save the ampere-seconds and can > >> return/deliver them later, it's reasonable to think that they stored > >> charge. > > >> The numbers work. Engineering is about what works. > > >> John- Hide quoted text - > > >> - Show quoted text - > > >Yeah sure, I use the same words. But isn't this the cause of the > >current .... 'confusion'? > > >NPI > > >It's kinda like the water in a hose analogy of current in circuits. > >It mostly works... but you can't spray charge out the end of a > >circuit, you've gotta have a conductor attached. (let's assume no > >high voltages.) > > >We really only measure voltages and currents, we all know there are > >charges moving around, but as soon as you try and get them to hold > >still they disappear. (Well you can use a Faraday bucket.) > > >I've been looking at flux leakage with our Keithley 601B > >electrometer. At the 10^11 ohms scale (with the multiplier set at X10 > >or X30) I can't walk near the circuit with out upsetting things. > > >George H. > > I have a 610C > > ftp://jjlarkin.lmi.net/Keithley_1gig.JPG > > On the 1e-14 amp range, with just an open Pomona plug as an antenna, I > can shuffle my feet on the carpet 10 feet away and pin the meter. > > Beautiful gadget. > > John- Hide quoted text - > > - Show quoted text - Opps, the number is 610B. (I'm always getting the numbers confused.) I've only got a 1 Gohm resistor to check the 'calibration' with. But it seems to work great. I'll post a flux report over on SEB. (bottom line, all rosin based fluxes seem to work fine for high impedance circuits. ) George H. |