Consul
Well-known member
... are a major pain in the rear end.
That is all.
That is all.
Kirchoff's Laws...
- Thread starter Consul
- Start date
Help Support GroupDIY Audio Forum:
Svart
Well-known member
or an absolutely beautiful thing if you work with touch sensors like I do..
Consul
Well-known member
I think it's just a matter of working out the kinks in my understanding. That doesn't make my homework any easier right this minute, though.
SSLtech
Well-known member
...Cherry picking from the two above quotes in order to try and create a single unified summary:
...are a major thing, if you work with -or touch- an absolutely beautiful rear end, like I do.
Keith
Echo North
Well-known member
Glad to see you're feeling better Keith!
bcarso
Well-known member
Consul, you should take comfort that you're not required to be completely rigorous. I have long been amused by Paul Slepian's treatise, Mathematical Foundations of Network Analysis (Springer, 1968). I forget how many pages it takes, but at the end you find out that it only gets through resistors---no reactive components!
bcarso
Well-known member
SSLtech said:
I saw one the other day that, were one to bounce a quarter off it, would have given back at least 35 cents in change.
Svart
Well-known member
Priceless!
> Kirchoff's Laws...... are a major pain in the rear end.
No. A functional grasp of Mr K is THE way to know what's happening.
There are two roads onto my campus, Mill and George.
Assume this traffic is a "closed system"; All students arrive in the morning and leave in the evening, nobody sneaks out through the woods, no students die in the cafeteria, no cars are left behind.
Use unit "Kcars" = 1,000 cars.
Each morning George has 2Kcars and Mill has 3Kcars.
Ah, but one evening we close Mill. Bridge collapse. Pothole improvements. VIP event. Everybody has to leave on George.
What is the traffic on George?
This is equivalent to K's Current Law.
Note that you DO have to keep track of coming versus going. I, a campus rat, have a pretty clear notion of direction relative to campus. However in circuits the "direction" is arbitrary. And instead of "in/out" we mark it "+/-". That means you can use rudimentary Algebra to get a general solution before you know which way is which. The exit flow on George is, to my on-campus POV, a negative number.
No. A functional grasp of Mr K is THE way to know what's happening.
There are two roads onto my campus, Mill and George.
Assume this traffic is a "closed system"; All students arrive in the morning and leave in the evening, nobody sneaks out through the woods, no students die in the cafeteria, no cars are left behind.
Use unit "Kcars" = 1,000 cars.
Each morning George has 2Kcars and Mill has 3Kcars.
Ah, but one evening we close Mill. Bridge collapse. Pothole improvements. VIP event. Everybody has to leave on George.
What is the traffic on George?
This is equivalent to K's Current Law.
Note that you DO have to keep track of coming versus going. I, a campus rat, have a pretty clear notion of direction relative to campus. However in circuits the "direction" is arbitrary. And instead of "in/out" we mark it "+/-". That means you can use rudimentary Algebra to get a general solution before you know which way is which. The exit flow on George is, to my on-campus POV, a negative number.
bcarso
Well-known member
Svart said:
Not mine, but heard some years ago from my friend Marshall Smith (also ex-Harman). Of course I've been looking for an opportunity to use it (thanks Keith) ;D
Consul
Well-known member
Oh, I'm well aware that grasping these concepts is important, but that doesn't make my brain work any better. I'm going to catch my prof during office hours later today to see if I can get some further help.
MartyMart
Well-known member
SSLtech said:
Going out with a girl at school who had "the perfect rear end" totally screwed my O Level exams !!!
( got dumped a week before they started .... and was heart broken :-( )
I've never seen a more perfectly filled pair of jeans in my life, before or since BTW
MM.
Emperor-TK
Well-known member
PRR said:
I've eaten in that cafeteria. I think that's a big assumption. Perhaps we can compare the cafeteria/ambulance route to sloppy solder flux on a high impedance circuit. It isn't an obvious pathway for transport, but it shouldn't be discounted.
> I've eaten in that cafeteria. I think that's a big assumption.
K.I.S.S. Consul is so new-to-this, he thinks a Professor can help him. Same Prof who already confused him.
Buddy-up to an elder/wiser STUDENT... that is how good edjcashun happens. Profs live on a higher plane of existence. Even if they felt they "should" help, most know too much to be much help. Freshmen should kiss-up to Juniors or friendly Seniors. You will have to buy beer, take shirts to cleaners, etc... if you swing it right, in a few years some punk frosh will be buying your beers to get a few insights.
> I've eaten in that cafeteria.
It's not as bad as it used to be.
Like wet-cloth insulation, 1% of electrons or students "vanish". If it is a student, we usually try to count the deaths. In a circuit, you can measure chassis leakage. It is just another node; Kirchoff can account it.
A full model of campus, or a medium chip, is the same idea with 6,000 loops and a huge number of nodes. You can zoom-in and know that, over a day, the number of students leaving the cafeteria (on foot or by hearse) is equal to the number who entered. You can map-draw 2 loops on 2 nodes; when you get to many loops the drawing gets cluttered and we need to use Algebra.
I must confess, I never invoke Kirchoff explicitly. I K.I.S.S. The first step is to cast-off huge chunks of circuit that I'm not immediately interested in. Then I eye-ball the loops that "do not matter". A 1Meg meter shunting a 10K resistor, the meter-leakage is 1% and anything up to 10% error doesn't matter. With blinders on, most circuits I grok come down to few-enough nodes that I can map them with a pencil. Very commonly a voltage-source and three impedances in series/parallel. And that reduces to a loaded voltage-divider, something any practical circuit-bender should read at a glance.
However I do not claim EE credentials. Real EEs get hired for stuff I could not touch, stuff which might actually "require" application of Kirchoff because it is too tangled for pencil-maps. You need to know this stuff, to the point that you would not mention it. Professional EE papers rarely say "using Kirchoff..."; that woult be like a carpenter saying "using a saw...".
Kirchoff's Voltage Law: you build a split-level house in Kansas. Start at the driveway, go up 8 feet for garage, 8 more feet for bedroom. Come over to the split, drop down 4 feet to living room ceiling, 8 more feet for living room... and you are still 4 feet off the level ground! So you go down 8 feet for basement (throw 4 feet of hole-dirt up so the front door opens to ground), then come up 4 feet to get back to driveway level. The sum of the up-studs MUST be equal to the sum of the down-studs. Or draw a loop and apply a negative sign to the studs you find on the way down, and just say "sum of studs must be zero".
Or fly an airplane and return to your starting point. The algebraic sum of the climbs and descents WILL be zero. If you "stop flying" 10 feet short of the starting level, you will traverse that 10 feet and uncomfortably. If you attempt to drop 10 feet more than you climbed, you won't.
None of which helps my Dell laptop. It goes to 100% charge, drops to 80% after 20 minutes, and then goes to zero in a few more minutes. Kirchoff is in there, somewhere, but I don't have the stud-lengths to put into the problem.
K.I.S.S. Consul is so new-to-this, he thinks a Professor can help him. Same Prof who already confused him.
Buddy-up to an elder/wiser STUDENT... that is how good edjcashun happens. Profs live on a higher plane of existence. Even if they felt they "should" help, most know too much to be much help. Freshmen should kiss-up to Juniors or friendly Seniors. You will have to buy beer, take shirts to cleaners, etc... if you swing it right, in a few years some punk frosh will be buying your beers to get a few insights.
> I've eaten in that cafeteria.
It's not as bad as it used to be.
Like wet-cloth insulation, 1% of electrons or students "vanish". If it is a student, we usually try to count the deaths. In a circuit, you can measure chassis leakage. It is just another node; Kirchoff can account it.
A full model of campus, or a medium chip, is the same idea with 6,000 loops and a huge number of nodes. You can zoom-in and know that, over a day, the number of students leaving the cafeteria (on foot or by hearse) is equal to the number who entered. You can map-draw 2 loops on 2 nodes; when you get to many loops the drawing gets cluttered and we need to use Algebra.
I must confess, I never invoke Kirchoff explicitly. I K.I.S.S. The first step is to cast-off huge chunks of circuit that I'm not immediately interested in. Then I eye-ball the loops that "do not matter". A 1Meg meter shunting a 10K resistor, the meter-leakage is 1% and anything up to 10% error doesn't matter. With blinders on, most circuits I grok come down to few-enough nodes that I can map them with a pencil. Very commonly a voltage-source and three impedances in series/parallel. And that reduces to a loaded voltage-divider, something any practical circuit-bender should read at a glance.
However I do not claim EE credentials. Real EEs get hired for stuff I could not touch, stuff which might actually "require" application of Kirchoff because it is too tangled for pencil-maps. You need to know this stuff, to the point that you would not mention it. Professional EE papers rarely say "using Kirchoff..."; that woult be like a carpenter saying "using a saw...".
Kirchoff's Voltage Law: you build a split-level house in Kansas. Start at the driveway, go up 8 feet for garage, 8 more feet for bedroom. Come over to the split, drop down 4 feet to living room ceiling, 8 more feet for living room... and you are still 4 feet off the level ground! So you go down 8 feet for basement (throw 4 feet of hole-dirt up so the front door opens to ground), then come up 4 feet to get back to driveway level. The sum of the up-studs MUST be equal to the sum of the down-studs. Or draw a loop and apply a negative sign to the studs you find on the way down, and just say "sum of studs must be zero".
Or fly an airplane and return to your starting point. The algebraic sum of the climbs and descents WILL be zero. If you "stop flying" 10 feet short of the starting level, you will traverse that 10 feet and uncomfortably. If you attempt to drop 10 feet more than you climbed, you won't.
None of which helps my Dell laptop. It goes to 100% charge, drops to 80% after 20 minutes, and then goes to zero in a few more minutes. Kirchoff is in there, somewhere, but I don't have the stud-lengths to put into the problem.
bcarso
Well-known member
One of the best presentations if I remember correctly: Ernst A. Guillemin, Introductory Circuit Theory , Wiley 1953 and later printings. OOP but readily available used.
Do you really need it? If you are taking a course and the prof. is determined to make you handle all sorts of bizarre topologies and floating sources etc., maybe it would help. For most electronics, IMO not really. Guillemin is insistent and adamant that everyone should know it at the level of detail presented, and he was known to be a great teacher IIRC.
If again I recall correctly, one of the intriguing things EAG proved was that you could realize a very much larger set of transfer functions that anyone had believed, if you could tolerate arbitrarily large amounts of loss. I forget where I read that, but I don't think it was in one of his books.
Do you know btw that you can get voltage gain out of an R-C (only) network? A good bar bet, if you have the misfortune of frequenting those sorts of bars.
I've also wondered if the famous math guy Victor Guillemin is related to Ernst. But although both have Wiki presence there's no mention of ancestral connection.
Do you really need it? If you are taking a course and the prof. is determined to make you handle all sorts of bizarre topologies and floating sources etc., maybe it would help. For most electronics, IMO not really. Guillemin is insistent and adamant that everyone should know it at the level of detail presented, and he was known to be a great teacher IIRC.
If again I recall correctly, one of the intriguing things EAG proved was that you could realize a very much larger set of transfer functions that anyone had believed, if you could tolerate arbitrarily large amounts of loss. I forget where I read that, but I don't think it was in one of his books.
Do you know btw that you can get voltage gain out of an R-C (only) network? A good bar bet, if you have the misfortune of frequenting those sorts of bars.
I've also wondered if the famous math guy Victor Guillemin is related to Ernst. But although both have Wiki presence there's no mention of ancestral connection.
bcarso
Well-known member
Two other good references found while looking for my copy of Guillemin: Desoer and Kuh, Basic Circuit Theory (very methodical and thorough, and beyond Kirchoff unusual for treating nonlinear and time-variant stuff in an introductory book) ISBN 070165750, and Van Valkenburg, Network Analysis, Prentice-Hall 1955.
Consul
Well-known member
PRR said:
I AM the elder/wiser student. :-\ This is a two-year school before I transfer to the big-time.
Besides, I'm starting to latch on to the concepts now.
It really is a simple as PRR is saying...
Take a circuit node for example. That is like a 3 way or 4 way traffic intersection in our hypothetical road... Assuming no cars stop and remain in the middle of the intersection, all cars leaving the intersection must add up to the number of cars that entered that intersection. No cars are created or lost.. # entering = # leaving.
Some text book descriptions may make this seem a little abstract but it really is quite physical and real.
JR
Take a circuit node for example. That is like a 3 way or 4 way traffic intersection in our hypothetical road... Assuming no cars stop and remain in the middle of the intersection, all cars leaving the intersection must add up to the number of cars that entered that intersection. No cars are created or lost.. # entering = # leaving.
Some text book descriptions may make this seem a little abstract but it really is quite physical and real.
JR
Consul
Well-known member
The big trouble in the class, I think, is how to apply such basic ideas to circuit analysis and end up with useful information. Like I said, I'm getting it now, but some of the other students...
Also, that's describing the Junction Rule. The Loop Rule is the one causing the troubles.
Also, that's describing the Junction Rule. The Loop Rule is the one causing the troubles.
Svart
Well-known member
I would say that real physics majors(as it applies to semiconductor design and such) can do stuff that I can't touch. But then again I don't really want to do their job either.. I'm happy just doing design work with their parts, I'll leave the real low level stuff to the ones who want to do it!
Besides, PRR, I would say that you continuously prove that you could work circles around most of the EEs I've worked with.
I had a couple professors that were really awesome but most were exactly as you describe. Even the awesome profs were a little too academic to really get into the details that we need.
I have to admit that I learned more from the first year on the job working under a couple of very friendly and supporting senior engineers than I did in any combination of classes from any of the schools I went to.
In fact, I have a harsh statement to put out in the open. Schools don't prepare you for the real world. They teach you how they want you to think and what they think is important. I tried to use what I learned in school and soon forgot most of it.
I think these 6 are the things that I have used the most:
Ohms law(every day!)
Algebra in all of its forms(every day!)
The ability to sort through datasheet marketing BS(every day!)
The ability to start seeing trends in designs(every day!)
SPICE and simulation tools(every day!)
Test equipment(every day!)
I think I used Diff eqs once too.
MASTER these things and you will never have any problems in the design world.
As for Kirchoff, I'm with PRR. Once you pass the test/course, you'll pretty much never hear anything else about Kirchoff unless you are dealing with something specifically using his work. You'll know it's there and you'll know what it's doing but it just won't be anything that you'll deal with.
Step back, forget about it for a couple days and then just start to think about what you've read. It'll come to you once you clear out the mind blocks caused by trying too hard.
