# Question on KVL

Discussion in 'Electronics Homework Help' started by chopnhack, May 29, 2016.

1. ### chopnhack

1,573
354
Apr 28, 2014
The video author mentions that when there is a current source in a loop, that loop automatically takes on that value. Is that true? I noticed that there was a voltage source in the loop and was somewhat confused since volt across a resistor would contribute to current! He says this at 1:07:15.

2. ### hevans1944Hop - AC8NS

4,608
2,151
Jun 21, 2012
I had to jump ahead in the video to see what you were asking about, but I did watch several minutes of the beginning. That was very painful. Although there were no errors (that I could see) in the subject matter, the voice-over presentation was absolutely awful and downright painful to listen to. The person speaking should have used a script, and perhaps edited and reviewed its contents, before committing to as a voice-over narration of a Power Point Presentation..

However, to answer your question, at this point (1:07:15) in the presentation the author is using KVL to write an expression for the voltage drops around loop I2. In doing so, he has to account for the current I2 as well as the current I3 through the 4 kΩ resistor to account for the voltage drop across the 4 kΩ resistor. And that's all that needs to be accounted for (in addition to the two voltage drops across the two 2 kΩ resistors) to write the KVL equation involving loop I2.

The voltage source to which you refer is not located in the loop under consideration for writing the KVL equation for loop I2. OTOH, current I3, which happens to pass through this voltage source, consists entirely of the current produced by the 5 mA constant-current source because that is the only current source in the I3 loop.

The contributions of the voltage source to the circuit analysis would be revealed if a KVL equation were written for loop I3, but unfortunately this was not demonstrated.

The main problem I have with this video is its presentation. It does show all the circuit analysis techniques you need to learn, but it does not tell you when and where to use any particular technique. At the end, the author simply suggests that you solve a lot of circuit analysis problems on your own (practice, practice, practice), and perhaps use LTSpice or a similar SPICE simulation program to check your work, and to play with values to find out their effect on circuits. Well, good luck with that. Perhaps eventually you will acquire an intuitive approach to circuit analysis. I have to agree with using LTSpice to check your problem-solving though. It's hard to find text books with problems and (more importantly) answers.

chopnhack likes this.
3. ### chopnhack

1,573
354
Apr 28, 2014
Imagine making it through the nearly two hours.... I cheated a bit and set the playback to 1.25X to get it to a more natural speed for me.

Thanks for clearing that up - when he said it, I was thinking that perhaps it was a rule to be applied for each situation, but I think I understand that it was a one of situation in this case because of the simplicity of the example.

I was willing to overlook that, my main issue was when he gets to setting up equations for KCL and totally skips the "clever, algebraic rearranging" (38:01).
Yes!

4. ### hevans1944Hop - AC8NS

4,608
2,151
Jun 21, 2012
I tried to download the Power Point Presentation, linked to in the YouTube site, and was informed that the website had been hacked and nastyware installed. So I abandoned that effort.

Is there anything in particular about Ohm's Law, Kirchhoff's KCL and KVL Laws, Thevenin and Norton Equivalents, and the Superposition Theorem that is giving you problems right now? I know the algebra can be confusing when trying to analyze complex circuit configurations, and I always fear that I will make a silly sign mistake and waste hours of time obtaining the wrong answer. But it really is straight forward.

The big advantage of LTSpice (besides its speed) is it can act as a sanity check for your circuit analysis. Or your circuit analysis can act as a sanity check for LTSpice. There is nothing that LTSpice does that you cannot do with a scientific calculator and some applied math. LTSpice just does the calculations faster. But you do have to make sure you provide LTSpice with an accurate model if the results are to be meaningful. And if doing a dynamic analyses, have some awareness of what time-steps are required to yield satisfactory results. The one thing I did like about the video presentation was the encouragement at the end to actually build circuits and measure what is going on.

5. ### chopnhack

1,573
354
Apr 28, 2014
I feel I have a good grasp of Ohm's laws now - unfortunately they only apply to linear, resistive circuits so that is only a small spectra of electronics. A good start, though
KVL/KCL - I need more time to practice this. As I said in a previous post, the concept is beautifully simple - conservation of charge, etc.. I will keep at it, I also have a copy of Schaum's Basic Circuit Analysis that I will be looking at tomorrow.

LTSpice is great, unless you feed it a situation without enough info - easy for a newb to do. You don't know what you don't know. I have done that a few times, not knowing that another parameter was needed.

meh, tomorrow is another day for the math. I am doing some layout on that last circuit we were discussing. Did you get a chance to review the schematic? I think I got it correct this time.
Did you make any progress on your toner/printer resurrection?

6. ### hevans1944Hop - AC8NS

4,608
2,151
Jun 21, 2012
I did review it, and thought I had left a reply on this thread, but it's not there. Schematic looks good except some components are labeled twice (capacitors) and some are not labeled at all (resistors). Not sure why you would want to place capacitors across the resistors in series with the two push-button switches. MCLR doesn't need it. Other switch can be (should be) de-bounced in software. Want to re-post the latest schematic in this thread?

Well, it's a Hewlett-Packard Laserjet III, and I think I remember the scan mirror failed sometime in the previous century. I haven't tried to turn it on.

Tried using the Canon laser copyier and discovered that some plastic-foam strip stuff, used as either a light-trap or maybe as a cushion, had disintegrated. Left a black smudge on the back of the paper I sent through it. And now it jams when I try to make a copy. The toner cartridge is dead. Shaking it didn't help. I am not sure it would be worth the money to replace the toner cartridge (assuming replacements are available) because there may be some more of that rubbery foam stuff falling apart inside.

I talked to my oldest son, who had journeyed this week-end from Atlanta to attend a funeral tomorrow in Cincinnati and decided to stop by Dayton to visit me... he said when he was a student at The Ohio State University working in their Physics Laboratory making PC boards, he used transparency film loaded in a laser printer to make "negatives" that he taped to photo-resist covered boards. After exposure to light, the boards were developed and etched. He said he never tried the toner transfer method of creating a resist pattern on un-coated boards.

However, I did look at some inexpensive laser printers, some for less than \$100, so I might try that unless I can find a source of pre-sensitized PC board stock at a reasonable price. Then I could try the transparency "negative" or maybe "positive" method (depending on which type of photo-resist the board has) using my high-resolution ink-jet printer to print on transparency film. Pretty sure I will need a mirror image of the artwork, so I can lay the side with the printing against the photo-resist for a contact exposure.

Not looking forward to this messy chemical process. The ferric chloride etch is bad enough, but IIRC the positive photo-resists use sodium hydroxide solution for the developer. I could apply positive photo-resist myself, maybe get some from my former employer if they haven't discarded their now-outdated supply of it. It was used many moons ago to coat semiconductor wafers, but I think this out-dated stock was also used a few years ago to make some superconducting Josephson junction arrays. The PhD who did that left UES, Inc. right after that when the contract ended.

Well, off to bed for now...

7. ### chopnhack

1,573
354
Apr 28, 2014
This seems pretty good! I have wanted to try this route too because I have heard the resolution is so much better, but I want to be able to build a UV meter so that I can check the output of UV leds before trying. That way I can quantify the UV energy and how long it takes. Interestingly enough this manufacturer gives you some good info, 5cm exposure distance using a 20w fluorescent bulb for 6-10 minutes.

8. ### hevans1944Hop - AC8NS

4,608
2,151
Jun 21, 2012
The boards you linked to are phenolic boards, which I would avoid because they absorb moisture. FR4 pre-sensitized boards here are more expensive but less problematical. Will continue looking for chemicals that I can use to coat my own boards. Mendelson's usually has a decent supply of inexpensive FR4 copper boards I can cut to size and coat myself. @(*steve*) purchased some while he was here. Maybe he can provide some feedback on how he is using them.

chopnhack likes this.
9. ### hevans1944Hop - AC8NS

4,608
2,151
Jun 21, 2012
As for exposure... I don't think a UV LED is necessary. I use a box with a #2 (IIRC) photoflood lamp inside, positioned about ten inches or so above the PCB. There is a squirrel-cage fan that brings in cooling air and circulates it around the PCB and lamp. Exposure time for negative photo-resist is in the neighborhood of fifteen minutes. There is a mechanical timer switch on the box to control exposure. I have an OttLite reading lamp as well as several 4-foot fluorescent light fixtures for shop lighting. Any one of these can probably be used to expose the positive-resist PCB. I also have a small UV lamp used to erase EPROMs but this is probably overkill. Might try it though: it's handheld.

10. ### chopnhack

1,573
354
Apr 28, 2014
That is cool, I will resist this method for now. It's definitely something on my to do list though.

11. ### (*steve*)¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥdModerator

25,491
2,833
Jan 21, 2010
Unfortunately they don't leave the US until July 10, and get here 3 months later.

12. ### hevans1944Hop - AC8NS

4,608
2,151
Jun 21, 2012
Yikes! So when they do finally arrive, are they "trade goods" to exchange with the natives, or are you going to etch some circuit traces on them? If traces there be, how do you plan to make that happen?

Uh, this is very important to KVL analysis... probably.

13. ### (*steve*)¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥdModerator

25,491
2,833
Jan 21, 2010
if you're interested in KVL and KCL then I can thoroughly recommend MITx 6.002.1x

Invest the hours in that and you'll come away with a pretty deep understanding of it. You'll also get to know about a heap of techniques that build on it.

And as for the various goods, they're for me and my hackerspace.

chopnhack likes this.
14. ### chopnhack

1,573
354
Apr 28, 2014
I did, but failed out about two years ago - very fast paced schedule to keep up to and I didn't have the requisite math (calculus) knowledge to pursue the next few weeks...