Activity for a concerned citizenâ€
| Type | On... | Excerpt | Status | Date |
|---|---|---|---|---|
| Comment | Post #287366 |
@#53497 Admittedly, seeing this for the first time without any prior knowledge might seem harsh but, if you would see all this user's history here (preferably in chronological order), you will get a better feeling of why Olin replied as he did (and why things evolved to this point). What I mean is, t... (more) |
— | over 1 year ago |
| Comment | Post #287232 |
@#52987 No, you're right, that sounds better. Bad wording from my part (but good intentions). (more) |
— | over 1 year ago |
| Comment | Post #287232 |
@#52987 Oh, forgot to say: k is the attenuation, related to a unity input. Maybe I should have used A, instead. I'll replace it now. (more) |
— | over 1 year ago |
| Comment | Post #287232 |
I searched online and found [this site](https://circuitsgeek.com/tutorials/t-pad-attenuator/) but, while the formulas are much nicer (no derivation) and the values for the resistors come out as they adveertise it, it doesn't seem to work as intended in terms of attenuation (I get ~51.4 mV, as opposed... (more) |
— | over 1 year ago |
| Comment | Post #287027 |
The schematic you shared uses a different symbol for the `ADA4530-1` -- the symbol appears disconnected. While I could correct the connections since they are fairly straightforward, this should have been done by you: either provide the custom symbol, or make a mention of it (preferably the former). A... (more) |
— | over 1 year ago |
| Comment | Post #286868 |
@#53586 Thank you for returning the favour. You had a nice "back of the envelope" approach. (more) |
— | over 1 year ago |
| Comment | Post #286836 |
I don't see any schematics. (more) |
— | over 1 year ago |
| Comment | Post #286784 |
@#53497 You're right, my words on the previous reply are a little bit ambiguous. I've modified the answer to include the distinction between the choice made for analysis, and the real life case. (BTW, `@` here needs typing the first three letters, no spaces, then the autocomplete should come up; fort... (more) |
— | almost 2 years ago |
| Comment | Post #286792 |
Are you doing it on purpose: not using a schematic editor? I think I was not the only one who told you to use one, so by now it looks like you simply "know better", or you are deliberately not listening. For my part, as long as you post doodles, I won't bother reading your questions. Even if you are ... (more) |
— | almost 2 years ago |
| Comment | Post #286784 |
@#54107 Yes but, as I said, their value is chosen such that it doesn't influence the overall oscillator -- after all, you don't want to have a polarizing network that, wehn added, changes the frequency of the oscillations, do you? Therefore, *for the purpose of analysis*, you can ignore them since, i... (more) |
— | almost 2 years ago |
| Comment | Post #286784 |
@#53497 You're right, that was a typo, I'll correct it. I was thinking in terms of L1=L2, so I'm probably guilty of "thinking with premeditation". About the input resistance: yes, I've omitted them because they're mostly part of the biasing circuit, and have little to do with the oscillation. They do... (more) |
— | almost 2 years ago |
| Comment | Post #286766 |
I don't have Altium, or OrCAD, but I think what OrCAD does with those "aliases" is it simply makes copies of them behind your back, either with the symbols or with the info, only, and maybe displaying them transparently to the user. It doesn't make sense otherwise, it has to have some separate info f... (more) |
— | almost 2 years ago |
| Comment | Post #286765 |
@#54107 And how is that working for you? The whole pronciple of an oscillator is based on the negative resistance. You need to approximate that, too, with another slope, and then the initial slope, too, if you want your oscillations to start. And you'll end up with a PWL *approximation* which will gi... (more) |
— | almost 2 years ago |
| Comment | Post #286730 |
It sounds like an [XY problem](https://xyproblem.info/#). What is it for? (more) |
— | almost 2 years ago |
| Comment | Post #286723 |
Olin, if you're reading [this](https://electrical.codidact.com/comments/thread/6474#comment-17851), could you please use it as a reference to force this user to explain every future question, especially when questionable content is asked? I'm not saying to ban the person but, given this blatant abuse... (more) |
— | almost 2 years ago |
| Comment | Post #286724 |
@#54107 I see. So you are using volunteers' time for your fun. That is good to know. (more) |
— | almost 2 years ago |
| Comment | Post #286724 |
At least now will you say what it is for? (more) |
— | almost 2 years ago |
| Comment | Post #286721 |
@#54107 I meant about the 10 kHz part because, as Olin says in his answer, that's not a characteristic of a bandpass filter. But you can define it in terms of center frequency and bandwidth (and attenuations in both pass- and stop-band). You haven't answered, though: what purpose will that filter ser... (more) |
— | almost 2 years ago |
| Comment | Post #286721 |
Did you mean "center frequency"? If so, even for a 20 dB attenuation at 1000 times the bandwidth in the stopband, I run into numerical issues and can't get such a high order, and thi sis for a Cauer/elliptic filter. You could try a 2nd order, but that would mean a Q of f0/BW=5000. You'll likely need ... (more) |
— | almost 2 years ago |
| Comment | Post #286700 |
@#54107 I don't understand. You have several people telling you that your view is not correct and they all bring arguments, yet you insist on your view. Now, I'm just a random nickname, but Olin is well known. If you think he isn't, take some time to read some of his answers here, or on ee.se. The wa... (more) |
— | almost 2 years ago |
| Comment | Post #286700 |
@#54107 "*For different input signals the frequency of the pole will have different values ,the pole as a pole exists for some specific values of L,C s but the frequency of the pole changes.*" -- This is false. For a transfer function of $1/(s+1)$ the pole is fixed, *at all times*, at $-1$. Otherwise... (more) |
— | almost 2 years ago |
| Comment | Post #286680 |
@#54107s Also, what do you mean with "*single phase* transmission line"? Are you thinking of a mains application? You should always say what it is you are after, not what you think you need (that's an [XY problem](https://xyproblem.info/#)), because there may be a different way to look at the problem... (more) |
— | almost 2 years ago |
| Comment | Post #286687 |
@#54107 Frankly, it sounds quite dishonest for you to ask about the design of the filter, then say it's only for "educational purposes". It's like me, asking for red fruits which taste like avocado because I'm interested in tasing one, only to say after people searched for me that "oh, I was just rea... (more) |
— | almost 2 years ago |
| Comment | Post #286680 |
You can still transmit signals, but they will be attenuated (no such thing as a brick wall filter). Have you seen [this page](https://prc68.com/I/Zo.shtml)? (more) |
— | almost 2 years ago |
| Comment | Post #286181 |
To improve a bit on the minimum sample time for accuracy, a buffer (simple repeater) could be added, which will eliminate the 4.7k. It will also add cost, which might not be justified, since the timings are already relaxed. (more) |
— | about 2 years ago |
| Comment | Post #286114 |
@#54288 Just so you know, the formulas that you gave imply a *positive* voltage output, but for the classical topology (non-isolated) the first formula you gave was the correct one. The second one is valid for flyback topologies (or for the awkward case of the classical one with negative input, or sh... (more) |
— | about 2 years ago |
| Comment | Post #286103 |
Can your load take the noise from a switched mode solution? If the load is mostly static (e.g. does not vary, or does not have fast transitions) you can oversize the output filter at the cost of bandwidth, to decrease noise. Otherwise, for 36 V max drop witn 0.1 A you'll have 3.6 W wasted, whereas wi... (more) |
— | about 2 years ago |
| Comment | Post #285861 |
@#53110 Well, if it was an artefact then it's on the owner of the video, feel free to ask him about that. I happen agree with the outcome, but that's just me. (more) |
— | about 2 years ago |
| Comment | Post #285861 |
@#53110 The whole problem was formulated in an ideal manner, so the lamp would glow instantaneously. It no longer matters that the "bulk" of the current would only appear after 1 s due to the transmission line effect. And it will not start glowing with a brief pulse, then wait for the rest, because t... (more) |
— | about 2 years ago |
| Comment | Post #285861 |
Have you seen this video? https://youtu.be/2Vrhk5OjBP8 (more) |
— | about 2 years ago |
| Comment | Post #285792 |
@#54288 I think the author may be saying something else, and that is dependent on different details that are not shown (I don't have the book). But, clearly, any transfer function can be plotted by replacing s->jw, since it's basic complex arithmetic. It's probably still related to the integral part ... (more) |
— | over 2 years ago |
| Comment | Post #285687 |
Unless I'm reading it wrong, the datasheet seems to specify the output with a nominal 10k/10p (general data table > output > load nom.). They could probably be around those values. [edit] The corner frequency for those values is certainly less than the maximum 45 MHz, though. (more) |
— | over 2 years ago |
| Comment | Post #284895 |
@#54668 The picture above is made with LTspice. It should be easy to replicate. (more) |
— | over 2 years ago |
| Comment | Post #284540 |
I don't know the term, but it seems that "switching transients" is a more generic term that can encompass "transient power". (more) |
— | over 2 years ago |
| Comment | Post #283926 |
@#54107s A current source cannot exist as opened (no load), in the same way a voltage source cannot exist as shorted out. The first will want to generate the specified current and since the impedance is infinite the voltage is infinite; similarly, a voltage source's internal resistance is zero, thus ... (more) |
— | over 2 years ago |
| Comment | Post #283265 |
Logisim has an option for the XOR gates in their properties, regarding `Multiple-Input Behaviour`, but it's either the "one true only", or "an even number of true". (more) |
— | over 2 years ago |
| Comment | Post #282635 |
@#54107 After the switch changes state, the only mesh you have is made of C, R, and L, with C precharged. Therefore there is no Vs, only Vc=Vr+VL, because Vc takes the role of Vs. With Vr=0, all that remains is VL. (more) |
— | almost 3 years ago |
| Comment | Post #282635 |
@#54107 Draw an imaginary vertical line separating `C1` from `R2+L1`. Now the voltage on the capacitor is mirrored to the inductor and the resistor. Since the resistor is not a reactive element, the whole initial condition applies now to the inductor, only, therefore `D` ends up with `2`, not `-2`. (more) |
— | almost 3 years ago |
| Comment | Post #282205 |
@Chupacabras I've updated the answer, maybe it clears up a few things. (more) |
— | almost 3 years ago |
| Comment | Post #280357 |
If it were me I'd give up the reputation points and only keep the relevant characteristics, such as the ones that already are. Reputation has the bad tendency that people will want to post for the points, rather than for the contents (just look at *.se). Having a counter that says "guru in this domai... (more) |
— | almost 3 years ago |
| Comment | Post #280152 |
Late comment, but that's a nice trick. A 2nd order Bessel might do just a tad better, though at the cost of an opamp. (more) |
— | almost 3 years ago |
| Comment | Post #280822 |
@PeteW I've also seen it once, during a storm. The wires failed locally, where the event happened, then the humidity and the aged wires (some from the 60's, 70's) acted as a welding electrode: the isolation burned and continued to burn down the wire. It was faster than what I see in OP's video, but I... (more) |
— | about 3 years ago |
| Comment | Post #279005 |
Did the edit not work? (more) |
— | over 3 years ago |
| Comment | Post #278756 |
@coquelicot I know that the practical part is what matters here, but the theoretical part might still help along the way. (more) |
— | over 3 years ago |
| Comment | Post #278756 |
The MathJax seems off? `\\` doesn't work, and nither does `\begin{align} ... end{align}`. Are there differences between this and the syntax used in ee.se? (more) |
— | over 3 years ago |
| Comment | Post #278614 |
@LvW There are people who prefer Boctor or Deliyannis because, at a first glance, you "only" need one opamp per stage. But when you consider the side effects, really, it really pays to add one more for an almost true zero (and not only). (more) |
— | over 3 years ago |
| Comment | Post #278614 |
I've updated my answer. Be sure to not omit the last paragraph. (more) |
— | over 3 years ago |
| Comment | Post #278614 |
Before I continue, you do realize that you'll get something close to (maybe a bit worse) the black trace in the last picture? If you look closely, you'll see that the passband is no longer flat(tish), it has a droop of maybe 3 dB, which means measuring a signal at 90 Hz and one at 100 Hz will give tw... (more) |
— | over 3 years ago |