2

There is a pulse going through a transmission line. There is a phenomena where if we put a pulse through a transmission line then after it reaches the IC we can have ringing instead of good pulse. In LTSPICE simulation connector is represented as pulse voltage source of 6V with resistor in series.

Transmission line is represented as RLCG. The load is 2 Ohm because 6V/3A=2Ohm. As you can see i could not get oscillations from this transmission line. I was expecting to see a picture as shown in the end, instead I got in the plot saw shapes which very stable response.

Where did I go wrong with this simulation?

[![enter image description here][1]][1]

[![enter image description here][2]][2]

Wikipedia

enter image description here

Eric Bogatin book on signal integrity

Editied:i removed photos from books and wikipedia because i still need to learn how to properly give credit.

Josh23
  • 1
  • 6
  • 1
    Josh23 - Hi, As you're new here, please see the [tour] & [help] for the main site rules. || Please note that if a post copies or adapts content (e.g. text, image, photo etc.) from elsewhere, that content must be correctly referenced. As a minimum, for online material, the original source webpage / PDF / video etc. should be named & linked (see that rule about references for books / articles etc.). Therefore please [edit] your question to name & link the original source of each copied/adapted image (and remember it's your responsibility to do that in future). Thanks. – SamGibson Nov 24 '23 at 19:08
  • Hello Sam ,i have tried to add description to the photos but it ruined the display of them .so i deleted them leaving only LTSPICE simulation photo.Could you please give me a link on a manual how to post photos and present links together on this fiorum? – Josh23 Nov 25 '23 at 05:57
  • Give that a look. Text between references is ignored, or breaks the references (embedded links). Put the text with the respective images. I'm not sure if I placed it appropriately; please review for accuracy. – Tim Williams Nov 25 '23 at 06:34
  • Question: where does 6V/3A = 2Ohm come from? This relation is not an explanation; it's just substituting one arbitrary number with two arbitrary numbers. Where did L and C come from, can you show your calculations for them as well? – Tim Williams Nov 25 '23 at 06:37
  • @Josh23 - Hi (a) Pages about SE editing help include here, and here from the FAQ. (b) I see you're getting help from another site member, so I won't get involved unless needed. (c) It's not enough to say that an image source was, for example, "Wikipedia". The source name should give a page title e.g. "Wikipedia - Resistors" & if the source is online, give the page link. If the source is a printed book then see the site rule I linked before. It explains about a citation's contents. TY – SamGibson Nov 25 '23 at 06:57
  • Hello Tim, Yes its a very important thing. I have a pulse of 0 to 6V going to an RF power amplifiers drain. The drain current of the amplifier is 3A under this pulse,so i assumed my amplifier could be represented as 2Ohm load to the PDN. is it a good assumption? Thanks. – Josh23 Nov 25 '23 at 11:57
  • I suspect several things wrong here. You might want to start a new question, describing exactly what you are doing to start with. – Tim Williams Nov 25 '23 at 13:26

1 Answers1

1

Where did I go wrong with this simulation?

You lumped all the cable's distributed capacitance and inductance into two components and that won't cut the mustard when looking at discontinuity reflections. They are called reflections btw and, although they look like ringing, they are not: -

enter image description here

Also, you appear to be expecting a 50 Ω cable but, the characteristic impedance is more like 316 Ω. You can of course you LTspice's built in transmission line but, don't ask me where it is because I don't use LTspice.

You also need to put links in your question for the images so you can credit them.

Ringing or reflections

Here is a simulated 1 metre, 50 Ω transmission line driven by a 5 MHz squarewave having a zero ohm source impedance. The cable has a 300 Ω termination: -

enter image description here enter image description here

Those effects on Vout are not ringing but reflections traveling back and forth down the cable summing with each other positively and negatively. And, this isn't some theoretical effect either. I was shown this at college in the 1980s where a prof set up an experiment with a few metres of coax just so that we didn't confuse ringing (an LC filter for example) with the reality about reflections.

Here's what happens when the load is significantly lower (10 Ω) than the characteristic impedance of the cable: -

enter image description here

Andy aka
  • 456,226
  • 28
  • 367
  • 807
  • Hello Andy,could you please recommend me a configuration in which I would see the desired effect of ringing for the input pulse? Thanks. – Josh23 Nov 24 '23 at 16:55
  • It's not called ringing (as mentioned in my quite short answer). What do you mean by configuration and, have you looked for the correct model in LTspice? – Andy aka Nov 24 '23 at 16:58
  • Hello Andy,i mean to have overshoot and under shoot oscillations as in the photo . by configuration i mean a transmission line RLCG model which would generation such phenomena when a pulse going threw it. I searched for it in google but i didnt find.could you please recommend me some thing which could help me with this? – Josh23 Nov 24 '23 at 17:05
  • 1
    When there are reflections at source and load, it does produce a ringing behavior and you will find plenty of authoritative sources that name it that way. Even if you personally disagree with the term, telling OP the term is wrong will only confuse them when they're reading material from other sources. – The Photon Nov 24 '23 at 17:45
  • 2
    For example here is Eric Bogatin discussing "ringing from unterminated lines". Here is Howard Johnson saying that "Terminations exist to control ringing". – The Photon Nov 24 '23 at 17:50
  • @ThePhoton surely you don't expect me to dumb-down my answer to make it easier for the OP to search for answers. Wikipedia (on Reflections of signals on conducting lines) never mention ringing as a counter-argument. So, just because you find two sources that call it ringing, does not mean it is. A closer look at the waveforms with a decent fast oscilloscope would show a series of step events. – Andy aka Nov 24 '23 at 18:58
  • @Josh23 (as per advice in my answer) you have not named and linked the sources of those images so I won't comment on them until you do. This is a site policy regarding giving credit to other material used btw. – Andy aka Nov 24 '23 at 19:00
  • 2
    @Andyaka, 1. Bogatin and Johnson are two of the most authoritative figures in the field of signal integrity. They're the authoritative authors Wikipedia should be citing for appropriate articles. 2. It's not a matter of one or the other: reflections are the cause, ringing is the effect. 3. Just because the ring signal has harmonics doesn't mean it's not a ring signal. And it will never be perfect step events because the (real-world) signal doesn't have infinite bandwidth. – The Photon Nov 24 '23 at 22:17
  • @ThePhoton calling it ringing is doing a disservice to the fundamental things that are happening. I'd never heard of them guys before so I have no problem calling them for using that term except that I'm not calling them; you brought them up so maybe you can justify why it's called ringing and leave an answer. – Andy aka Nov 24 '23 at 22:32
  • Anyway if you like WIkipedia as a source, here is an article that identifies "ringing" as one of the main issues in signal integrity (last paragraph of the lede), and calls for "using terminations to control ringing" as one of the methods of improving signal integrity (section "Fixing signal integrity problems") – The Photon Nov 24 '23 at 22:34
  • Also notice the first two references in that article are Johnson and Graham, and Bogatin. And no, I've never edited that Wiki article. – The Photon Nov 24 '23 at 22:37
  • 3
    @Andy aka - Really??? You have 442,000 reputation points, opine all the time on SI issues, yet you claim to have never heard of Drs. Eric Bogatin and Howard Johnson? How 'out Lee Ritchie (The Speeding Edge)? All acknowledged experts (as The Photon said) in SI matters. And all use the term "ringing" to describe the effects discussed here. – SteveSh Nov 25 '23 at 01:32
  • Real systems that "ring", like literally "ringing a bell", also exhibit transmission-line effects to various degrees; I have absolutely no problem calling it "ringing" in this case. I would still call it "ringing" even if it weren't simple harmonic motion (e.g. nonlinear oscillation of ceramic capacitor inrush, overshoot in a switching converter). – Tim Williams Nov 25 '23 at 06:15
  • Hello Andy, This is a good example .I am still learning the syntax of formatting ,so I have deleted the photos without credit till I learn how to do that. Could you please give a link on how to post photos in the comment section? Also I recreated your simulation and when i changed source resistor from 0 Ohms with 50 ohms then i got overdamped response instead of overshoot. I know its a sum of reflected waves and we need reflection coefficient calculation. Is there sum mathematical logic could explain why by changing source resistor from 0 to 50 we get a overdamped response? Thanks. – Josh23 Nov 25 '23 at 07:33
  • @Josh23 calling it an overdamped response is the wrong steer on what might be happening. The term overdamped applies to filtering and there is no filtering going on here. Echoes or reverberation are the equivalent terms in acoustics. Anyway, to leave an image in a comment... Pretend you are making an answer and put the image in that answer but, don't press the button to make the answer visible/live. Instead copy the url of the image and use that in the comments to embed that image. Then delete the answer you were creating (the image will still be stored and usable). – Andy aka Nov 25 '23 at 09:17
  • @SteveSh Never heard of him either. Let me be clear about this so I don't get pointless repeat comments of amazement from anyone: I had never heard of Bogatin, Johnson or Ritchie until they were mentioned in comments in these last few hours. Maybe they are more popular in the US or maybe you and photon have heard of them because your schooling didn't explicitly cover signaling and communications (1980s I did mine)? – Andy aka Nov 25 '23 at 09:23
  • @ThePhoton I'd like to make a quotation from Eric Bogatin that I just found: Reflections and distortions from impedance discontinuities give rise to a degradation in signal quality. In some cases, this looks like <-- in other words he says it looks like* ringing. This was from a possibly illegal on-line source at the start of chapter 8 (Transmission lines and reflections). Here's an image in case you don't believe me: https://i.stack.imgur.com/Zvgbh.png <-- The book is signal and power integrity simplified 2nd edition. – Andy aka Nov 25 '23 at 09:43
  • He then goes on to refer to these things explicitly as ringing and, to me that is lazy and contradictory. – Andy aka Nov 25 '23 at 09:58
  • @Andyaka "The term overdamped applies to filtering and there is no filtering going on here" -- are you sure? If you check the AC steady-state analysis, it might show interesting results. – Tim Williams Nov 25 '23 at 10:51
  • @TimWilliams feel free to leave an answer that shows where the filtering is taking place and, add a few words that show that it is a dominant force in producing the effects the OP is alluding to. I can be clear about my simulation and state that filtering is not taking place. – Andy aka Nov 25 '23 at 11:36
  • Hello Andy, Two simulations were performed. The first circuit creates an overshoot while the second doesnt have an overshoot. I understand that the result is a sum of reflected waves going back and forth. Is there some theory in which I could predict if my pulse on the load will be with over shoot or not? Thank. enter image description here – Josh23 Nov 25 '23 at 11:36
  • @Josh23 if the load resistor value is less than the characteristic impedance of the cable then there will be a staircase ramp up and down. If the load resistor is higher than the characteristic impedance it will be as per the image in my answer. If we are done here, please take note of this: What should I do when someone answers my question. If you are still confused about something then leave a comment to request further clarification. However, transmission line theory is a massive subject and cannot be delivered in one Q and A. – Andy aka Nov 25 '23 at 11:41
  • Hello Andy, I voted for your very helpfull responce.Could you please give me a stackexchnge scripting manual so my questions and comments will be well written? – Josh23 Nov 25 '23 at 11:48
  • @Josh23 thanks for that but, I'm not sure what scripting manual you might refer to. Is there some feature you've seen on SE that particularly interests you? – Andy aka Nov 25 '23 at 11:50
  • 1
    @Andy aka - I finished my university level schooling in the mid 70's. Signal Integrity was not much of a concern back then. And I never took an SI or related course. All my knowledge and experience with SI has been OJT (On The Job) training. I only came across Bogatin, Johnnson, Ritchie, and others around the turn of the century when SI became an issue with the increases in IC and system level clock rates. – SteveSh Nov 25 '23 at 11:51
  • @SteveSh I came at it from a different direction; a mixture of radio transmission (at work) and my college's degree course. In the late 80s RS485 also came into the mix of learning experiences re t-line theory. – Andy aka Nov 25 '23 at 11:55
  • @ThePhoton There is a repeated error in the link (image) http://www.signalintegrity.com/Pubs/edn/bothendsterm.htm . One must add at the midden "transition" 2 factors (K12, from left to the right ... and K21 from the right to the left ... and these are not necessarily equal). – Antonio51 Nov 25 '23 at 15:42