Talk:Planar transmission line

Dielectric Lines
The article includes image line in its catalogue of formats. This is consistent with the definition of planar lines as flat, ribbon-shaped lines in the lede. The scope of the article might reasonably have been restricted to those lines constructed using PCB technology, but if dielectric strips are to be included, then all variants (rib-line, NRD etc.) ought to be mentioned. I only raise this issue because the article is a good-article nominee, and coverage of the subject is a good-article criterion. --catslash (talk) 00:26, 16 February 2017 (UTC)


 * I'm not convinced that NRD can be described as planar whereas at least one of my sources explicitly lists imageline as planar. Numerous sources actually seem to make a distinction, e.g "The hybrid integration technique of planar and NRD-guide circuits for millimeter-wave applications".  I've no idea what rib-line is.  Do you think it is prevalent enough to justify a section of its own?  Feel free to add something if you have sources. SpinningSpark 12:39, 16 February 2017 (UTC)
 * Ah, here's a source that discusses rib guide (I assume that is the same thing as rib-line) and other dielectric waveguides.  Maybe I'll include these as a variants diagram. SpinningSpark 13:03, 16 February 2017 (UTC)

Citation format
I would like to upgrade the citations in this article to use the Harvard format and reflinks. It and WP:SFN. WP:CITEVAR requires that I ask first. Does anyone have an objection? 7&amp;6=thirteen (☎) 15:50, 6 March 2017 (UTC)
 * Yes, I would object. I would have done it that way in the first place if I thought that it was any better. SpinningSpark 22:35, 6 March 2017 (UTC)
 * It is better for the readers.
 * But you are entitled to your opinion and your liberum veto. <b style="color:#060">7&amp;6=thirteen</b> (<b style="color:#000">☎</b>) 22:51, 6 March 2017 (UTC)

Quasi-TEM modes - radiation
This section says that it is an inhomogeneous medium or the consequent quasi-TEM propagation that causes radiation at discontinuities. Of course it is true that microstrip (inhomogeneous) will radiate at discontinuities while stripline (homogeneous) will not. However, stripline can be made with different dielectrics above and below the strip, resulting in quasi-TEM propagation, but it cannot radiate. Furthermore, if the materials in this stripline are lossless, and if the line is nicely via-fenced so that energy cannot escape through other modes, then any discontinuities must present a purely reactive load (by conservation of energy). Conversely, discontinuities in microstip will radiate (a little), even if the substrate dielectric constant is unity.

The section also says that the resistive component of the impedance presented by a discontinuity results in radiation. Radiation is surely a logical consequence of the resistance (if the materials are lossless), but it is like saying that it is the heat in your brake-discs that slows you down - it violates common notions of causality. [Fixed] --catslash (talk) 00:03, 14 January 2018 (UTC)


 * I agree on the causality issue. The resistance is a measure of the radiation, not the cause of it.  What you say on inhomogeneous stripline makes sense, but I would like a reference before changing anything.  Modes that cause "sideways" radiation are a known issue in stripline and these too will show up as a resistive element to components.  Yes, perfectly fenced stripline cannot radiate.  Another case is boxed stripline, also completely shielded.  Perhaps the solution is to add the qualifier "...most quasi-TEM..." to indicate there are edge cases without making the modes section more of a tome than it already is. <b style="background:#FAFAD2;color:#C08000">Spinning</b><b style="color:#4840A0">Spark</b> 16:13, 14 January 2018 (UTC)

Stripline circuit
I just want to respond here to some of the points raised about the stripline microstrip circuit in the article at the peer review. This was the lead image, but is currently in the "General properties" section. As I said at the peer review, a detailed description of the minutia of the circuit would be inappropriate. Firstly, because it would be tangential to the article, and secondly because it was uploaded without a detailed description so adding such a description would be WP:OR. But for what it's worth, here is my understanding of the points raised. Hope that's helpful, <b style="background:#FAFAD2;color:#C08000">Spinning</b><b style="color:#4840A0">Spark</b> 18:05, 1 August 2018 (UTC)
 * "what are the light vs dark parts? Conductor vs substrate?" In this circuit, the conducting lines are darker than the insulating substrate.  The substrate is unusually bright probably because it is a ceramic.
 * "What parts are the planar transmission line and what parts are not?" The planar transmission lines are all the dark lines on the ceramic substrates. Everything else is either discrete components or not planar.
 * Are the screws important?" I don't believe the screws are important electrically. They are there merely to fix the parts together.  Rows of screws or holes often have something to do with screening in circuits such as this (see via fence) but in this case they are a result of the construction of the device.  The circuit is built of four separate pieces of ceramic substrate.  Underneath the ceramic is another board, probably a more traditional glass-epoxy board.  The ceramic substrates are fixed on to the underlying board with clips retained by the screws.  It was probably constructed this way because the device is a prototype or one-off.  Ceramic is difficult to drill.  It needs special drills and takes a long time.  Clips are more convenient for a prototype.   A production unit would probably be made on a single piece of ceramic with the necessary holes for fixing down.
 * "Why are wires mixed in with the planar patterns?" The only wires I'm seeing are in the top left-hand corner.  These are providing connections between the planar lines and external connections (mostly power connections, I think).  The only other wires I can see are the windings of two very small choke inductors near the transistor.
 * "What is the size of the wave relative to the size of planar elements?" The thick line in the centre shaped like half a swastika is a stub bandpass filter.  Each arm is likely a half wavelength (since the ends are open circuit rather than short circuit).  The absolute size of the wave is not possible to determine without either a scale reference in the picture or knowledge of the operating frequency.  However, the whole device is going to be tens of centimetres across.


 * Could the underlying embossed surface be metal? It looks a little shiny and there seems to be a dent in it just above the left-hand board.  Also, if it was metal it would provide ground-plane continuity of sorts behind the two bridges linking the lower board.


 * The stub comprising the top-left limb of the swastika appears to be grounded very close its end, which would modify the response considerably. --catslash (talk) 23:01, 1 August 2018 (UTC)


 * Well the bottom line is that we have no hard information. I had noticed the grounding link (if that is what it is) but couldn't explain its function.  It is reminiscent of impedance matching arrangements used on inverted-F antennae and helical resonators. <b style="background:#FAFAD2;color:#C08000">Spinning</b><b style="color:#4840A0">Spark</b> 18:45, 2 August 2018 (UTC)

FA preparation
I've finally got back on this one after letting it moulder in the compost heap of my to do list for months. I think I have now addressed all the points that came out of the peer review except for a couple of things. One is field pattern diagrams, which I am working on now. The other was a comment by user:Mark Viking to the effect that "the planar format fits in well with the manufacturing methods" should be replaced by "the planar format is easy to manufacture". I haven't done that because that is not the point, at least, it is not entirely the point. The great advantage of the planar form is it can be made by the same process as a pcb. Components made of planar transmission lines can be made in the same manufacturing step as the pcb traces. Essentially, the cost of these components is zero - or no additional cost. That is more than just being "easy" to make. <b style="background:#FAFAD2;color:#C08000">Spinning</b><b style="color:#4840A0">Spark</b> 22:17, 28 November 2018 (UTC)