"John Larkin" <jjlarkin@highNOTlandTHIStechnologyPART.com> schrieb im Newsbeitrag news:fkt8t25shgq88ejbi56j4ubcaahnlntp3c@4ax.com...> > But FR-4 varies a lot, so there's no definitive data. > > What are you trying to do? >Hi John, I'm trying to compare several loss tangent values (I think that's the decisive parameter..!?) from several materials, FR4-, PE, etc. over the frequency. But it's hard to find such diagrams.
Building Coaxial transmission line on PCB?
Started by ●February 12, 2007
Reply by ●February 16, 20072007-02-16
Reply by ●February 16, 20072007-02-16
On 15 Feb 2007 22:26:30 -0800, "werty" <werty@swissinfo.org> Gave us:>On Feb 13, 7:51 am, John Larkin ><jjlar...@highNOTlandTHIStechnologyPART.com> wrote: >> On Tue, 13 Feb 2007 20:46:04 +1300, Jim Granville >> >> >> >> >> >> <no.s...@designtools.maps.co.nz> wrote: >> >werty wrote: >> >> On Feb 12, 2:30 pm, "john jardine" <j...@jjdesigns.fsnet.co.uk> wrote: >> >> >>>"Geronimo Stempovski" <geronimo.stempov...@arcor.de> wrote in message >> >> >>>news:45d04b34$0$27624$9b4e6d93@newsspool2.arcor-online.net...> I think transmitting high-speed signals is very easy when you have a >> >> >>>>360-degree ground reference, round conductors, >> >>>>and no other nearby signals like in coaxial cables. My aim is to design >> >> >>>PCB >> >> >>>>tracks as much like a coaxial cable as >> >>>>possible. Anyone tried this before? Is it possible with regular FR4, >> >> >>>anyway? >> >> >>>>Thanks for your help. >> >> >>>>Gero >> >> >>>Had trouble with crosstalk on a mass of video signals. Cured with a >> >>>multilayer board where each signal was 'boxed in' by ground plane to the >> >>>sides, above and below. Sort of square coax. >> >> >> ---------------------------------------------------------- >> >> >> Boxed ! the wavelength is far greater than >> >> your dimensions , thus higher modes can not >> >> exist , thus you do NOT need sides . >> >> When you reach 10 Ghz , then maybe >> >> you need sides in ur boxed "coax" . >> ><snip> >> >> >Seems video distribution is not one of your areas of expertise ? >> >> > Note that John was talking about crosstalk, in the analog domain, >> >and yes, what he did certainly will have a measurable improvement, and >> >is somewhat 'industry practise' when minimising crosstalk. >> > Note he says 'cured', that means he is on both sides of the problem, >> >and it is a brave (or something else?) person that counters such direct >> >experience >> >> >I imagine in extreme digital cases, such as where you are worried not >> >only about sending the clock, but also about the ps/fs of jitter, then >> >this design approach would also help : it's not hard to do. >> >> >-jg >> >> I do stuff down to a few ps RMS jitter on a regular 6 or 8-layer >> board, microstrip traces, with switching supplies and uPs and display >> drivers on the same board. Picoseconds aren't tough these days. >> >> John- Hide quoted text - >> >> - Show quoted text - > >____________________________________ > > You are hoping that we believe switchers > cause lots of noise .... > > Zero ripple is what switchers do ! > The sudden pulse of current is only > around a very short loop , it does > not cause noise . > They dont even have "ground loops" > >We cannot use switchers to feed to rails on our 2 to 12 GHz designs. NOISE CAN AND DOES get injected into such systems BY SWITCHING POWER SUPPLIES. YOU may not be aware of it, but those of us that work in such bands are aware of high frequency switching noise, and it DOES show up under spectrum analysis. Your brain has a ground loop.
Reply by ●February 16, 20072007-02-16
On 2007-02-16, werty <werty@swissinfo.org> wrote:> BTW > Study CAT5e Ethernet cable . > garbage ! > USB cables are much faster . > > > I think ppl limit themselves to > whats avail in PCB , then complain > when it dont work , but if they'd > experiment , they'd find the problem > is using thin PCB . > Then they limit on putting down > 100 transmission lines per mm . > You cant learn , unless you experiment.Shite man, what are you using for a computer, a 40-column Commodore PET? [...]> Sending signals that will be amplified > use high Z ( ~65 ohms ) and > sending power needs low Z . > These rules can't be bent . > Thats what you're doin , is bending > rules ...Even a PET wouldn't explain the random indentation. -- Grant Edwards grante Yow! Remember, in 2039, at MOUSSE & PASTA will visi.com be available ONLY by prescription!!
Reply by ●February 16, 20072007-02-16
Grant Edwards <grante@visi.com> writes:> On 2007-02-16, werty <werty@swissinfo.org> wrote: > >> BTW >> Study CAT5e Ethernet cable . >> garbage ! >> USB cables are much faster . >> >> >> I think ppl limit themselves to >> whats avail in PCB , then complain >> when it dont work , but if they'd >> experiment , they'd find the problem >> is using thin PCB . >> Then they limit on putting down >> 100 transmission lines per mm . >> You cant learn , unless you experiment. > > Shite man, what are you using for a computer, a 40-column > Commodore PET? > > [...] > >> Sending signals that will be amplified >> use high Z ( ~65 ohms ) and >> sending power needs low Z . >> These rules can't be bent . >> Thats what you're doin , is bending >> rules ... > > Even a PET wouldn't explain the random indentation.And the random word and punctuation spacing. Some kind of mobile phone? The Microsoft Word Usenet Export Filter? I think it's the usenet equivalent of green ink. <http://en.wikipedia.org/wiki/Green_ink> -- John Devereux
Reply by ●February 16, 20072007-02-16
On Fri, 16 Feb 2007 08:01:13 +0100, "Geronimo Stempovski" <geronimo.stempovski@arcor.de> wrote:> >"John Larkin" <jjlarkin@highNOTlandTHIStechnologyPART.com> schrieb im >Newsbeitrag news:fkt8t25shgq88ejbi56j4ubcaahnlntp3c@4ax.com... >> >> But FR-4 varies a lot, so there's no definitive data. >> >> What are you trying to do? >> > >Hi John, I'm trying to compare several loss tangent values (I think that's >the decisive parameter..!?) from several materials, FR4-, PE, etc. over the >frequency. But it's hard to find such diagrams. >No, what are you trying to *do*? WHY do you want a "coax on a pc board"? John
Reply by ●February 16, 20072007-02-16
On 2007-02-16, John Devereux <jdREMOVE@THISdevereux.me.uk> wrote:> Grant Edwards <grante@visi.com> writes: > >> On 2007-02-16, werty <werty@swissinfo.org> wrote: >> >>> BTW >>> Study CAT5e Ethernet cable . >>> garbage ! >>> USB cables are much faster . >>> >>> >>> I think ppl limit themselves to >>> whats avail in PCB , then complain >>> when it dont work , but if they'd >>> experiment , they'd find the problem >>> is using thin PCB . >>> Then they limit on putting down >>> 100 transmission lines per mm . >>> You cant learn , unless you experiment. >> >> Shite man, what are you using for a computer, a 40-column >> Commodore PET? >> >> [...] >> >>> Sending signals that will be amplified >>> use high Z ( ~65 ohms ) and >>> sending power needs low Z . >>> These rules can't be bent . >>> Thats what you're doin , is bending >>> rules ... >> >> Even a PET wouldn't explain the random indentation. > > And the random word and punctuation spacing. Some kind of mobile > phone? The Microsoft Word Usenet Export Filter? > > I think it's the usenet equivalent of green ink. > ><http://en.wikipedia.org/wiki/Green_ink>I love it! I'd never heard the phrase "green ink" before. It's a keeper. -- Grant Edwards grante Yow! Now, I think it would at be GOOD to buy FIVE or SIX visi.com STUDEBAKERS and CRUISE for ARTIFICIAL FLAVORING!!
Reply by ●February 16, 20072007-02-16
On 15 Feb 2007 22:22:50 -0800, "werty" <werty@swissinfo.org> wrote:> In coax for 2.5 Ghz , for example , > it WILL have a large diameter and > the center will have an exact dia and > ratio .. No substitutes .Never heard of micro hardline, I guess. Or non-TEM propagation modes in large-diameter coax. John
Reply by ●February 16, 20072007-02-16
On 15 Feb 2007 22:26:30 -0800, "werty" <werty@swissinfo.org> wrote:>On Feb 13, 7:51 am, John Larkin ><jjlar...@highNOTlandTHIStechnologyPART.com> wrote: >> On Tue, 13 Feb 2007 20:46:04 +1300, Jim Granville >> >> >> >> >> >> <no.s...@designtools.maps.co.nz> wrote: >> >werty wrote: >> >> On Feb 12, 2:30 pm, "john jardine" <j...@jjdesigns.fsnet.co.uk> wrote: >> >> >>>"Geronimo Stempovski" <geronimo.stempov...@arcor.de> wrote in message >> >> >>>news:45d04b34$0$27624$9b4e6d93@newsspool2.arcor-online.net...> I think transmitting high-speed signals is very easy when you have a >> >> >>>>360-degree ground reference, round conductors, >> >>>>and no other nearby signals like in coaxial cables. My aim is to design >> >> >>>PCB >> >> >>>>tracks as much like a coaxial cable as >> >>>>possible. Anyone tried this before? Is it possible with regular FR4, >> >> >>>anyway? >> >> >>>>Thanks for your help. >> >> >>>>Gero >> >> >>>Had trouble with crosstalk on a mass of video signals. Cured with a >> >>>multilayer board where each signal was 'boxed in' by ground plane to the >> >>>sides, above and below. Sort of square coax. >> >> >> ---------------------------------------------------------- >> >> >> Boxed ! the wavelength is far greater than >> >> your dimensions , thus higher modes can not >> >> exist , thus you do NOT need sides . >> >> When you reach 10 Ghz , then maybe >> >> you need sides in ur boxed "coax" . >> ><snip> >> >> >Seems video distribution is not one of your areas of expertise ? >> >> > Note that John was talking about crosstalk, in the analog domain, >> >and yes, what he did certainly will have a measurable improvement, and >> >is somewhat 'industry practise' when minimising crosstalk. >> > Note he says 'cured', that means he is on both sides of the problem, >> >and it is a brave (or something else?) person that counters such direct >> >experience >> >> >I imagine in extreme digital cases, such as where you are worried not >> >only about sending the clock, but also about the ps/fs of jitter, then >> >this design approach would also help : it's not hard to do. >> >> >-jg >> >> I do stuff down to a few ps RMS jitter on a regular 6 or 8-layer >> board, microstrip traces, with switching supplies and uPs and display >> drivers on the same board. Picoseconds aren't tough these days. >> >> John- Hide quoted text - >> >> - Show quoted text - > >____________________________________ > > You are hoping that we believe switchers > cause lots of noise .... > > Zero ripple is what switchers do ! > The sudden pulse of current is only > around a very short loop , it does > not cause noise . > They dont even have "ground loops" > > > >Please post the schematic of a zero-ripple switcher. John
Reply by ●February 16, 20072007-02-16
"John Larkin" <jjlarkin@highNOTlandTHIStechnologyPART.com> schrieb im Newsbeitrag >> No, what are you trying to *do*? WHY do you want a "coax on a pc > board"? >Ah, okay, what I am actually trying to find out is what makes FR4 act worse than e.g. teflon at data rates beyond 2,5 Gbps. Is it the loss tangent or the epsilon r? How is the frequency-dependent attenuation physically describable? Where does the energy go? Heat, ...? It was my opinion that higher frequencies can be transmitted over coax but not over FR4 because of the geometry. Because in a coax there is (almost) no energy loss because the TEM wave is "captured" by the outer shield and in a planar setup like stripline or microstrip there are E-field and H-field lines vanish into the air environment (or somewhere else...). Therefore I'm trying to design a coax on a PCB. Am I right with my thoughts, anyway?
Reply by ●February 16, 20072007-02-16
On Fri, 16 Feb 2007 18:58:54 +0100, "Geronimo Stempovski" <geronimo.stempovski@arcor.de> wrote:> >"John Larkin" <jjlarkin@highNOTlandTHIStechnologyPART.com> schrieb im >Newsbeitrag > >> No, what are you trying to *do*? WHY do you want a "coax on a pc >> board"? >> > >Ah, okay, what I am actually trying to find out is what makes FR4 act worse >than e.g. teflon at data rates beyond 2,5 Gbps. Is it the loss tangent or >the epsilon r? How is the frequency-dependent attenuation physically >describable? Where does the energy go? Heat, ...? It was my opinion that >higher frequencies can be transmitted over coax but not over FR4 because of >the geometry. Because in a coax there is (almost) no energy loss because the >TEM wave is "captured" by the outer shield and in a planar setup like >stripline or microstrip there are E-field and H-field lines vanish into the >air environment (or somewhere else...). Therefore I'm trying to design a >coax on a PCB. Am I right with my thoughts, anyway? >A couple of things make pcb's lossy: the loss tangent of the material (and FR4 is pretty bad) and the copper losses. Copper loss gets bad on conventional FR4 boards because 1. FR4's Er is high, so for a given impedance traces are skinny. 2. The underside of the copper is treated to bond to the epoxy/glass, and the treatment (black oxide or something) greatly increases skin losses. Peel some up and look... it's gross. 3. In the case of microstrip, the current is concentrated on the underside (the dirty side) of the trace, so losses are that much worse... the shiny topside of the copper is underutilized. Stripline would be better, with balanced current density, except that the trace will be much thinner, which has its own penalty. A good microwave pcb has a low Er, low loss dielectric; is thick, for low current density and wide traces; has very smooth copper, which means traces and pads peel off easily. I don't think any simple geometry tricks (ie, emulating coax) will make FR4 any better, and would probably make it worse. For low losses, microstrip on a thick board is probably as good as it gets. John
