At 07:23 AM 1/28/2007, [EMAIL PROTECTED] wrote: >Jim, see in text >73 (sorry no spell checker on this system.......) >groeten Peter ><mailto:[EMAIL PROTECTED]>petervn(a)hetnet.nl ; ><mailto:[EMAIL PROTECTED]>pa0pvn(a)hetnet.nl ; >pa0pvn(a)gmail.com ; pa0pvn(a)amsat.org . > > > >---------- >From: Jim Lux [mailto:[EMAIL PROTECTED] >Sent: Sun 28-1-2007 15:55 >To: [EMAIL PROTECTED]; k5nu; FlexRadio@flex-radio.biz >Subject: Re: [Flexradio] sdr swr > >At 03:08 AM 1/28/2007, [EMAIL PROTECTED] wrote: > >Mike > > > > > > > >What is still wondering me the most is that when the SWR is close to > >1, the transmitter impedance > > > >should not have any influence on the SWR. > >Actually, no.. think about this... what if the Transmitter output Z >were 100 ohms... You'd still see a mismatch. > >In the simplistic sort of view, a 50 ohm line hooked to a 50 ohm load >presents a 50 ohm impedance at the other end of the line, so you'd >think that the mismatch localized at that point. BUT, say you broke >that line half way, and put the meter there. Looking back towards >the transmitter, you see 100 ohms, with some length of 50 ohm line >(i.e. probably something other than 100 ohms, and almost certainly >NOT 50 ohms), looking towards the load, you see 50 ohms. So the >mismatch is apparent no matter where along the line you look. > >If there in no power coming back then you do >not see the impedance of the transmitter.
So it would seem, but if you grind out the math, it turns out that's not the case. I thought so too.. but, say you had 100 ohm source, a meter of 100 ohm feedline, a meter of 50 ohm feedline, and 50 ohm resistor. There's a mismatch no matter where you look. There's a real elegant explanation of this I found a while ago.. I'll see if I can find it. It was one of those "oh, of course that's the way it has to work".. before that I kept getting wrapped up trying to figure it out... and grinding the math, while proving what's necessary, didn't really provide any conceptual insight. It was the clever picture that did it. I'll see if I can find it. >Most SWR meters, also, make the assumption that they are in a 50 ohm >line. That is, they don't actually measure the voltage along the >line at several points and calculate the standing wave ratio in a "Z >independent" way. (they're not a slotted line with a >probe...) Therefore, while the can measure SWR (which is pretty much >the same everywhere in the line) they can't measure absolute Z very >accurately (that is, they can't tell you WHERE the mismatch is) nor >can they accurately measure the forward and reflected power separately.. > >As far as I know most SWR meters are in fact 2 directional couplers, >and are working pretty well. Actually, at HF, no.. directional couplers with good directivity and wide bandwidth are hard to build. Most HF SWR meters either use some form of bridge, or essentially sample the (complex) current and voltage at a single point, and from that, infer or compute the SWR. > To measure without bothering to much >over the tranmitter impedance one can put a -3 or -6 dB Power attenuator >between the Transmitter and the SWR meter. I found that this is verry >usefull if you are mesuring in high SWR conditions. Then the tramsmitter >impedance is working with the reflected power. yes.. putting in a pad is a very, very useful thing. >I always think of an endless coax in an properly terminated system. > >All the Thevenin match criteria says is that if source and load Z are >conjugates, you get maximum *net* power transfer. (consider a source >with 50+50j Z and a load with 50-50j... complex conjugates.. the 50j >on the source cancels the -50j on the load..... so this transfers >power from generator to sink most efficiently.. but there's "reactive >power" circulating back and forth between the two.. While it's >theoretically the same 100% efficiency as a 50+0j to 50+0j, in a real >system with IR losses in the connection, the reactive termination is worse) > >Good point Jim, reactive part of the impedance can put quite a bit of >sand in the system. Half of the working of antenna tuners is to campensate >for that. Idially the tuner should be at the arial connection, now you get >1/4 labda and 1/2 labda effects of the coax on the reactive parts as well. Or, as it usually happens, something like 15/17ths of a wavelength or something weird. Life is rarely so helpful as to provide exact 1/4 or 1/2 wavelengths, and especially not over a wide band. Putting the tuner/matching network at the feedpoint is certainly the way to go. And now, with cheap microprocessors, communications, etc., there's no reason not to do it. Back when you HAD to turn a knob to control something, the tuner had to be in the shack, so it was within arm's reach. But not any more. In fact, I'm an advocate of putting the PA and the matching network at the feedpoint. I'd much rather send my signals with low loss inexpensive RG-6 (cable tv coax) at a few milliwatts, then then send AC or DC power in something more appropriate. (Although, RG-6 foam can easily take several hundred volts and can carry 3 or 4 amps without any trouble.. ) I'm hoping that the recent FCC order doing away with the gain limitation on external power amplifiers will prompt more mfrs to build devices like this. We do it now on receive for UHF, with mast mounted preamps... the PA should be the same. >Jim, W6RMK > James Lux, P.E. Spacecraft Radio Frequency Subsystems Group Flight Communications Systems Section Jet Propulsion Laboratory, Mail Stop 161-213 4800 Oak Grove Drive Pasadena CA 91109 tel: (818)354-2075 fax: (818)393-6875 -------------- next part -------------- An HTML attachment was scrubbed... URL: http://mail.flex-radio.biz/pipermail/flexradio_flex-radio.biz/attachments/20070128/4946a834/attachment.html _______________________________________________ FlexRadio mailing list FlexRadio@flex-radio.biz http://mail.flex-radio.biz/mailman/listinfo/flexradio_flex-radio.biz Archive Link: http://www.mail-archive.com/flexradio%40flex-radio.biz/ FlexRadio Homepage: http://www.flex-radio.com/ FlexRadio Knowledge Base: http://kb.flex-radio.com/
