Bolometers work at pretty low (milliwatt) power levels and you would need to use an attenuator, which could add some error to the measurement. Hewlett Packard made the HP434A Calorimetric Power Meter, which could measure up to 10 watts by matching the temperature in a load with that caused by DC power (which can be accurately measured). I often used one in the early 1960s (this was at work--they were a bit pricey for ham use).
Bob, N7XY On Mar 10, 2012, at 4:56 PM, Fred Townsend wrote: > Ron, Rick, Erik: There are many reasons why temperature measurements are a > poor way of evaluating a tuner. There is a method of power measure based on > temperature rise. It's called a bolometer. It only works well if the mass > involved is small. Wiki bolometers if you want the details. > > I think there is an even more important reason not to use this approach. > Let's consider the case where the transmitter has a very strong second > harmonic say -10db. If the tuner is doing its job, tuning, not just > matching, the 2nd harmonic will be tuned or filtered out. 10% of the energy > will be lost. What becomes of that energy? In an ideal world that energy > would be reflected into a dummy load, like when you use a circulator. Since > most tuners don't have circulators, part of the energy will be dissipated in > the tuner and part will be reflected back to the transmitter. That energy > that is dissipated in the tuner is a measure of how well, not how poorly, > the tuner is working. BTW in this case measuring SWR at the transmitter will > also suggest the tuner is not doing its job well when it is. > > Now we know good rigs don't have strong harmonics but I think I have > illustrated why measuring heat is not a good measure of performance. > > 73, Fred, AE6QL > > -----Original Message----- > From: elecraft-boun...@mailman.qth.net > [mailto:elecraft-boun...@mailman.qth.net] On Behalf Of Erik Basilier > Sent: Saturday, March 10, 2012 12:31 PM > To: 'Ron D'Eau Claire'; 'Rick Stealey'; elecraft@mailman.qth.net > Subject: Re: [Elecraft] high-power tuner > > As has been clearly demonstrated in this thread, there are multiple methods > of measurement. The one that gets my vote for elegance is the one with two > tuners back-to-back. With respect to the method that measures temperature > rise, taking into account the mass of the tuner, one also needs to take into > account the specific heat capacity of the tuner. One kilo of material x > doesn't heat up at the same rate as one kilo of material y when the same > heating power is applied. The tuner will of course be a mix of materials, so > one would have to measure the rate at which the tuner heats up when heat is > applied through a know heating source rather than TX power. If it is done > that way, one needs to know neither the mass nor the specific heat capacity, > since what one is measuring is essentially the mass times the specific heat > capacity. > > 73, Erik K7TV > > -----Original Message----- > From: elecraft-boun...@mailman.qth.net > [mailto:elecraft-boun...@mailman.qth.net] On Behalf Of Ron D'Eau Claire > Sent: Saturday, March 10, 2012 11:47 AM > To: 'Rick Stealey'; elecraft@mailman.qth.net > Subject: Re: [Elecraft] high-power tuner > > I was investigating tuner losses a few years ago and ran into many of these > same questions. > > A physicist buddy pointed out to me that the normal approach to measure loss > in something like a tuner is to put it in a well-insulated chamber and > measure the rise in temperature over time while transmitting. From there on > can calculate the energy required to cause the temperature rise which can be > used to calculate the number of watts of RF that never make it through the > box. > > 73, Ron AC7AC > > -----Original Message----- > > There seems to have been no answer as to how to accurately measure the loss > in a tuner. > Here is a solution but requires two tuners or at least one calibrated one > that could be used to measure others. > Take first tuner and tune it into the mismatch, say 600 ohms. Use an > antenna analyzer. > Then remove the load, and connect another tuner to the output of the first > (back- to-back - antenna port on first to antenna port on second tuner.) > Then put a 50 ohm load on the second tuner where the transceiver would be > connected, and tune the second tuner to a match. It will have the same > settings as the first tuner, complete symmetry. > Then measure the power in the 50 ohm load to get the loss. > > Since both tuners are matching the same load, and the system is symmetrical > the loss contribution by each tuner is half. Repeat for other types of > loads, and now you have a calibrated tuner to use with any tuner you want to > test. > > Rick K2XT > > ______________________________________________________________ > Elecraft mailing list > Home: http://mailman.qth.net/mailman/listinfo/elecraft > Help: http://mailman.qth.net/mmfaq.htm > Post: mailto:Elecraft@mailman.qth.net > > This list hosted by: http://www.qsl.net > Please help support this email list: http://www.qsl.net/donate.html > > ______________________________________________________________ > Elecraft mailing list > Home: http://mailman.qth.net/mailman/listinfo/elecraft > Help: http://mailman.qth.net/mmfaq.htm > Post: mailto:Elecraft@mailman.qth.net > > This list hosted by: http://www.qsl.net > Please help support this email list: http://www.qsl.net/donate.html > > ______________________________________________________________ > Elecraft mailing list > Home: http://mailman.qth.net/mailman/listinfo/elecraft > Help: http://mailman.qth.net/mmfaq.htm > Post: mailto:Elecraft@mailman.qth.net > > This list hosted by: http://www.qsl.net > Please help support this email list: http://www.qsl.net/donate.html _____ N7XY DX Cluster Node - telnet to n7xy.net, port 7300 ______________________________________________________________ Elecraft mailing list Home: http://mailman.qth.net/mailman/listinfo/elecraft Help: http://mailman.qth.net/mmfaq.htm Post: mailto:Elecraft@mailman.qth.net This list hosted by: http://www.qsl.net Please help support this email list: http://www.qsl.net/donate.html
