What about the output jack on the back of the Talking House 5.0 that is used for an ATU? that output can be 50-75 Ohms as I was told. So if you make your own ATU you would essentially have a 50 ohm antenna system No?
Good question. I have no experience with the Talking House or the ATU but I do have experience with antenna tuners in other applications.
Generally, the purpose of an antenna tuner is to present an acceptable load to the transmitter. In ham radio this is accomplished by tuning to provide the minimum SWR measured between the antenna tuner and the transmitter. Used this way the antenna tuner does nothing regarding the match between the transmission line and the antenna. Also, minimum SWR in this case does not necessarily mean the transmitter sees a 50 ohm load, it only means that the reflected power is minimum. But this IS NOT HOW THE TH ATU WORKS.
The TH ATU, according to a schematic which I reviewed, tunes the antenna to resonance. The meter indicates the relative voltage at the base of the radiator element and tuning is to peak this voltage. This does not mean that the transmission line or the transmitter is presented with a 50 ohm load. For example, in my antenna system the peak radiator voltage happens when the system is resonant and presents about a 30 ohm load to the transmitter.
In none of the situations which I have presented is the load seen by the transmitter assured to be 50 ohms. It may be but it may not be 50 ohms. Therefore, assuming that the power measured with a 50 ohm load is not likely to be the same as the power delivered to an antenna system.
The proper way to measure power delivered to an antenna is to measure the RF current to the antenna and if the antenna Z is accurately know the power can be calculated. Another method is to use a directional wattmeter (not a SWR meter) which can measure the forward and reflected power. The power can also be measured by measuring the voltage, current, and phase angle at the antenna feed point or at the input to the transmission line if one is used.
Neil
Rich, your AMT3000 posts have been most useful. While waiting for delivery of an AMT3000 kit I'm simulating the AMT3000 RF output stage in LTSpice. The aim is to optimize the circuit for a 50 ohm load.
It's an interesting circuit that runs 100 mW DC under all conditions. But, the efficiency isn't the best. R18, the 820 ohm collector resistor eats up a lot of the power and I suspect it's there to provide a load no matter what. Removing R18, and using a 900-to-50 ohm L-network the transmitter outputs 26 mW carrier and 40 mW at 100% modulation.
Interesting results from your simulation. If you install the L network and make make measurements, please let us know what you find. I found that simulations using LTSPICE are fairly good at predicting performance of RF circuits at BCB frequencies.
Any resistance in the final output circuit will waste power so you are on the rignt path by removing this resistor.
I am interested in your choice of 50 ohms as the impedance to transform to 900 ohms.
Neil
I chose 50 ohms because that's what I have instrumentation for and so that the modified AMT3000 is not set up for one particular antenna. Antenna adjustment will be performed with an MFJ-259B. Any antenna I build will have a feedpoint impedance lower than 50 ohms and transforming that to 50 ohms only takes a small feedpoint shunt inductor.
The 900 ohm collector load allows the PA transistor to swing half of the available 12 volt negative swing with an unmodulated carrier. With carrier it swings down to 9 volts and with 100% modulation it swings to 3 volts.
The PA is so inefficient that when set to run 300 mW DC input the RF output of 65 mW is what an efficient 100 mW class-C stage would run. So, I'm looking at changing the PA to run 100 mW DC input with high efficiency. Tripling the RF output power gives the station 3X the coverage area.
It appears that the final can run class C but I am not sure. Your experiments will yield interesting and probably useful results.
Neil
I have the new AMT3000 RF output stage and the audio modulator working nicely in SPICE. The RF stage is operating class-B with an efficiency of 67%. It's simple (much simpler than what's already in the AMT3000) and can be grafted into the exisiting AMT3000 PCB. The new circuit does not have automatic DC power control and must be set manually. So, 67 mW RF out vs 23 mW (in simulation) for the stock AMT3000.
I think I'll build the kit stock and have the new RF output stage on perf board. I can then explore both circuits and modify as needed. The next project might be to build a class-E stage with close to 100% efficiency.
Wouldn't Class D be easier to build and adjust?
I was looking for a one transistor amp and that would be class E. I'm not liking what I'm seeing with the class-E amp in simulation as the fast collector slew rate is causing high current spikes into the L-network capacitor. On the other hand, class-E is attractive because of the high efficiency. It could output 40% more power than the class-B amp.
If I understand it, a class-D amp is a push-pull amp using an output transformer and I'll explore that at a later time.
The class-B amp is nice in that it's tame and looks hard to break.
davec,
Though not applicable to the SSTRAN you are working with you might be interested in these posts and the threads which contain them. They report on a design which produced an 86% efficiency. The design now uses a PLL instead of a crystal and has been on the air here for several years feeding a base coil loaded antenna. Range observed with a car radio is one mile good copy and 1.3 miles recognizable signal.
http://www.part15.us/comment/12797#comment-12797
http://www.part15.us/comment/20498#comment-20498
Neil
Neil,
thanks for the links and I'll play with your transmitter schematic in SPICE.