If you have a high power export and use it on AM, you may notice it's very easy to modulate these radios into negative peak cutoff long before the positive peak clips. No setting of the carrier control can make both peaks symmetrical. You either end up with too little carrier that clips the negative peak or too much carrier that can't reproduce a proportional positive peak.
The first problem is the radio makes too much power to drive the internal high power amplifier. The manufacturer has tried to remedy this problem by only installing a single final on the dual final radio board. That helps prevent overdrive to the main amp but it's not enough. The next trick they used was the problem, reducing the carrier control to get down to the right level of drive.
On a Galaxy Saturn Turbo the collector voltage on the 2SC2312 under carrier conditions is a mere 1.79 volts to maintain a 25 watt carrier through the amp! Now when a symmetrical waveform enters the series pass modulator it takes very little audio to drive the 1.79 volts down to RF cutoff and has much more headroom reaching the 13.8 volts on positive peaks. Bypassing the amplifier revealed the radio had a carrier of 750 milliwatts and a modulated PEP of 18 watts. Producing a drive ratio of 24:1 with the peaks driving the amp well into gain compression.
In order for any series pass modulator to reproduce a symmetrical waveform with any power, we must set its unmodulated DC output closer to 50% of the total voltage. That means the carrier voltage of 1.79 volts must increase without producing more than 750 milliwatts of carrier. Without dedicating a lot of time to this project, I simply placed a 10 ohm 1/2 watt resistor between the driver and single final on the board. In series with the 471 pf coupling cap from the collector of the driver to the base of the final.
This allowed me to set the DC carrier voltage off the modulator at just over 3 volts to maintain the 750 milliwatt carrier. Positive peaks were also reduced to 9 watts PEP, permitting a PEP output from the main amp of just over 100 watts without gain compression and far less RF cuttoff on the negative peaks just by adding over a volt of headroom at the modulator.
I would have been happier to achieve this power at a 6 volt modulator carrier setting but didn't have the time to chase it further and was pleased with the results. Adding attenuation in the pre driver stage did not work because it cut the positive peaks faster than the negative peaks. If you have too little RF feeding a modulated stage, it won't have enough RF to develop a good positive peak.
The next step would probably be attenuation between the radio final and the amplifier. That would require larger non inductive resistors and more attention to impedance between the two stages and is why the simple 10 ohm resistor appeared more tempting to me. I've also heard reports that newer MOSFET radios with built in amps may have addressed this problem.
More than one source is claiming the new radios will not swing more than 5 watts with the internal amp bypassed. I'll bet the old 18 watts from the single final was giving the extra MOSFET amplifier a fit and had to be reduced to keep it together. If this problem was corrected on the new radios it would be easy to check. Set the radio for a 25 watt carrier and measure the DC voltage present at the 3 bias test points on the removable conductive PC strip in front of the driver and final. The closer it is to 6 volts the better the job.
The first problem is the radio makes too much power to drive the internal high power amplifier. The manufacturer has tried to remedy this problem by only installing a single final on the dual final radio board. That helps prevent overdrive to the main amp but it's not enough. The next trick they used was the problem, reducing the carrier control to get down to the right level of drive.
On a Galaxy Saturn Turbo the collector voltage on the 2SC2312 under carrier conditions is a mere 1.79 volts to maintain a 25 watt carrier through the amp! Now when a symmetrical waveform enters the series pass modulator it takes very little audio to drive the 1.79 volts down to RF cutoff and has much more headroom reaching the 13.8 volts on positive peaks. Bypassing the amplifier revealed the radio had a carrier of 750 milliwatts and a modulated PEP of 18 watts. Producing a drive ratio of 24:1 with the peaks driving the amp well into gain compression.
In order for any series pass modulator to reproduce a symmetrical waveform with any power, we must set its unmodulated DC output closer to 50% of the total voltage. That means the carrier voltage of 1.79 volts must increase without producing more than 750 milliwatts of carrier. Without dedicating a lot of time to this project, I simply placed a 10 ohm 1/2 watt resistor between the driver and single final on the board. In series with the 471 pf coupling cap from the collector of the driver to the base of the final.
This allowed me to set the DC carrier voltage off the modulator at just over 3 volts to maintain the 750 milliwatt carrier. Positive peaks were also reduced to 9 watts PEP, permitting a PEP output from the main amp of just over 100 watts without gain compression and far less RF cuttoff on the negative peaks just by adding over a volt of headroom at the modulator.
I would have been happier to achieve this power at a 6 volt modulator carrier setting but didn't have the time to chase it further and was pleased with the results. Adding attenuation in the pre driver stage did not work because it cut the positive peaks faster than the negative peaks. If you have too little RF feeding a modulated stage, it won't have enough RF to develop a good positive peak.
The next step would probably be attenuation between the radio final and the amplifier. That would require larger non inductive resistors and more attention to impedance between the two stages and is why the simple 10 ohm resistor appeared more tempting to me. I've also heard reports that newer MOSFET radios with built in amps may have addressed this problem.
More than one source is claiming the new radios will not swing more than 5 watts with the internal amp bypassed. I'll bet the old 18 watts from the single final was giving the extra MOSFET amplifier a fit and had to be reduced to keep it together. If this problem was corrected on the new radios it would be easy to check. Set the radio for a 25 watt carrier and measure the DC voltage present at the 3 bias test points on the removable conductive PC strip in front of the driver and final. The closer it is to 6 volts the better the job.
