This problem was discussed on the moon-net list in May-June 2016. I thought it was worth documenting what is going on with many radios that produce a “power spike”.
The problem is that the gain of the amplifier chain (from say, final mixer to output) will vary from rig to rig, also the drive level will vary. Therefore the actual gain required of the amplifier chain varies from rig to rig, purely due to component variations and even due to alignment settings, which are probably done fairly quickly at the factory.
The use of ALC to control the gain of the amplifier chain is a typical and common approach taken by manufacturers. It is a technique that does not work very well for modes other than FM and where external amplifiers are used, where the output power required is less than the maximum rated power of the radio (actually, less than about 1.5 X the rated power).
Other manufacturers also have these problems. They address them in different ways.
In one example, the Yaesu FT8*7 series, there are two controls for each band set (HF, VHF low, VHF high and UHF), being drive gain and output power limit.
Another example is Icom’s IC910 where I understand that the power level control has a dual action, one is to reduce the amplifier gain and the other is to change the output power limit level. So it achieves the same kind of result as the Yaesu 8*7 series, and should result in no power spike when first transmitting.
In any radio, if the driver gain is too high, the output power may momentarily exceed the preset level intended from the radio and set by the power level preset feeding the ALC circuit. The time constant of the ALC circuit determines the attack and delay times but cannot prevent the power level rising above the preset value, momentarily, and that’s all we need to exceed the limits of a solid state amplifier device.
So in the 8*7 series you can set the gain level appropriately so that the radio cannot output any more than your chosen power level, and it is a matter of alignment procedure to adjust output limit and stage gain appropriately to get the result you want. They do (cleverly) offer three power levels and you can set the gain and the output power limit for each power level.
For a radio capable of 100w output it is never going to be enough to set the output limit (driving the ALC) to (say) the 25w level. As already stated, that will still result in a power spike while the ALC sets the output level to what the user requires (via the output power control). It is more obvious and easiest to see in the constant carrier modes like fm and CW.
If the drive level is sufficient to allow the power amp chain to produce 100w, then the initial output (on say CW mode) will be 100w, and if you have set the output limit to 25w, feedback via the ALC circuit will reduce the power to 25w. But the initial spike will always be there. It may only last a few milliseconds, but with solid state circuits it is not a matter of heat or averages, it is whether the input voltage exceeds the correct level at all, for even the first sine wave at 144 MHz, ie. for 1/144 microseconds.
What produces the spike? There is enough drive to the final amplifier for 100w. The output limit setting may be set for a lower power level, if so the ALC line is used to send a gain control voltage back to the gain controlled stage(s).
To make it impossible for a power spike to be produced, the drive level has to be reduced. We need to limit the drive to the power amplifier chain to whatever is needed to produce the nominal power level, whatever it is, 1w, 25 or 75. This could be done in several ways.
The first and typical way to reduce the drive level, when using audio source and ssb mode, as for WSJT and other AFSK type modes, is to reduce the audio level going into the radio. This would work, but if the reduction in drive required is significant (more than say, 10 db) that decreases the signal level without reducing the level of noise and other inevitable spurious signals, including the suppressed carrier of the ssb signal. Eg a carrier suppression level of 45 db may be specified by the manufacturer, referenced to its performance at full rated power. By using the audio drive to reduce total output we are accepting that the suppressed carrier can remain at its current level, and that may be ok for some radios.
The second method of reducing the output is to use a high power attenuator between the radio and the external amplifier. This attenuator would be in that circuit on receive mode too. For EME use many operators using the separate receiver antenna input to the receiver, or use another receiver anyway. But the impact of the attenuator on receive mode is another factor to consider.
Other options include:
- Modify the radio’s internal gain in the transmitter chain, preferably in the section amplifying at the transmitting frequency. Depending on the design of the radio, there may be a point where the final mixer output is fed to the amplifier chain, which would be a good place to insert a suitable attenuator.
- Insert a voltage on the ALC line, setting the gain of the transmitter to the highest it is allowed to be – reducing the output level to the highest it can be to safely drive the external amplifier.
- Bypass amplifier stages.
Some types of mods would render the radio incapable of higher power output, so would need to be reversed when moving the radio on to another purpose. Whatever method is chosen, it must prevent the drive chain from producing enough power to drive the output stage to full power.
Importantly, it should be done in a way that is absolutely foolproof. A casual mistake by the operator that destroys the external amplifier is something to prevent entirely.
The metering on the radio would be meaningless if some of these options were taken. Separate methods of metering the drive level and adjusting for best operation would be required.
This is not a plug and play application. We are using a radio in a way that is outside its designed purpose.
The inability of the TS2000 (or other radio) to be used without modification for lower power purposes is no reflection of its suitability for other purposes. All commercial radios are built for the most common use by the majority of buyers. When we apply these general purpose radios to special uses such as for EME amplifier drivers, we cannot really be surprised that they are not ideal for EME drivers “out of the box”.
