The question and response below only consider thermal limitations. There are also signal quality limitations in some data modes, PSK31 in particular.
PSK31 is very intolerant of non-linearity. The "crest factor" of PSK31 can be up to 2, i.e. the peak power can be twice the average power. Since it is clipping of the peaks that causes splatter or IMD, with most transmitters you need to keep the output power below one-half the rated maximum in order to ensure linearity. Linearity is not nearly as much of a problem in RTTY, but it matters in PSK31. If you have the means to measure IMD of your transmitted PSK31 signal, you can test this. A couple of years ago I did some tests using a "PSKmeter" (a device that monitors a transmitted PSK31 signal and reports the level of IMD products in the signal) observing the output of my K3 in PSK31 with the K3 in DATA A mode. What I found was that IMD levels were low at powers below 5 watts (KPA3 not in line) and at powers above 12 watts but below 50 watts (KPA3 in line), but as the power was increased above the 5 watt or 50 watt level, the measured IMD started to increase quite rapidly. The reported IMD at full power (10 watts without the KPA3 or 100 watts with the KPA3) was high - much higher than I would consider acceptable. I did not do similar tests in the K3's PSK D mode, but I have no reason to believe they would be better. See W7AY's measurements of PSK31 quality in PSK D vs. DATA A (for power levels below 5W and 50W) at <http://homepage.mac.com/chen/Technical/K3/Digital/digital.html>. I would suggest to anyone who plans to transmit PSK31 at powers greater than the recommended 5 watt (one-half full power) level that they should monitor the quality of their transmitted signal to ensure that the signal quality is acceptable. The simplest "low tech" way to do this is to have another station report IMD figures for your signal at various power levels. Note that the other station should be located such that the signal they receive from you is neither too strong (a very strong signal can produce IMD in the receiver) nor too weak (the signal must be well above the noise level for the reported IMD measurement to be meaningful). In RTTY such precautions do not appear to be necessary. Qualitative observations of the spectrum of my K3's RTTY signals in both FSK D and AFSK A modes did not show a significant observable difference between the signals at 50 watts and the signals at 100 watts. 73, Rich VE3KI Wayne Burdick wrote: > Jim Dunstan wrote: > >> The KX3 specifications indicate a nominal 10 watts output. When >> operating a mode such as PSK, which presents a continuous output, >> what is the recommended power output? > > We recommend using 5 W for data modes and keeping transmissions > reasonably short. However, you can definitely use full power (up to 12 > W) if the duty cycle is low, such as during hunt-and-pounce in a > contest. Of course as the duty cycle goes up, so will the PA and case > temperature, and eventually the KX3 may automatically roll back your > power output. > > I just did a test using the built-in PSK-D mode, alternately > transmitting and receiving for 20 seconds each (simulating short > contest contacts). At 12 watts (20 meters), the temperature never got > high enough to roll back power during ten minutes of this. > > I then tested the KX3's high-efficiency TX mode, which reduces current > drain by roughly 50% for a given power level. This mode kicks in at at > 5.0 W or less in CW/FSK-D modes, and at 3.0 W or less in all other > modes. So I set power to 3.0 W in PSK-D mode, which is amazingly > effective given the S/N ratio of PSK31, and transmitted continuously. > After 5 minutes the PA temperature was still increasing slowly, but > I'm guessing I could have gone on for another 5-10 minutes. > > The lesson, here, is to let the other station transmit once in awhile :) > > Wayne > N6KR ______________________________________________________________ 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
