Galaxy DX98VHP SWR's

If it is too loud turn the mic gain down not the power. I usually run mine with the mic gain open about half way, maybe just a bit higher.

 

Wow. The thread kinda wandered around in this one, but I thought it was worth noting a couple of comments. Outlaw and MsJamie seem to have kinda summarized the output power proportioning correctly. There have also been some empirical "Yeah, but..." responses. If I may respectfully submit.....

If you look at a 20 watt unmodulated carrier on a scope, there's only so far the power can vary and still be linear. And it's also easy to confuse "average power" with "peak envelope power", especially when watching a meter -- more especially a mechanical meter with an actual swinging needle which develops momentum once it gets moving.
Watch a professional tech tune an AM transmitter sometime. Even without the relative luxury of a spectrum analyzer, the microphone is typically disconnected, and an audio oscillator (tone generator) of the desired frequency and a level that's adjusted to *not* be distorted is fed into the modulator input (or the mic input, on a two-way transmitter). As the audio level (possibly specified by the service manual, but can be 'scoped by the tech by itself if not) is allowed to have greater and greater influence on the carrier signal, the "rectangular" trace of the carrier develops sine waves along the top and bottom of the trace. As the modulation level increases, the tops (and bottoms) of the sine wave shapes on the edge of the carrier's earlier box shape grow taller, and the narrowed parts of the carrier signal between the peaks get closer together until they seem to pinch off, just at the point where the sine waves hit the zero line on the bottom. That point is where the output stage has been driven to cutoff, and is no longer capable of accurately reproducing what's being fed into it. Once the modulation gets to that point, the limiter is turned to allow no more than that, and any additional "transmit volume" adjustments can be made by the user with either a front panel mic gain pot or a power mic's output adjustment. But if the limiter circuit is defeated (or "just turned up a ways to make it sound better") we're into spurious transmission time.
If the modulator is running linearly, the tops of the peaks (and their mirror images below the baseline on the scope) will still retain a curve, which can be described as a sin function (which I can't do without taking off my shoes, or I'd now be a world-renowned surgeon...)
That's 100% modulation. On a different kind of display, like on a spectrum analyzer, the carrier power will have 25% more power added to it by *each* of the two sidebands. So average power output of an AM signal is increased by only 50% over the "dead key" carrier. But overdrive the stage by forcing it to try to conduct during the only part of its power cycle where there's actually current available at enough collector (or plate) voltage to amplify, and the output becomes decidedly "non-linear". By the way, Class C amplifiers are made for applications where there is no amplitude change at all, except on & off, like FM.

Peak output power is figured a little differently, and can still be figured from a scope by using a formula. Peak envelope power is usually applied to SSB transmissions, because there is zero carrier when there is no modulation, and the "opposite" sideband's power gets to get added arithmetically to the "correct" sideband's power, as well as the absent carrier's value, hence the higher permitted P.E.P. output in Part 95. But the output device(s) still must be biased into their linear operating region, and are often fed in push-pull so that when one device cuts off, the other comes on.

Once you drive the modulator so that the output is going non-linear, the device(s) that make the power do so in much the way that any other junction device does without proper bias to keep it within its linear range -- it can become a wideband noise generator.

For folks who are putting out 10 or 20 watts dead key but 250 peak, some careful listening might reveal that they're only 100% audible within the range that their dead key power covers. That's because AM receivers don't have a "product detector" to be able to decode the missing carrier. So listeners to a station with a 20:1 peak to carrier ratio, who are beyond the dead key's range, are probably only hearing the occasional explosive consonant sounds and RBs and short blips of the highest audio that are making the transmitter put out its maximum power, and only for the brief amount of time that they last. The rest of the time, they're probably down in the mud.

After doing this for 50 years, starting out when even most ham phone ops were using AM unless they built their own SSB adapters, I've been able to sorta note that, for AM, "Watts are for talking." If I'm talking to someone on a band where the conditions require 100 watts, I'll set my carrier to about 100 watts and adjust my modulation on the scope. My average output will only tick upwards a little bit as I talk, because meters have lag. But it sure is pretty on the monitor scope. "Swing" wouldn't exist if it weren't for the mechanical momentum of meters with high mass needles and little control over their standards of calibration.

Dead keying 20 watts and "swinging" to 250 is nearly as likely to be putting out much of that wasted power on harmonics and other out-of-band spurs as it is on adjacent channels, although it's usually easier to spot if it's tearing up the half-dozen channels above and below whatever is showing on the channel indicator. Having it not audible to another casual listener on a band full of skip, especially if he's beyond the range of the original 20 watts, I don't believe is that good a test of the spectral purity of a signal.
Especially when it flies in the face of years of solid engineering and physics. If that were the case, then commercial AM broadcasters, including shortwave propaganda stations like VoA, Radio Moscow, etc., would be using the technology.
As it is, the FCC only allows slightly above 100% modulation for commercial AM broadcast in the U.S. because those stations have tailored radiation patterns and power limitations, and only one licensee per frequency withing certain geographic limits, with specified guard bands between them for protection against splatter. At night, of course, even lowered output powers on MW bands can still carry a long way, hence the hobby of broadcast DXing.
But meanwhile, if you want to be heard reliably, I suggest keeping carrier power up where it needs to be to get heard, and not counting on the incremental boost of modulation level to keep your range up by much more of a margin. If your output stage runs too hot, my next suggestion is turn down the exciter to where it's happy and then boost the output -- linearly -- afterwards.
There's a good reason I don't refer to most RF amplifiers sold for the CB market as "linears": they're just not.
The whole part about echoes and reverb & such are a separate topic, part of what I think of "intentional distortion". I fervently believe that nobody would use them if they didn't have talkback so they could enjoy their own voices. Take a listen to the professional AM communicators who fly airliners: anyone hear any of that stuff?
But I digress -- again.
Rutroh -- Wifey says my fingers have shortened another inch since I started typing. Time to go.
73, all.

 

The antenna analyzer that you seem to have described the CB shop as having used to check the antenna, puts out a very clean signal on whatever frequency is dialed up on it. Since your SWR changes when it's on high power, it's pretty clear that the "high performance option" that you got done had the effect of making your radio put out maximum out-of-band garbage to make all those extra watts. Your antenna is working as a bandpass filter, and is "rejecting" all the stuff that's not within the range of frequencies ("channels") that it's made for, and reflecting the extra power back down the coax, where it gets read by your SWR meter as a high SWR. (I'm using terms loosely here, but the result is the same.)
So the high SWR is telling you that something is terribly wrong in your radio, especially when it's on high power.
Safe to use? Dunno. Since the big expensive parts that make the high power are operating outside the range they're supposed to work in, they and the parts around them are probably running much hotter than they are supposed to have to. If any one of them fails, it's hard to tell what will go next.

I typically figure that a radio manufacturer already wants his/her/its products to perform as well as they can, so get happy customers telling other customers to buy them. The idea that some retailer can re-engineer a product and magically make it work substantially by just turning adjustments seems, to address your question, unsafe to the radio.

I don't work for the place who sourced yours, and have no idea just what constituted their sooper-dooper toonup thingy, but I can nearly guarantee you that if your radio were put on a spectrum analyzer, it would look like a forest full of trees, rather than a flat plain with a single narrow pine tree in the middle.

You can probably get by with your radio on low power, since the SWR shows as low, and as long as you keep all your mic stuff low (to keep from putting out overmodulated spurs, doing the same thing as the high SWR as you talk). But high power, and/or too many audio things turned up, will all put your output stage at risk. SWR is measured with dead key, but overmodulating still does the same as a badly tuned spurious transmitter being dead keyed, but it does it at a varying rate.

I'd see if the vendor will un-do its super tune thingy; or just take the radio to a commercial two-way shop with a spectrum analyzer and pay them to tune it right and chalk it up to experience.

Low power, with your mic gain low enough to be clear instead of "loud and proud" will probably work fine and not fill your cab with smoke, and certainly cost the least.

Hope that helps,
73