ALC, Keying, & Other Amplifier Problems
For some reason, there is great confusion about interfacing amplifiers to transceivers, especially legacy models of both. Part of the problem is the popular amateur press, as often seen on these very pages. These admonishments often contain anecdotal information and personal preferences. Further, most respondents have never owned the equipment in question, so their advice is based on what they have owned or used. Thus, some suggestions result in a costly mistake being made.
The first place you should look for interconnect information is in the associated manuals of the equipment you're trying to interface. If you don't have manuals, buy them! There are dozens of places on the net to buy and/or download amateur radio equipment manuals. The URLs appear almost daily within these pages. However, the typical scenario is a late model transceiver with a legacy amplifier. In these cases, in order to safely interconnect them, requires you to know a little more about than what's typically in the manuals.
Relay Keying
Far to many late-model transceivers do not contain a relay to control an amplifier. When they do it is usually not adequate for legacy amplifiers (sometimes not even for new model amplifiers). The requisite open-key voltage maybe exceed (or peak to) 100 volts, and might be negative or positive depending on the model. Most late-model units are just 12 volts, but sometimes the current requirement exceeds one amp, typically twice the maximum current allowed.
Most legacy transceivers can easily key a late-model amplifier, but the reverse is not always possible. This is where a keying interface makes a lot of sense. Both AmpKeyer and Ameritron (and others) make transceiver/amplifier interfaces which will interconnect just about any transceiver to any amplifier.
There are a few companies who manufacture kits to be added to legacy amplifiers to reduce their keying requirements. This might be okay if your transceiver has a relay, but if you own an Icom IC-706 for example, you'll still need a keying interface of some kind.
One important item to keep in mind when using any keying interface is this: The ALC voltage is unbuffered. That is to say, what ever goes in, comes out. Remember this when reading the next section.
ALC
ALC, which stands for Automatic Level Control, is essentially a feedback circuit (internally and/or externally) which gain-limits one or more stages. Its purpose is to prevent over driving the finals as well as any attached amplifier. Due to design considerations of most legacy tube-type amplifiers, the first ALC output circuits (which started to show up in the late 60s) were negative going. That is to say, their resting output (no ALC) was zero. As overdrive was approached, the ALC voltage starts going negative, thus lowering the transceiver's drive (output power). The ALC output voltage may range from -6 to as high as -100 volts.
Most amateurs erroneously believe that all amplifiers require 100 watts of drive. Since this is all their transceivers deliver, many don't bother with the ALC connection. The truth is, most tube amplifiers only need about 65 to 80 watts for full rated output, and a few just 40 watts. Some solid state designs require no more than 25 watts. In any case, over driving results in excessive IMD, manifesting itself as splatter and distortion. Just for the record, unless you own a laboratory-grade storage scope, and know how to use it, you can't see the distortion caused by overdriving. In other words, if you're using a station monitor scope, or an ALC indication to check for distortion in your transmitted signal, you're kidding yourself!
Nowadays, the ALC output supplied by most late-model amateur power amplifiers is between -4 and -20 volts, which is about perfect for most late-model transceivers. For example, an Icom IC-746's output power will drop to nearly zero when the ALC input exceeds -4.5 volts. This presents two problems.
First, while -4 and -20 volts range is correct for late-model transceivers, it is not enough for some legacy transceivers like a Kenwood TS520 or Heath SB102. There is no easy way to correct this particular malady.
The other problem is, the ALC output voltage of some legacy amplifiers can be as high as -100 volts. Apply this to the ALC input port of a late-model transceiver, and you'll have a melt down! Thankfully, the current requirements are low, so a series 10k ohm resistor and a 5 volt zener diode is all you need to protect yourself in these cases.
To be sure, the ALC output pot on the amplifier (if there is one) should be turned to its lowest setting (no ALC output) before doing any interconnections or adjustments. This is true regardless of the age of your amplifier. In any case, the ALC adjustment procedure for your particular amplifier should be followed. Remember, proper use of ALC will increase the average talk power without overdriving the amplifier.
There are almost as many ALC adjustment procedures as there are amplifiers. For example, no two Ameritron amplifiers have the same procedure. This is another reason why it is very important to have manuals for your equipment. In simple terms, the ALC should be adjusted to limit the transceiver's output to the maximum the amplifier is designed to handle. With very few exceptions, this is less than 100 watts!
How Much Drive
It seems to be common practice to drive one's amplifier right to the ragged edge and beyond, and then wonder why the finals died after just a few hours of operation. Add in the fact most users don't know how to tune one, or take way too long to do so, and tube life can be measured nearly in minutes.
One of the more popular amplifiers these days is the four tube, Ameritron 811H (and the 811 3 tube model). If you care to read the manual, it says: This amplifier is designed to operate at full ratings when it is driven by an exciter that has approximately 70 watts (55 watts for the 3 tube version) of RF output. You can use an exciter that has lower output power, but the amplifier's output may be less. If you use an exciter that delivers more than 70 watts, carefully adjust the driving power to avoid "over drive" and the creation of spurious signals, which could create needless interference to other operators. Pay no heed to this statement, and you'll be buying new finals.
Speaking of which, it's all the rage these days to replace the 811As with 572Bs which have more plate dissipation (they also have more plate capacitance which effects tuning in some cases). The general thought is, now that the plate dissipation is about double, you can drive the amplifier with more power. Wrong!
The power supply in these amplifiers, is minimal at best. Overdriving not only taxes the finals, it also taxes the power supply. Factually, it doesn't make much difference which is which, the net result is increased IMD products which cause splatter and distortion.
Hand in hand with this, is the belief that 20 or 30 more watts will magically garner that rare DX contact. It won't! The amount of power increase it takes for a receiving station to notice any real difference in signal strength is about 3 dB. In the case of the 811H, that's equivalent to about 1,100 watts out. The only way to achieve this amount of power increase, is to trade the 811H in on an Al-572!
Tuned Inputs
A lot of the legacy amplifiers do not have tuned inputs. Depending on a lot of factors, this fact may or may not be of concern. When it is, placing an antenna tuner between the transceiver, or using an internal one, to match the input impedance is wrought with problems.
So is using an amplifier without WARC band coverage. While the input tank (if it has one) may be adjusted, the final tank may not have enough bandwidth, and there is almost no way to tell save for reduced output which results in heating up the various final tank components, to say nothing about the finals. And it doesn't address any spurious emissions which might occur.
Finals
Some of the tubes used in legacy amplifiers are no longer available, or are VERY expensive. Eimac 8874s, and all of the various sweep tubes are in this category. As a result, a lot of tube substitution goes on. Like every coin, there are two sides to this.
If you know what you're doing, tube substitution may be the only salvation for that much-beloved amplifier you've owned since high school. Virtually, it requires redesigning the final tank circuitry, changing tube sockets in most cases, and having the necessary tools as well as the knowledge to use them. It is not a job for the uninformed, or faint hearted.
Far too often, folks just stick in a substitute and pray for the best. While it may indeed put out power, you have no clue how clean it is. As stated previously, it takes special laboratory equipment to measure IMD. Meters and on-air assessments are meaningless.
Another inane, ill advised, and selfish act, is deliberately modifying the drive and/or bias of the final stage of an otherwise properly operating transceiver to get a few more watts out. The net result is to drive the finals past their linear response curve which creates additional IMD products that can be clearly heard.
The Presumption of Power
Why amateurs (especially neophyte ones) think they just have to have POWER to make every contact is beyond me. If I may be so brash, I think it is the CB inheritance a lot of us share. This manifests itself in the need to use compression as if it were a necessity, replace otherwise decent microphones with ones enhanced with parametric amplifiers, and driving their amplifiers way beyond reason. As my good friend Steve Katz, WB2WIK/6, would say, "All knobs to the right."
The use of compression is a double edged sword. Properly adjusted, using a well-matched microphone in the proper manner, will indeed boost your average to peak power. If you're already on the verge of overdriving your amplifier, the use of compression will certainly put you there, and unless someone complains about your splatter, you'll never know.
Part of the problem is rooted in these simple facts: Built in metering is universally inaccurate; Most after-market wattmeters aren't much better than built in ones are, including and especially peak reading ones; And a total lack on understanding of the ratio between average power and peak power.
The very best of wattmeters are typically rated at 5% of full scale reading. For example, if the scale you're reading from is 0 to 200 watts, and the actual power is 100 watts dead carrier, the reading could be from 90 to 110 watts and still meet specs. Wattmeters of this accuracy are upwards of $500, and some as high as $1,500. What's more, use one on the edge of its frequency range, and all bets are off. Therefore, you can't rely on a meter to indicate overdrive, and the resulting splatter and distortion.
Regardless of the meter's ultimate accuracy, using one to measure SSB peak power isn't much more than speculation. Individual speech patterns, meter dynamics, transmitter dynamics, an amplifier's dynamics, compression settings, microphone gain settings, and even the background noise level will all effect the reading. And contrary to popular belief, a monitor scope won't help either. Nor will any rule of thumb.
This is to say, if we knew the actual SSB peak power was 100 watts, our non-peak-reading wattmeter might read from 10 to perhaps 40 watts. Wattmeters with peak reading ability are under the same constraints, so their reading aren't the ultimate either. This said, they can be of benefit if for no other reason than to indicate that there is a real difference between average and peak power, or at least should be. If there isn't, it's a sure bet the knobs are turned too far to the right!
Tuning the Beast
It is not uncommon to hear someone tuning up for what seems like hours, striving to get the last ounce of power out. The fact is, tuning an amplifier shouldn't take more than 30 seconds, and if you've done it before, even in less time. And remember this; Improper amplifier tuning is the single most prevalent cause of component failures, especially in minimally sized amplifiers like the Ameritron 811H.
If you don't know the proper procedure to tune your amplifier (perhaps even if you do), you need to read this: http://www.w8ji.com/loading_amplifier.htm
What's more, you should print out the page and put it in your manual as reference material. Pay particular attention to the second paragraph in blue print, and the last paragraph.
Spares
While a little off the subject, this is another consideration to keep in mind if you own a tubed (or solid state) amplifier, especially a legacy one. Over the last 20 years, the number of final types suitable for amateur service has declined to the point you can almost count them on one hand. With very few exceptions, most currently available ones are made in Russia and/or China.

If your amplifier uses 3-500Zs, 811As, 572Bs, 6146s, 8877, or 3CX800s, I don't suppose you have to worry about spares, yet, but this will not always be the case. Sooner or later, they'll all go the same place 8874s and all of the sweep tubes have gone.
Conclusion
Unfortunately, the real people who need to read these lines, won't. The best the rest of us can do is set an example by operating moderately and courteously.
Alan Applegate, KBG
http://www.k0bg.com