Why Class Matters in an Amplifier

We all know Alpha amplifiers are "high class". But I want to talk a little bit about another meaning of "class" as it pertains to amplifiers- the class of operation, and what it means for the efficiency of the amplifier.

The job of a linear amplifier is to reproduce at its output a larger replica of the RF cycles appearing at its input. The input signal appears between the grid and cathode, and acts as a gate to control the flow of electrons between the cathode of the tube and its anode. As the input signal increases, so does current flow in the tube. This current, when converted by the so-called "tank" circuit, becomes the amplified output of the unit. You would think that in order to produce an entire sine wave cycle at the output, current must flow in the tube over the whole RF cycle. That is one possible solution to making an amplifier, but it turns out we can do better. By allowing the tube to be cut off (zero current flow) for part of the RF cycle, some desirable properties can be obtained, with acceptable side effects. But if the tube is cut off for part of the cycle, how can the output waveform be a faithful replica of the input? It turns out that the output tank helps in this regard. It acts a lot like a flywheel.  When the tube is conducting, the output waveform is mostly determined by the current in the tube. When the current drops to zero, the stored energy in the inductors and capacitors which make up the tank continue the cycle, completing the sine wave. This is exactly like a steam engine driving a flywheel- the piston of the steam engine only delivers power during part of the rotational cycle of the flywheel, and the angular momentum of the flywheel carries it around the rest of the cycle.

There are various ways of describing the class of the amplifier. The letters A, B and C are the most common, and have been in use for many decades. One can also characterize an amplifier by its "conduction angle", which is the number of degrees (out of a possible maximum of 360) over which the tube draws any appreciable current. Conduction angle affects a number of things, chief among them being efficiency. Conduction angle can be varied over a wide range in a real amplifier by changing the grid bias and the RF driving voltage. Assuming a perfect device, it is relatively straightforward to compute the efficiency as the conduction angle is changed. This is shown below for the case of maximum output power, all other things being kept equal. Two important points are shown- class A, on the right is capable of no more than 50% efficiency at full power;  class B, in the middle is capable of 78% efficiency at full output. Class C is the region from the middle to the left, and the efficiency increases to 100% as the conduction angle goes to zero- however, at this point the power output is also zero! So the 100% efficiency point is a theoretical asymptote only.

The region we are most interested in is the so-called "class AB", which consists of all the conduction angles between 180 and 360 degrees. This is where Alpha tubes are biased. Class B would be nice from an efficiency point of view, but produces unacceptable distortion. Lowest distortion would occur in class A, but the efficiency is unacceptable- so the compromise is somewhere between these extremes. You may also notice a number added- class AB1 or AB2. The number refers to whether grid current flow or not- 1 denotes no grid current, 2 denotes that grid current flows over part of the cycle.

This is the case for maximum output power at a given plate voltage, such as for CW or FM.  What happens (as in the case of SSB or AM) where the signal spends much of its time below maximum (or PEP)? It turns out that this can be computed as well, and this is shown below.

This shows the efficiency at partial output, plotted as a function of output voltage. So on the bottom scale, which goes from 0 to 1, 1 represents full output power, whatever that may be- 100 watts, 1.5 kilowatts etc. The numbers- 0.1, 0.2 etc. represent the RF voltage compared to the RF voltage at this full output. Since power is voltage squared, the number 0.5 represents (0.5*0.5) or the quarter-power-output point. Half power is close to 0.7.

There are three curves for efficiency- classes A, B and AB. The class B curve is a straight line from 0 to 78%. Class A is a parabola, and falls off steeply for powers less than the maximum, which is only 50%. Class AB is somewhere between the two. It has a maximum efficiency around 70% at full output, and falls off more steeply than class B, but is still better everywhere than class A.

The fourth curve, labeled "Pd" shows the waste power that is generated by a tube in class B at each output level. It is expressed in percent as well, in this case percent of PEP. So if we had a 1000 watt PEP amplifier, then 40% represents 400 watts of waste heat. The important thing to notice here is that the most severe case of heating for the tube occurs between one quarter and one half of maximum output. So if you think you are "saving" your tubes by operating the amplifier at less than its design point, think again!

The final interesting point to be made is that this analysis applies to all type of amplifiers- triodes, tetrodes, even MOSFETs and bipolar transistors. There are, of course, secondary effects which change the details for each of the device types, and which are important in equipment design. But a class B amplifier will have roughly 78% efficiency whether built from a tube or a FET. In fact, the analysis here follows closely the approach suggested in the book "Solid state radio engineering", by Krauss, Bostian and Raab.

The other thing to notice when comparing real amplifier efficiencies is that this analysis is for the power delivered into the tank circuit. There will be other losses in the tank circuit that will reduce the total amplifier efficiency. And if other power requirements are taken into account, the "wall plug to coax" efficiency will be lower yet. Finally, the properties of the signal which is being amplified have a big influence on the average efficiency, but that is a topic for another time.

73, Gordon, W0RUN 

Happy Summer.  I hope that all of you had a great Field Day!   

We are working away here at the South Sunset facility, repairing amps and shipping new amps and answering technical questions from many of you.  I am still digging into solving lots of old documentation and process issues.  Things are getting better; many Bills of Material have been cleaned up and I know a lot more about the insides of an amplifier than I ever expected.  It has been a learning experience and I appreciate the support from all of the hams out there who I have spoken with since my arrival here.

One of my biggest concerns these days is addressing our sales backlog and keeping the pressure on our vendors and suppliers in order to reduce that and get more of our amplifiers out the door.  As you can imagine, it takes a lot of attention to detail to stay on top of the purchasing activities and ensure that a major component or a tiny little part doesn't hold up the completion of an amp.   

I am really excited about the prospect of the new antenna tuner that Steve is leading the charge on.   I think our entire staff is stoked over additional products on the horizon to complement our current suite and keep us on top of the market.  We'll be waiting for that final design and ready to implement it. 

73, Kathy, KB2HDT 

Good day everyone. I trust the July 4th  Weekend for USA folks will be a nice
break for all.

First and foremost, Dayton was a great event for us. Thanks so much for those of you who stopped by to say hi. Molly, Gordon, Michael, Harold, Ryan, Steve and Jack all enjoyed visiting with you.

We are pleased to announce our new and only Alpha Amplifiers distributor outside of North America. Andrews Communications Systems will be representing us within Australia, their home, as well as Asia and New Zealand. They will be selling and  warranty servicing our complete line of products as well as representing us at the many events throughout their territory. 

Andrews has just celebrated its 35th year in business as Australia's leading radio and optical distributor. Check out their web site at Andrews Communication. Welcome Lee Andrews and his team to the Alpha team.

A few weeks ago I decided to telephone some recent Alpha Amplifier 8410 owners to see how the product was working for them, why they decided to buy the 8410 over the competition and why they purchased the 8410 instead of the 9500.

Though sampling 14 owners is far from a majority opinion I thought their comments were worth sharing with you.The following is a consolidation of their feedback.

How is the 8410 working for you?

13 customers said----"Works well, protection device does what it is supposed to do, fire it up and it works, stable, puts power out, just great, awesome piece of equipment, perfect design, no service issues, so pretty the way it is built, had two Alpha amps before and an Ameritron. Found Ameritron amp to be hard to tune, not legal limit, power transformers vibrate and noisy. Alpha best out there and I can afford the best"

2 customers had some initial questions and offered some very positive points included in the above paragraph. "Couldn't get amp to tune up on 15 meters at first. Tech support was huge help in resolving problem"

Why the 8410 over competition?
Customers said "Made in USA was big reason why, competitive
chart within our web site was very helpful" Download it HERE. "I can afford the best, Alpha reputation is the best, I can pay more and it is worth it, chose based on conversations with Alpha customers, quality of competition not good (actually verbiage not exactly correct-was more colorful but not suitable for print), holds value, not overpriced.

Why the 8410 instead of the 9500?
"Didn't have another $3,000" (now the difference is $2,555 with
8410 6/1 price increase)," like tuning myself, too many additional
parts to go wrong, may get bored, tube life and tube replacement issue, if had 4 year warranty on 9500 tubes may have gone that route, not a price issue, too much $s for the value" (in some cases this may be our fault for not explaining the extensive benefits and features of the 9500 - we do now-take a look at www.rfconcepts.com for 9500 details), not overpriced but did not
need an auto tuner, owner friend suggested 8410".

Other unsolicited comments
Wild praise for Molly, Glenn and troops at RFC for customer support and inquiries.I know, at times, we miss hitting our high expectation mark but we are always trying to be the best.     Loves our Wattmeters and uses all features.

AND THIS IS INTERESTING:
4 8410 customers love the 8410 but wished they had purchased
the 9500. I reminded them of our very attractive trade in program and that their Alpha amp may have a very high trade in value. Here is an opportunity to check with Molly at 303-473-9232 to see how reasonable it could be to move up to the 9500

Considering the risk of conclusions based upon 14 8410 owners I am going to make some anyway.
-Our Alpha Amps are designed for your feature set need. Pick the one that is best for what you want to accomplish
-Check with friends. As we say, "Ask the Ham Who Owns One"
 -Always keep in mind we have a very favorable trade in program. There is always the possibilty of getting another or different Alpha amplifier based upon your new needs.
-And, if you can afford it, buy the best.

Thanks for your continued support,  

Mike Adell