Ameritron AL-80B

more power to ya!

© 2018 by KV5R. All Rights Reserved. Rev. 2/9/2019.

Introduction

There comes a time in every hobby where we have to decide to either back out of it or get serious. Frankly, I’ve always resented the idea of having to spend a bunch of money just to overcome someone else’s noise problem, but the fact is that most people don’t want to talk to you if they can’t hear you well. That’s just the way it is. I ran barefoot (100 watts) for twenty years, with the convenience of an auto-tuner and push-button tuning. But recently I realized it was time to move up and as they say, run with big dogs. We are currently in the low part of the 11-year sunspot cycle, and most of the HF activity is on 75 meters — and 75 is a very noisy band, especially in the summer. When the distant storms are running S9+10, you need to be at least +15 or 20 if you expect to be heard.

I studied the situation for a few weeks, and finally settled on the Ameritron AL-80B amplifier. It easily runs about 800 watts PEP, using a single 3-500ZG tube. I read all the articles and reviews about it; it’s well-designed, reliable, and a good dollar-per-watt value. Everybody that has one loves it. Running a nominal 800 watts is 9dB (1½ S-units) above barefoot, and less than 3dB below the full US legal limit of 1500 watts—at about one-third the price. Sure, I’d rather have a solid-state KW and a high-power auto-tuner, but that option runs about 4 grand, and I determined to spend well under 2 for this upgrade.

I also considered the ALS-500M and the AL-811H, both of which are around $500 less than the AL-80B. But with the -500M you still have to add a 75 amp power supply, which is several hundred dollars more, and then you have a 500 watt amp that cost almost as much as the -80B. The -811H, called an 800 watt amp, is really 600, has flimsy 811 tubes that like to pop, and has no RF power meter. I did not consider the ALS-600, because it’s just way too darn expensive for 600 watts; and the -1306 even more-so. So in the end, I stand by what many others say, that the AL-80B is the best value per watt in new HF linear amplifiers. And I got it on sale for $1425 (GigaParts, no tax, free shipping), which is well below the normal $1645+shipping retail price.

Since I run a multi-band ladder line-fed dipole, and my auto-tuner is rated at 300 watts, I also purchased an MFJ-989D, their largest antenna tuner. It’s actually quite a bargain, considering that it also provides active peak-reading meters, antenna switch, a big current balun, and a dummy load, in addition to a legal-limit roller-inductor tuner. Yes, there are better antenna tuners, but only at twice the price.

Please enjoy my little “QRO Day” story in pictures. After the pictures I’ll discuss amp tuning & operating.

QRO Day


“Something strange is coming… I don’t recognize that sound…”


Yay! The wait is over.


Let’s both carry that 55-pound box…


The larger one, containing the tuner and the tube, is much lighter.


Two big boxes—now the work begins…


GigaParts shipped the tube in the big box with the 989D tuner, and it was double-boxed and well-padded in foam rubber.


The business-end of this project, the Taylor 3-500ZG tube (yes, it’s a China tube, they all are now). Just think, there’s 14.7 pounds of air on every square inch of that glass! An implosion would ruin your day, so handle with extreme care. I unpacked it on the bed, and left it there until installed.


MFJ-989D, Versa Tuner V, legal-limit antenna tuner. It also has active peak-reading meters, antenna switch, 1:1 bifilar-wound current balun, and dummy load, which if purchased separately would cost as much as this tuner. I did have to do two minor repairs, which we’ll see in another article.


Tuner’s rear panel, with connections for 2 coaxes, long-wire, and balanced feed line.


The new tuner goes under the 7100 and 993, since the AL-80B will occupy the other side of the little table. I should have opened it up and inspected it, because a few days later I discovered the turns counter was slipping and I had to pull it out of there a do a little alignment on the gears (see article). Later, I ended up putting the tuner on top of the amp, because it’s just too hard to get to the back of everything in this corner.


And there it is. Time to open it up, inspect it, and install the tube. An RF amp may be the only piece of electronics where the first thing you do is remove the cover. Never ship an amp with the tubes installed!


Rear panel input tuning slugs. Hopefully they will not need adjustment (the 15/17 did).


Input, output, keying relay jack, ALC jack, 12V accessory jack, and ALC limit control. The fuse holders are near the power cord.


Amps are hand-made, so carefully inspect it for any loose hardware or bad solder connections. The plate cap is wrapped and taped down, and the fuses are in a little package. Everything that I could check was tight, and the soldering looked good. I took a bunch of extra pictures of the insides for future reference.


The high-voltage plate cap, which is also a heat-sink to cool the tube’s anode connection.


I guess this is what passes for tube sockets nowadays. I was expecting a ceramic socket, but oh well, as long as it works… Notice the pin arrangement: there are 4 in a half-circle, and 1 by itself.


Carefully install that $220 glass thang. The manual cautions to not twist or rock it, just line up the pins and push straight down. As it spread the contacts, it landed with a surprising thump, so it might be better to put a finger under it next time.


Attach the plate cap and snug its set-screw. The manual cautions to not over-tighten it. Just snug. Be very careful here! Just having that screwdriver that close to that high-vacuum glass envelope made me nervous! That’s just for the photo; when I actually tightened the screw I had my other hand under the plate cap and screwdriver.


Years later, I’ll know when it was put in service. Always date your tubes.


Install the 2 fuses in the rear panel sockets.


Switches off.


Put meter in HV position.


Connect cables. See also my homebrew amp interface article.


Plug it in already! I put in a dedicated 20 amp circuit for it. #12 AWG cable from breaker panel to 20-amp duplex receptacle. The -80B will draw about 12 amps at 120 volts at full output. It can be re-wired for 240, but with a dedicated circuit it should not be necessary (it wasn’t). If your outlet can run a 1500-watt space heater, it’ll run the -80B just fine on 120 volts, unless you have a very long wiring run. At greater than 50 feet, I’d switch it to 240.


Switch on! No smoke, no pops, just a hair under 3200 volts, exactly what it should be. Yay! It’s alive! By the way, the forward and reflected meters are active peak-reading meters, so you see real PEP, not average, when running SSB voice. And it’s a fast-acting peak meter, not the old slow capacitor-discharge type.


The filament works, too.


And there it all is, in position, and ready to put big signals on some new friends’ meters!

So How Does it Work?

Tuning Up - First Time - Load Discovery

NOTE: This is brief overview of amplifier tuning, and a lot of things are not covered here. Please follow your manual, and do some research before tuning up for the first time. I’ve owned an SB-220 before, and I still made a thorough study of amp tuning before proceeding.

Tuning a high-power tube amp is no simple matter for newbies. An out-of-tune amp generates excess heat and burns up expensive things, as well as producing splatter. I found that reading and following the manual didn’t really give me a good feeling for the whole tuning process, so I went on-line and quickly located lots of opinions about how to tune an RF amplifier. Still not satisfied, I then discovered the writings of W8JI, who just happens to be the designer of Ameritron amps. That’s good info, and I finally got a good idea of how an amp should be properly loaded, not from sundry opinions, but from the guy who designed it.

The two controls, Plate and Load, are interactive, but they each act very differently, and must be understood:

  • The Plate control must always be peaked (max RF output), as it resonates the tube’s output circuit. It has a specific peak point for every frequency and Load position, but it does not vary with drive power. Exact setting of the Plate control is much more critical than the Load setting. It must always be peaked! Note that some people say “dip the plate current,” because it happens that at output power peak, the plate current dips a little bit. You can do it either way, since the -80B has meters for both.
  • The Load control determines the coupling between the tube and the output. It is dependent on both frequency and power level. It cannot be accurately set at low power, but must be set at maximum drive, then left there as the drive power is reduced to the normal operating level, which is about 800 watts PEP for the AL-80B.
  • Since the controls interact, the Plate must be peaked after every move of the Load.

The most important indication when tuning up an amp is the grid current. An under-coupled amp will have high grid current, which will quickly (in a few seconds) destroy the tube. One high grid current accident will cost you a cool $220. So read the manual, and W8JI’s articles, and keep your grid current down in the safe range, at all times. The maximum continuous grid current rating for the 3-500G is 110 milliamps, but for very brief max-power tune-ups, 160mA probably will not destroy it, but 200mA will. The manual says 180mA is the never-exceed, even very briefly, limit. The Load setting, as well as drive level, will control the grid current. My amp, properly loaded and driven to about 800 watts, will run the grid at about 135ma, and the plate current at about 400-450ma. Remember that that number is seen accurately only with a steady carrier, not voice, as the grid current meter shows average, not peak current. So, you tune up and see 135ma, but then talking you only see maybe 80ma. Nevertheless, the voice peaks (not seen) are hitting 135ma, but it’s very short duration, low-duty-cycle peaks.

Please don’t take my numbers as gospel, but do your own research, then carefully get to know your own tube. I have found that, on mine, around 130-140mA grid, and 400-450mA plate, produces about 800 watts. Pushing it to 1kW pushes the grid into the danger zone, around 175mA grid and 550mA plate. Even at 900, the grid current is just too high for comfort! Ameritron can call it a 1kW amp all they want to, but running it that hard will surely shorten tube life, and there’s no margin for error.

We see, then, that finding the Load setting, at maximum power, must be done in several very short tests, 3-5 seconds on, 15-20 seconds off. This process is sometimes called “load discovery,” where we discover where the Load should be set for a given frequency. We then make a chart of settings and return to them for subsequent operations, at a lower drive power. The manual gives suggested initial Plate and Load settings, but you still have to determine your own, because each tube is a little different.

You can’t just peak the Plate and Load for max output and be done. The Load is moved along little by little, and each time the Load is moved, the Plate must be re-peaked. After several short-duration max-power tests, an optimum setting will be found for the Load, which will be maximum output power, but with the grid current in the safe zone. This is repeated at several frequencies across each band, and recorded in your chart. If you’ve never done it before, it takes all day to find settings for several frequencies on all the bands. Take your time, and watch the tube’s anode color. If it’s getting orange, slow down, decrease your on-time and increase your off-time. A dull red glow is okay, but orange is getting too hot. You also need to watch your dummy load temperature and stay below its duty cycle spec.

Most tuning instructions say to reduce your drive power to a very low level, tune up, then increase it a little and tune up again, and so on, until you reach maximum power. That’s a good idea if you have no idea where the Plate and Load should be to start with, but if you start with them at the Manual’s settings, and peak the Plate at low power, then you can go straight to full power and commence the Load discovery process. In my opinion. If you feel better sneaking up on it, by all means do so, at least the first time, per band.

The reason the Load position must be determined at max power is that, as stated above, the Load position is dependant on the power level (the output impedance of the tube changes with power). For example, your Load might be best at 4 when driving the amp to 800 watts, but at 3 when driving it to 500 watts. The problem is that if you use a Load position for 500 watts, then drive the amp higher, it will be “under-coupled” and non-linear, have high grid current, and produce splatter. So the Load position is determined at maximum power, then left there for operation at a lower power, so it’s always slightly over-coupled. This ensures that the amp is never driven into non-linearity, or hits an excessive grid current level. Being a bit over-coupled is always better than being under-coupled. Also, if the Load is set way too low, the tube will arc internally and blow a hole in the grid, so always err toward a higher Load setting. Furthermore, severe under-coupling can arc your tuning capacitors, which leaves rough spots on them that will easily arc again.

Once Load discovery is done and the settings are recorded in a chart, you shouldn’t need to do that max-power Load discovery thing again (until you change the tube). Do your Load discovery into a dummy load, and then always operate into a 50-Ohm load (antenna tuned to flat SWR), and your Load settings will be right, provided you determined them at max power. How many Load settings do you need to determine? I don’t really know, but I recorded a Load (and Plate) setting every 25 kHz across every band. After using it a while, I think every 50 kHz would be sufficient, as you can easily interpolate for in-between settings. The bandwith for Load settings is fairly wide, but the bandwidth for Plate settings is very narrow. But the more data points you have in your chart, the closer your settings will be before you even hit the key. You’ll then set the Plate and Load according to the chart, then briefly key up and peak the Plate, which can be done at low or normal operating power. But remember: Never peak the LOAD at reduced power! Use the charted setting you determined at maximum power.

Normal Operation - How I do It

  1. Power up the amp and check the BAND switch and STBY (standby) switch.
  2. Check the frequency and do not QRM. On LSB move 1kHz above, and on USB 1kHz below, the active QSO. That puts your carrier in their opposite sideband, where it will be suppressed. But make sure you’re not 1kHz into an adjacent QSO! A panadapter is really handy here.
  3. Set radio to RTTY, and set output to about 70 watts (for an AL-80B). Or you can leave the radio at full output but reduce its power with the ALC feedback loop from the amplifier. I prefer to keep the radio output at 70, then set the ALC on the amp to reduce it just a tiny bit more. This ensures that the amp is never over-driven.
  4. Key the radio and tune the antenna tuner to a flat 1:1 SWR. You should also have tuner settings in your chart, so you are very close before transmitting. ALWAYS set your tuner at low (barefoot) power! And do NOT roll the inductor while transmitting more than just a few watts!
  5. Set the Plate and Load controls to the chart, as previously determined at max output, for the current frequency.
  6. Put the amp on-line (switch to Operate).
  7. Key the radio and amp and quickly peak the Plate. It is more sensitive than the Load control, so you can’t just use the chart setting. The Load, however, will be fine as set by the chart.
  8. Briefly key once again and check output power and grid current, just to make sure everything is nominal.
  9. Switch to SSB and operate!

It always sounds (and reads) a lot more complicated than it is, but after a little while it becomes second-nature, and is no more difficult than unlocking your car, getting in, cranking up, and driving away without crashing. You follow a procedure, develop a habit, then do it the same way every time.

Remember that driving any amp to its maximum is a bad idea. Others will not see or hear any difference between 750 and 1000 watts, so do your tube a favor and run it 20-30% below maximum rated output. As they say, if your car will do 120 MPH, would you drive that fast all the time? But people buy an amp that says 1,000 Watts in the ads, and expect to run it at 1,000 watts forever! 750-800 is a good speed limit for the -80B, and that’s only half an S-unit (and $2500) below legal limit.

Another thing to consider is duty cycle. If an amp will comfortably do 800 watts on SSB voice peaks, that’s a light duty cycle. Running a higher duty cycle mode, like CW, requires a reduction in power (500 watts), and running a very high (100%) duty cycle (RTTY) requires a further reduction, to well under 400 watts. For AM, the AL-80B manual says do not exceed 200 watts carrier power.

NEVER reduce power by de-tuning the amp! Reduce the radio output, either with the radio’s RF Power control, or the ALC on the amplifier, or both.

About ALC: You may hear some hams saying they never connect ALC. That’s just plain stupid! Connecting and using the amp’s ALC allows the amp to control the radio, and thus protect itself from over-drive. Even if you don’t want any ALC action from the amp, you can set it right above your normal operating output level and it’ll be there to protect the amp if you accidently transmit with the radio at full output.

Conclusion

After a year of use, I’m very happy with the AL-80B. It’s a workhorse, is well-designed, and once that Load Discovery bit is done, very easy to use. I wish it was a solid-state amp, but it’s around $2000 less than an ALS-1306, so I don’t mind twisting knobs occasionally. The only thing I don’t like about it is it’s big and heavy!

On-air reports are good; when switching it on, the signal goes way up, with no distortion or change of audio quality, just as it should. Also, I have recently built an RF current sampler and connected it to my oscilloscope, and the signal is nice and clean.

Please see the other articles in this series, where I added an I.F. panadapter to the 7100, built an amplifier interface, a salt-water dummy load, and repaired the 989D.

Next, we’ll look at some accessories you can build for your amp, and save some money.

73, — KV5R

Leave a Reply

Your email address will not be published. Required fields are marked *