Author Topic: Linear Amplifiers  (Read 11021 times)


  • Global Moderator
  • Hero Member
  • *****
  • Posts: 1940
Linear Amplifiers
« on: December 26, 2012, 02:49:40 PM »
Linear Amplifiers

Being an avid Dxer I have owned a linear amp for several decades, yes I have upgraded at various times and during the course of several decades have either built, or serviced several models of linear amps.

For those of you interested in your first, or used amp hopefully my experiences will be helpful in your choice.

My first amp I built way back in the mid 70s and it was a Heathkit SB220 took a full week to build it and it served me well for many years. Of course these are still available today at Ham fests and web sites selling ham radio gear. There are a few things to consider with this amp; yes you can find it at a modest price. However, Heathkit is long gone and any product support has vanished.

The SB 220 used a pair of 3-500 z tubes, and it did produce about 600 to 800 watts output, remember this model was rated by the old standard and that was 1Kw input. Plate voltage times plate current; normally the SB 220 at legal limit then produced 2000 volts at 500 ma. 1KW input, on the ssb mode it would average over 1kw input and reached about a kW peak output, or a little more.

In the ssb mode the plate voltage was raised to about 2500 volts with load, so the output increased, thing is these units worked.

If you find one of these units, you will have to interface it with your more modern rig. Your newer rig has a small relay for switching the t/r line in the amplifier. In vintage models they used a 120-volt relay and switched one side of the A/C to activate the T/R relay in the amplifier.

In the new rigs, only a contact closure is needed since we are now using low voltage logic circuits to switch the T/R relay. Unless you have an interface box such as the Amp Keyer, or other device in line to switch 120 A/C and connect one of these old amps to your modern transceiver, well the results could be catastrophic! 

You will fry the small relay, do damage to the circuit board holding it and may cost you several hundred dollars in repair.

A box called the Amp keyer is available from the Heathkit shop:

The SB220 ran fine until you had an arc in one of the finals, as a result of this it would blow the regulator diodes on the filament supply. These diodes were located behind the two meters on the front panel. Another problem was that of the meter shunts for the high voltage meter, the shunts were of the old carbon resistors and ranged 10 to 20 megohms in value. These resistors in time would drift way out of the tolerance level; normally they were about a 10% tolerance. However, they would drift much more then that figure, the results were the High voltage readings were not correct, either the meter indicated a low High voltage, or much more then the actual figure. There was not a problem with the power supply at all, rather it was the meter shunts changing in value and giving an erroneous reading. Normally, using the cw/ssb select for cw you should see 2000 volts under load at 500 ma, 1 Kw Input and about 700, to 800 watts output. In the SSB position you should see 1 Kw output or more.

Heathkit also produced another model the SB 200; this one had an output of 1200 watts PEP, or about six hundred watts in the key down mode. The SB 200 used 572B final tubes and there are two of them mounted horizontally, one problem I caught with this model that belonged to my deceased brother law.

I serviced this unit for him due to the fact it was locking up in transmit mode, as soon as you turned it on, it went into transmit mode and would not drop out to receive mode. My brother in law thinking it was a problem with the T/R line had completely disconnected the entire line in an effort to find the problem.

It came down to a shorted final, it was a grid to cathode short in one of the 572B tubes, and this was found with a simple ohmmeter check on the pins of the tubes. Initially, I removed one tube, then the other and checked see if the t/r opened to receive mode. Once I found the tube, some ohmmeter checks between the pins revealed the culprit.

You see the short caused completion of the T/R line and completed the circuit; hence closing the T/R relay and remaining in transmit condition.

In the SB 200 you need tubes that are made for Horizontal mounting, regular tubes will have element sag due to gravity, eventually shorts will result and the effects become obvious.

The tubes are available from RF parts, and if you can find these two models in good condition that is no modifications and a good construction job you are in business. Use them for as long as possible, power transformers are available from Peter Dahl Company, but are quite expensive. In this scenario, you may pay more for the transformer then you did for the whole unit.

Heathkit also made a SB 230 same basic amp, different cosmetics, again take your chances.

Consider also the aging filter capacitor banks in these units; last I heard there was a replacement board available from the Heathkit shop for a nominal price. It is well worth the effort to consider this upgrade.

R.L. Drake Company for many years produced high quality ham radio gear, Drake got out of the ham radio business about twenty-five years ago. However, their L4B, L7 series amps are still around today and can be had for a modest price.

Unlike the Heath amps, the Drake had a separate power supply, it was not in the deck, but a rather heavy separate unit connected by a screw on connector covered by a rubber boot for the high voltage lead. The rest of the required connections were a Cinch Jones plug with twelve or more connections that plugged into the back of the deck.

The L4B and L7 used 3-500 z tubes in the finals, again, these units are rated in the old standard, one KW input, however in the ssb mode you will see about 1200 to 1400 watts output. This of course is due to the higher plate voltage in the ssb mode. Normally, in the cw mode about 800 watts output, some people have claimed tubes are bad if they do not see 1KW output in this mode.

Not so, read the manual, and if you have a device that is one hundred percent efficient you better market it. You can make a million dollars over night. I KW input will produce about 800 watts output on the cw mode. This is based on what class the tube is biased, most amps use AB2, some use Class B and cw amps may switch to Class C.

An indication of bad tubes, or aging tubes are the following, in key down and full output and proper loading does the power sag after the amp has been in key down for about thirty seconds or longer? Does the grid current rise, or continue to climb during this test?

Is the grid current high, near, or over the maximum amount?

Are filaments lit for both tubes? Lastly, are there any arcs, or flashes from any of the tube finals?

Grid current readings are the best indication of aging, or weak tubes, If the power output sags under full load and continues to sag, the grid current continues to rise, chances are you have a soft tube, or tubes.

The Drake Power supplies have two boards, one on each end of the power supply; they contain the rectifier stacks and the filter capacitor banks. However, as with all amps of this era, those caps are at the end of their useful life. If you can find the required value and working voltage namely 450V they will be quite expensive and will be NOS. That is New Old Stock, so you may not be any better off then with the old capacitors.

At one time, and I assume they are still available is a replacement board from the Heathkit shop, I bought the board and replaced the old aging caps for about $120. It worked fine, very easy to replace, all mounting holes match, in fact it drops in the mounting holes for the old boards.

All eight caps and sixteen diodes are on one board, you remove the old ones, reconnect the primary and HV leads to the power supply and you are back in business.

As far as problems with the L4 and L7, two areas surfaced, one the T/R relay needs cleaning, the contacts become soiled, or tarnished. This causes loss of receive signals, or an intermittent loss of signals. Second problem is that of the feed thru caps in the finals, sometimes they split, it causes one final to not be driven, these were available from Drake for a while, but they may no longer be available, check and see.

If you have the old boards and power supply, on occasions the rectifier boards would arc at a connection of the diodes, this caused the board to burn; it created an awful stink in the house. If the boards are still available, they costs $20.00 each from Drake, there is no repair after they burn, only replacement, so take your chances, or buy the replacement board from the Heathkit shop if it is still available.

One other problem noted with these amplifiers was the dual rocker switchers, one was used for power on and off while the other one selected the cw/ssb mode. There was one design problem with the on off switch. Across the terminals of this switch you hot switched the 220 ac in order to turn on the amp. However, the contact points of the switch in time would weld together and you could not turn off the amp.

For a short period of time you could buy the assembly from Drake, however I managed to get one of the remaining assemblies from Drake. I replaced my ailing switches and this is about a two-hour job. You must remove the front panel, then remove the switch, replace the switch and replace the front panel. The original switch assembly was available only once, I replaced these switches for two friends of mine and I was sent an assembly that was not red and black, but all white. With it you received a modification kit, yes it worked fine, but not the original red and black rocker.

One of the modifications offered was to remove the 220 AC line from the switch, then use a relay of lower voltage controlled by the switch to actually switch the 220 ac. If you have the original rocker assemblies in your L4 or L7 you have certainly gotten your money’s worth out of them.

I used my L4 for years before selling it out as is; the power transformer gave up the ghost with an internal short, replacing it would have cost more then the original cost of the amp. The only source was Peter Dahl Company and the cost was quite a large sum of money.

Kenwood T L 922, this is by far in my opinion the best of the lot, a brute of an amp and will run for years without a problem. Only thing is it is somewhat noisy, the power transformer hums and the blower fan creates an additional noise. Couple this with the noisy t/r relay and you have a cacophony of racket on your operating desk. However, this amp will perform for you and do it for years.

Again, there is no product support Kenwood no longer stocks parts for this amp. After well over a decade with zero problems, I had an arc over and it trashed the plate meter and the shunt board. I found the meter, replaced it, the shunt board I had a friend of mine who could fabricate one, once done it was a matter of replacing all the parts, moving them from the old board to the new one.

After all this, I found another problem with the one of the RF chokes to the final tubes; I was only getting power out of one tube. Again, the search began and I finally found the part I needed. By this time, I felt I had gotten all the usefulness I could from this unit and sold it off to a ham in St. Louis who collected TL 922 amps.

By swiping parts out of really dead units, this fellow could repair some of the amps and sell them to hams wanting a modest priced linear. The sale price of mine equaled about what I paid for it at a ham shop about ten years ago. The TL 922 can be found for around a $1000 and if you find a good one, hold on to it and use it.

There is no wattmeter in the 922, just a relative power meter in my opinion rather useless, of course if it changes abruptly, one can assume a problem. But, really Kenwood you could have included a wattmeter.

My brother in law brought me a Collins 30L1 to check over for him, he had found this gem at an estate sale. It was full of dust and dirt, but once I cleaned out the filth, fired it up it came to life. I advised him to consider the replacement power supply board offered by the Heathkit shop. The Collins quality always reflects excellent design, long dependability, and a product that will give you many years of service.

If memory serves me correctly I think this amp used 811A tubes, or 572B, in the final, PEP was 1200 watts.

About a year ago, I made the decision about tube amps, replacing tubes has become very expensive, 3-500 z tubes have doubled in price over the last few years. Ceramic tubes although they last much longer then glass tubes are ridiculous in price. I had admired the Tokyo High Power HL 2.5 model. It is solid state, no tune and built with quality, fit and finish in the unit.

I had written a product review of this unit on the former W9OG forum site and received a lot of interest. It is a great amp, no arcs, makes 2kw key down output with ease. However, the recommended level is 1500 watts and indeed it makes this with no problem.

I did find a good bargain for a reconditioned Alpha 86 as well; I have two stations one for 20 where I use the THP and the other for 40, where the Alpha 86 is now employed.

The Alpha so far has performed, they have a great reputation, this amp will make 1500 watts with ease and can do 2300 watts output without a problem. Only thing is the blower fan is somewhat noisy, but I can live with since I rarely use a speaker in the shack. The 86 uses a pair of ceramic tubes 3CX 800 and they are brutes. Unlike other amps the 86 have a lot of protection circuits that protect the tubes and power supply.

In summation you can expect over time for a linear to go down for some reason, tube shorts, internal arcs on the tubes, high voltage break downs, or even filter caps going bad over years of usage.  After all, they are high voltage devices and the odds are against you from the very first time you plug it in and turn on the amp.

I used all of my previously owned amps in contests and long duration, they proved them selves for reliability and dependability. In fact the number of hours I racked up on each unit during heavy contests, were about three times the normal usage most ops use their amplifiers. In normal operating I used my amps every day, so my demand is much more then an average op.

I may have missed some brands for the most part I have either serviced, or used the ones mentioned. If you are in the market for a used amp and one of vintage ratings consider my comments and the problems I serviced.

If you decide to make the jump and go for a new one, I can certainly recommend the THP HL 2.5 kw and the series of Alpha Amps on the market today. In time, I am sure all manufactures will convert to solid state units. It just makes sense due to the increasing cost of tubes.

A linear amp can open the DX flood gates, it can make the difference in an average strength signal, or one that stands out in the crowd. The choice is yours, there are plenty to chose from, but be careful in your choice.

If you have questions, be sure and ask, no problem.

Bill, K4LRX


  • Global Moderator
  • Newbie
  • *****
  • Posts: 5
Re: Linear Amplifiers
« Reply #1 on: January 21, 2013, 09:40:53 PM »
Good article Bill.  I've owned a Collins 30S1 and an Alpha 78.  I liked the Alpha 78 because it is so simple to tuneup, 2000w out.  Today I'm barefoot!


  • Newbie
  • *
  • Posts: 2
Re: Linear Amplifiers
« Reply #2 on: November 30, 2013, 06:37:46 AM »
My heart is set on an Alpha because of their excellent design and build quality. I'm not sure if ETO or the new owners build better equipment. It seems ETO amps are still in demand.  The only HF amp I've used was a Dentron Clipperton using 4 572Bs. But after seeing the plate voltage sag under load, and not considering this was used as negative feedback to stabilize the amp, I designed a beefed up full bridge rectifier, which replaced their 1/2 bridge voltage doubler with a boat load of filtering caps. In addition, I had a 240 vac line installed, rather than using the 110 vac. The plate voltage was as solid as a rock.  No more sag, but oscillation melted the chokes resulting in a number of exploded tubes. It sure did put out 2X the original power. Next time, I should have not pushed it so hard. I had stations in Japan and Russia reporting S9 plus 40db, louder than their neighbor down the street on 15m.