Print

[VK3BL]

Look at the end of the day we need to remember the following:

1) Directly Heated 1930s Triodes are 500-1000 hour tubes at best.  And that's filament hours, not cumulative transmission time.

2) You can STILL buy 1970s vintage Joint Army Navy RCA NOS tubes, as well as the Russian G-811 substitute.  What does this tell you?  That they made *a lot* of them.  I am sure they didn't do this just for fun; it was almost certainly done because they *never* had a particularly long life.  Maybe communist countries liked spitting out tubes just to keep people employed, but I guarantee you that was not the motivation behind all the JAN orders.

3) When they were released, NO ONE built Linear RF Amplifiers.  They were used for radiotelegraphy in Class C, or as AF Modulators in a push-pull pair in Class B.  RCA later released Linear / Class AB2 maximum ratings for ICAS usage (defined as 1/2 half the standard lifetime), based on the fact Amateurs would be happy with this whilst tubes were cheap.

4) Back when the Collins 30L1 and other 811A amps were produced, most Amateurs still had jobs (and families).  They might have used the station ~10 hours a week at most.  That's worst case 1 year of service life from a set of tubes.  This is a *very* different usage scenario from retired life where one can ragchew 100+ hours a week if they feel like it.  It should not be surprising that the Collins 30L1 is seen through rose coloured glasses, but the Ameritron AL-811 is not; they were sold decades apart, to a market that had matured.

All that said, if you are willing to accept that 811As have a relatively short service life (by today's standards) in Class ABx Linear Service, you will consider them a bargain for the money.  That said, with the advent of RF Power Transistors, changing tubes frequently does of course seem a little old hat.  Changing cheap tubes once a year isn't going to elicit many complaints, but do it every 2 months and its a different story.  We need to acknowledge that expectations AND usage patterns have changed.

Just because something seems like a technical problem by today's standards, doesn't mean it is in the context of the original design.

[G3RZP]

RCA suggested 1000 - 1500 hours at CCS ratings for 811, 809 and the like. So in a BC set up running 24 hours day every day, that's about 6 weeks life.....RCA also reckoned on amateur service being about 300 hours of transmitting a year. RCA Ham Tips, Oct 1939.

[VK3BL]

RCA suggested 1000 - 1500 hours at CCS ratings for 811, 809 and the like. So in a BC set up running 24 hours day every day, that's about 6 weeks life.....RCA also reckoned on amateur service being about 300 hours of transmitting a year. RCA Ham Tips, Oct 1939.

Thanks!  I'd been looking for an authoritative source on the lifetime for ages.  I usually got about 500-750 hours out of a set with heavy usage, at which stage I'd throw them out as they were down to around 350 watts for a set of 3.

The failure mode was always a loss of filament emission - I've got a few floating about the shack that still look NOS / unused but are only useful for decoration now. 

RCAs lasted about 30% longer than the Russian G-811 and Chinese FU-811.

[VK3BL]

RCA Ham Tips, Oct 1939 can be found here: http://n4trb.com/AmateurRadio/RCA_Ham_Tips/issues/rcahamtips0205.pdf


As aforementioned, it was designed to be utilised as a Class B AF Modulator in Push-Pull, or a Class C Telegraphy Final Amplifier, with a CCS life expectancy in the order of 1000 - 1500 hours.

Basically, lets not expect miracles from 1939 technology folks.  $80 USD or a little more will get you a quad, and you can comfortably put out 600-800 watts if attention is paid to proper tuning and technique. 

As annoying as it is to remove 18 screws and change tubes every 500 hours (ICAS), in terms of dollars spent / hour they're still a competitive tube.  The cost per hour with current market prices is around 16 US Cents, and at ~$20 a tube, its not the end of the world if you get a bad one...

[G3RZP]

The failure mode was always a loss of filament emission - I've got a few floating about the shack that still look NOS / unused but are only useful for decoration now. 

Have you tried bringing them back to life with something like twice filament voltage for 10 minutes, followed by 50% over for 20 minutes and 25% over for 20 minutes? Vary the volts and time as like, but people have claimed success in rejuvenating thoriated filament valves that way.

[W8JX]

NOT A FAULT CONDITION

Guess again. A AL811 as designed operates tubes in grey area and beyond to start with and it is not normal for plates to glow because they have so little thermal mass that once glowing it will not take much more dissipation to warp or melt them. 572's on other had have thick graphite anodes that are designed to glow without damage. Higher plate/anode temperature does increase rate of heat being radiated out of tube but a 811 is not designed for glowing as SOP while 572's are.  (as well as 3-500's too)

[KM3F]

8JX is correct, you can't compare 811 with 872 for plate dissipation robustness.
If you have one 811 out of three that glows, you have a problem with the tube or the position.
Position testing will answer both questions.
If your wondering, yes I have worked in the tube industry for two companies for type development.
I know the initial activation methods and life testing, sampling and internal tube investigations.
Doing position testing is nearly the same as done in the plant for 100% testing and sampling on a fully  instrumented test facility.
Test conditions are set up. Metering is set up for accept or reject limits on all test parameters.
The testing is done by non technical people so no bias is introduces in the testing.
Good luck.

[VK3BL]

W8JX & KM3F: Do either of you own an AL-811?

Did either of you read the supplied *RCA* reference?
Have either of you tested every brand of 811A manufactured in an AL-811, to the point of filament emission loss?

I have done all of the above, and I can promise you that a plate or two glowing under no signal key down conditions is completely normal.

Understand the AL-811 is a Class AB2 amplifier biased for ICAS usage.  As such, the standing plate dissipation is high, and as per the design trade offs, there is no circuitry to aide tube balance.

No one runs there amp with the key down and no signal, so it is NOT A PROBLEM.

And no, an 811A plate will not melt or incur any visible damage from a dull red to cherry red glow.  RCA states this, and they designed the tube.  The 811A plates can take a beating, as long as you don't do stupid things like make a coffee whilst tuning up or try and run full output on 11M AM.

[W8JX]

W8JX & KM3F: Do either of you own an AL-811?

Did either of you read the supplied *RCA* reference?
Have either of you tested every brand of 811A manufactured in an AL-811, to the point of filament emission loss?

I have done all of the above, and I can promise you that a plate or two glowing under no signal key down conditions is completely normal.

Understand the AL-811 is a Class AB2 amplifier biased for ICAS usage.  As such, the standing plate dissipation is high, and as per the design trade offs, there is no circuitry to aide tube balance.

No one runs there amp with the key down and no signal, so it is NOT A PROBLEM.

Not sure what you are smoking but a red plate on a 811 is not normal by any stretch (unless it is in a AL 811 which eats them)

[KM1H]

Did either of you read the supplied *RCA* reference?
Have either of you tested every brand of 811A manufactured in an AL-811, to the point of filament emission loss?

Have you ever used a RCA, GE, Amperex, or other US made 811A when they were still fairly new say in the 1970's and run a life test?

Did you KNOW that old JAN NOS RCA 811A's are well known to be gassy these days? Maybe also other brands.

Did you KNOW that Chinese 811's do not have the dissipation of the US versions?

Did you KNOW that Chinese 811A's are poorly/ randomly evacuated to no set standard as well as not gettered to the degree of US tubes?

Did you KNOW that Chinese 811A's do not have the reserve emission of US versions?

Did you KNOW that the 1939 811 is NOT the same tube as the ~1949 811A? Maybe the Chinese and Russians built to the 1939 design.

Required reading:
https://www.w8ji.com/811a_tube.htm

Have you or anyone else increased the bias slightly to place the tubes deeper into AB2 in order to reduce the idle current?  Ive done it numerous times with 572B's when used as AM linears. The resulting spectrum analyzer distortion on AM and SSB shows an increase well within measurement error...ie: maybe a 1dB increase. All it takes is adding 1-3 more diodes to the bias string; the actual number depends upon line voltage and the actual operating bias with no drive.

Carl

[KM3F]

VK3, I have to wonder what your smoking as well.
A red plate at 'idle' shows you there is no plate dissipation left for head room  for when the tube is RF driven.
What don't you understand about that?
Have you not saw examples of an 811 plate with big holes burned in the plate structure?
An 811 plate that is red under 'non-driven idle plate current conditions' has no safe dissipation left to offer.
811 tubes are already on the edge in these amplifiers.
All it takes is an inexperienced op to over dissipate these tubes during tuning and over drive them in service..
Good luck.

[VK3BL]

Perhaps there is a miscommunication.

I am referring to 811As used in the Ameritron Al-811 series ONLY.

I agree with Carl, if the AL-811 bias was changed, then the tubes wouldnt glow in that amp in key down no signal positions... but the AL-811 is biased for Class AB2 ICAS, so its a tube eater.

And fyi, tube dissipation is often lower under full drive than standby.  Thats why AM is so hard on linears.

And PS, its absolutely rediculous to state that a given tube shouldn't glow without taking the amplifier design into consideration.  I mean for crying aloud, the AL-811 manual tells you they are running the tubes in ICAS, and the TUBE VENDOR tells you plate glow is to be expected / accepted IN ICAS USAGE.

If you don't want the plates to glow, either change the load line so the bias current is lower, or just use them in Class B Telephony or Class C Telegraphy like they were made for!

I ran mine flat out with cherry red plates, and got around 500 hours out of a set.  Once again, THIS IS WHAT RCA STATES a ham should expect in ICAS.  I could sell them on eBay they still look so good, but have no emission left.

I repeat, WHERE IS THE PROBLEM?!  If you don't like running 811As in ICAS, or want longer life from a set, DO NOT USE AN AL-811!!!

[VK3BL]

VK3, I have to wonder what your smoking as well.
A red plate at 'idle' shows you there is no plate dissipation left for head room  for when the tube is RF driven.
What don't you understand about that?

Its not ideal, we all know that.  But its a cheap amp with cheap tubes, and 1 of the 3/4 doing more work for a while isn't the end of the world.

As I have repeatedly stated: If you are using more than 2 tubes, and biasing them at 2/3rds of rated dissipation under zero signal, you should not be surprised if one of them draws more current than the others and develops a glow. 

Thats why RCA explicitly states that an attempt should be made to balance the tubes.  W8JI did that in the original design, but the shitty tube quality meant they kept arcing and taking out the tube balancing components (grid resistor and grid bypass cap), resulting in the grids floating, and possible exciter damage. 

At some stage, Ameritron / W8JI decided the customer would be better off with an amp that had less tube balance, but a lower risk of exciter / amp damage under fault (arc) conditions. 

After all, its a cheap ICAS amplifier, with cheap tubes that the manufacturer states will glow in ICAS.

Are there any owners out there who would like to chime in?  Just key the amp up in sideband, and watch the plates.  In under a minute, I guarantee one of the tubes will be glowing.

[KM3F]

This is exactly what I have been trying to say.
I do not need to own this amplifier to look at the schematic and see all plates are connected in parallel and all grids are grounded to see that if one tube falls outside emission standards or the grid ground is missing that  it may draw more current.
That is all I have been saying but it's going by some heads.
If you took the same three tubes and built a test circuit minus the plate tank and input circuits it would amount to the same conditions without ever calling it an amplifier by any name.
You would be looking at a fixed set of test conditions.
Good luck.

[W8JX]

And fyi, tube dissipation is often lower under full drive than standby.

Must be some really strong stuff to smoke down under to make this  statement. At zero signal tubes are very close to or at continuous rated safe dissipation. It can only go one way when you drive it to max power, way up and well pasts tube ratings. Basically you are pay Russian roulette with the tubes. When you spin the cylinder and xmit, how many "bullets" in gun depends on how hard you drive it and whether gun goes off when you pulled trigger.

Thats why AM is so hard on linears.

AM (like RTTY) is hard on them because the gun is full of bullets and its just a matter if you pull trigger hard enough for it to go off.