Print

[W8JX]

I think there should be a special forum for "peanut whistle" amps.  Leave this forum open for full power amps, OK, 1200 watts and above.  1200 watts and less goes to the peanut whistle forum.  Too much bandwidth being spent on amps that can't get out of their own way.

Nothing wrong with 572 or 813 amps.  But 811 and 807 based amps could be moved.  LOL

Maybe use a benchmark that amp has to use tubes that have minimum CCS rating of at least 100 watts plate dissipation. Anything less in its own light duty amp forum (like 811 and sweep tube amps)

[K9AXN]

Its not really that controversial if you understand the tank circuit is tuned to provide maximum efficiency at maximum output.  As you decrease output from your tune point, you lose efficiency.

As efficiency decreases, plate dissipation increases.

There is nothing worse for set of 811As than tuning up at full power, then decreasing drive (say to 400 watts output) and then running a continuous mode. 

Things to think about.  Have to guess that BL is probably loading the amplifier quite a way below max out capability.  You could conceivably load the amp at or close to the point where it goes from Class A to AB mode.  Under those conditions the plates could glow at idle, then cool as drive increases.  It would seem like a silly use of an amplifier.  Why not simply go barefoot. 

If you are to exploit the output capabilities you have no choice but to load it at the max out level that you choose and sacrifice the efficiency while the tube is driven through the class A to AB operating range.

For BL, At idle there is no grid current.  If the designer said the grid resistors are to balance the tube load at idle he may not be the person that designed it.

Regarding the grid capacitors.  RCA inserted the .01uf caps in the grid circuit to equalize the inductance of the grid lead.  At approximately 7Mc the capacitor to ground provides a far better ground than a copper strap. 

Collins used 200pf grid caps to ground in the 30L1 and others followed.  That cap changed the inductive lead to capacitive throughout the ham bands and added 3db to 6db of negative feedback for linearity.  Removing them because a wayward tube shorting damages them is questionable.

Regards Jim     

[VK3BL]

For BL, At idle there is no grid current.  If the designer said the grid resistors are to balance the tube load at idle he may not be the person that designed it.
   

I hear what you're saying; I agree they would probably be of little benefit at idle, if any.  However the designer claimed the grid resistor / bypass capacitor was put in place to achieve overall balance - no specific claims about idle balance were made by them.

I personally always tuned my amplifier up to full output into a dummy load, and then reduced drive so that I made 400 watts.  All of my spent tubes looked NOS, so I reject any argument I mistuned.

That said, if I were to key the amplifier up with no drive, at least one of the plates would usually end up glowing.  If I went through my rather large 811A collection, I could obtain a bunch of 3 that load shared well enough at zero signal to not exhibit a glow at idle, but my general experience is that with any 3/4 set of fresh tubes, there is enough variance that one will end up glowing under zero signal bias conditions.

Once again, I never saw this as a significant design deficit, as nobody runs an amplifier with zero signal input for any amount of time.

Also remember when running AM, an amplifier will spend considerable time closer to carrier output levels than it will PEP; average power output during modulated AM is often 150% of carrier alone.  Efficiency is much worse at carrier level, leading to a much greater average plate dissipation.

I furnish the opinion that the AL-811 series and 811As in general should not be used in AM.  Recall, the Collins KWM-2 did not even have AM, so its matching linear the 30L1 would never have been abused in such a fashion.

If you do wish to use 811As as part of an AM transmitter, use them as push-pull class B modulators driving a class C rf amplifier.

[KM1H]

I furnish the opinion that the AL-811 series and 811As in general should not be used in AM.

 

There are many 30L1's plus Ameritron 811 amps on AM these days using low power exciters for around a 10dB gain.

A safe rule of thumb is to keep the CARRIER power to 1/4 the total plate dissipation. With carrier control modulation and/or biased deeper into Class AB2 or B, depending upon the amp, the tubes will run cooler or more output can be obtained before anode color. That is up to the user.

I have absolutely no problem running a Dentron Clipperton L with 4 x Chinese 572B's at 1200W using the (switched) bias method and steady carrier AM. An older Hunter Bandit 2000C with Chinese 3-500G's installed in place of the missing 3-400Z's will run 1200W (PS Limited) all night as will a Dentron DTR 2000L at 1500W with its original 8877.

An even older, 1964 National NCL-2000 prototype, loafs at 1200W with near fleapower exciters thanks to a very robust forced air cooled pair of 8122's.

Carl

[VK3BL]

A safe rule of thumb is to keep the CARRIER power to 1/4 the total plate dissipation. With carrier control modulation and/or biased deeper into Class AB2 or B, depending upon the amp, the tubes will run cooler or more output can be obtained before anode color. That is up to the user.

I have absolutely no problem running a Dentron Clipperton L with 4 x Chinese 572B's at 1200W using the (switched) bias method and steady carrier AM.

Thats some good advice Carl.

Out of interest do you get much plate glow at 1200 watts when running AM? 

I think its a bit of a shame amp manufacturers don't include manual bias change features; I cannot stand EBS/Dynamic Bias, and have disabled it in my AL-572.  As far as I am concerned, all it contributes is distortion when I am monitoring my transmissions with a second receiver.

I should design an aftermarket 'Bias Board'; sure might help save quite a few 811As from the landfill!

[VK3BL]

On a side note, I think we scared 1Ham off.

If you're still around, I'd love to know how you went with your amplifier.

[K9AXN]

BL this is your comment.

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

Please explain how this can be in an amplifier that is properly biased and loaded to exploit the dissipation of the tubes --- AM or Linear (AB1 or AB2).

[W8JX]

BL this is your comment.

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

Please explain how this can be in an amplifier that is properly biased and loaded to exploit the dissipation of the tubes --- AM or Linear (AB1 or AB2).

I would love to hear them explain this one too.

[W1BR]

Class A??   

[W1QJ]

BL this is your comment.

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

Please explain how this can be in an amplifier that is properly biased and loaded to exploit the dissipation of the tubes --- AM or Linear (AB1 or AB2).

=

I'll take a guess at what BL is trying to say here.  Lets assume an Al-82 amplifier.  In SSB ZSAC of .225A, plate voltage 3500v=787 watts plate dissipation ZERO WATTS OUTPUT.  Load amp up to .5amps plate current, plate V=3400v Pin=1700 watts Ameritron says the amp has 65% efficiency .65X1700=1105 Pout=1105 watts.  Plate dissipation =Pin-Pout  1700-1105=595 watts. 
So in standby, I assume he means IDLE the plate dissipation is less when the amp is making 1105 watts output than when making zero watts output.  I am assuming that is what he means.

[K9AXN]

Class A??   

Is that a question?

[G3RZP]

Could someone explain how having a reactance from grid to ground - which at some presumably lower frequency is capacitive and at higher frequencies is going to be inductive - provides negative feedback?
As in providing the equations?

I am aware that W8JI is more than dubious on this matter, and although I don't always agree with Tom, I do feel that he has an exceedlingly strong point here.

[VK3BL]

BL this is your comment.

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

Please explain how this can be in an amplifier that is properly biased and loaded to exploit the dissipation of the tubes --- AM or Linear (AB1 or AB2).

=

I'll take a guess at what BL is trying to say here.  Lets assume an Al-82 amplifier.  In SSB ZSAC of .225A, plate voltage 3500v=787 watts plate dissipation ZERO WATTS OUTPUT.  Load amp up to .5amps plate current, plate V=3400v Pin=1700 watts Ameritron says the amp has 65% efficiency .65X1700=1105 Pout=1105 watts.  Plate dissipation =Pin-Pout  1700-1105=595 watts. 
So in standby, I assume he means IDLE the plate dissipation is less when the amp is making 1105 watts output than when making zero watts output.  I am assuming that is what he means.

That is what I mean yes, and I did mean IDLE / Zero Signal rather than standby.

[KM1H]

That is what I mean yes, and I did mean IDLE / Zero Signal rather than standby.

Id suspect that 99.99% on here thought that was what you meant and didnt need to pop off about it.

Carl

[K9AXN]

Could someone explain how having a reactance from grid to ground - which at some presumably lower frequency is capacitive and at higher frequencies is going to be inductive - provides negative feedback?
As in providing the equations?

I am aware that W8JI is more than dubious on this matter, and although I don't always agree with Tom, I do feel that he has an exceedlingly strong point here.

If your referring to this comment: Collins used 200pf grid caps to ground in the 30L1 and others followed.  That cap changed the inductive lead to capacitive throughout the ham bands and added 3db to 6db of negative feedback for linearity.  Removing them because a wayward tube shorting damages them is questionable.

The inductance of the grid lead in an 811 is 40 to 50nH and combined with the 200pf cap are series resonant to ground at approximately 45Mc.  Below 45Mc the series pair behave as a capacitor to ground not inductor.  There is a bridge composed of the plate to grid 5pf and 200pf to ground that provides 1/40 ratio of Negative feedback.  The center of the bridge is attached to the grid and 200pf capacitor to ground.  The negative feedback from the 5pf capacitor is injected at junction of the grid and 200pf cap.  

Solves a few problems, the grid lead to ground is capacitive throughout the ham bands not inductive, adds negative feedback, and reduces the gain.  

W8JI does not have a strong case or any case at all.  He concocted a theory that the Collins 30S1 and 30L1 designed by Bruene and Senti generated negative feedback using a bridge composed of the cathode to grid capacity and 200pf grid to ground --- which is absurd.  This bizarre theory has been passed around for the last 40 years to discredit Orr and Senti both superb engineers.

The facts:  The 30S1 negative feedback is generated by a bridge made up of a 5pf capacitor connected from the plate to CG through a parasitic suppressor.  This is connected around the screen.  The lower leg is a 200pf capacitor to ground.  This is a 1/40 ratio of negative feedback.  It also prevents any parasitics at
the plate reaching the CG.  An eloquent design.  W8JI simply didn't see or wouldn't acknowledge the 5pf capacitor connected around the screen to the control grid.  He apparently assumed the screen would prevent any plate feedback because the screen would block it.  With this in mind an alternate reality bridge was developed.  The Cathode to grid and grid to ground.  Hence decades of a false narrative.  

The 30L1 has 5pf plate to grid, a perfect fit for the same ratio using 200pf capacitors.  The added value is the grid to ground profile becomes capacitive not inductive throughout the ham bands.  The reduced gain also reduces the need for neutralization.

Would I neutralize a GG amp --- of course.  Would I use the standard inductor wrapped around the input technique --- no because if it is neutralized at 29Mc it will become progressively under neutralized as you move to 3.5Mc because of the effect the inductance of the transformer has on capacitive reactance.  Ok, so what.  This is why the IMD and linearity of the lower bands progressively become worse using that method.    

All of this and legions more are available in (Single Sideband Principals and Circuits. Pappenfus, Bruene, and Schoenike), The Collins archives design letters, many QST articles written by Bruene, and just a bit of common sense.  

Regards Jim