Linear Amplifier Gain

[flintstone mop]

Hello Everyone
I have been reading that the gain on U.S. made linear amplifiers is set at 15dB, orders from the FCC. So that we cannot go beyond the legal limit.
What is done to change that gain? Is it just the tube picked in the PA.? Or are there other parameters involved in the circuitry surrounding the tube itself?
Thanks
Fred

[WQ9E]

Fred,

Using grounded grid circuitry is the primary reason for the gain and the tube type chosen is a contributing reason.

It is possible there is a FCC rule about gain for high power amps but I have never run across it.  The grounded grid design quickly became the near universal choice for amplifiers given its simplicity (some early designs didn't even have tuned input circuits), stability without requiring neutralization, and a number of tubes were available for GG service which provided the desired power level with the typical 100 watt output transceivers using a pair of 6146 tubes.   

Some of the earlier amplifierss will reach their rated input with very low drive since they are "traditional" grid driven designs which can be switched to class AB1 for linear operation for SSB/CW.  My Hallicrafters HT-33 and Heathkit KL-1 Chippewa can both be driven to nearly rated input using my Yaesu FT-817 QRP rig.  The Johnson Thunderbolt is similar although it was intended to be driven into AB2 for full rated input where significant grid current will occur.  My homebrew amp uses three 4CX800 tubes running as grid driven class AB1 but with these I used a passive grid input (resistive loading) which provides simplicity and stability without neutralization and it takes a little over 30 watts to drive to 1,500 watts output.

[W3RSW]

With two 3cx800a7 triodes in AB GG, you can get the same 1500 very clean watts output with 35 watts drive via a decent Q=2 tuned PI input network.

[WQ9E]

There are many ways to get 1,500 watts of clean output including solid state at this power level.  I went with a tetrode design in case I ever wanted to be able to use extremely low excitation power and with regulated screens and some cathode bias a grid driven 4CX800 is very clean.  Plus they were $25 each when I designed and built this amp. and I bought 2 sets of spares at the time so I am set for life.   Two tubes would have been sufficient at 1,500 watts output but I overbuilt the amp to stand up to continuous contest duty so the additional dissipation is just a bit of insurance.

I looked into re-tubing a recently acquired Alpha 374 with a pair of 3CX800 tubes and the cheapest new price I could find was over $800 for a pair of Chinese sourced tubes and at the price point of 3CX800 tubes I don't care to take a chance on pulls or questionable Chinese quality.    I expect that the 3-500 and 572B families will continue to be available for a long period of time at reasonable cost but there isn't much industry demand for tubes in this power range anymore so pricing/availability on a lot of transmitting tubes in this power range is going to get dicey in the future and we would all be well advised to have a set of running spares for any modern era tube type amps.

[flintstone mop]

I copied a link here, where this question came up and a responder gave the part 97.317 about gain.

"§97.317 Standards for certification of external RF power amplifiers.-

(a) To receive a grant of certification, the amplifier must: (1) Satisfy the spurious emission standards of § 97.307 (d) or (e) of this part, as applicable, when the amplifier is operated at the lesser of 1.5 kW PEP or its full output power and when the amplifier is placed in the "standby" or "off" positions while connected to the transmitter. (2) Not be capable of amplifying the input RF power (driving signal) by more than 15 dB gain. Gain is defined as the ratio of the input RF power to the output RF power of the amplifier where both power measurements are expressed in peak envelope power or mean power. (3) Exhibit no amplification (0 dB gain) between 26 MHz and 28 MHz. (b) Certification shall be denied when: (1) The Commission determines the amplifier can be used in services other than the Amateur Radio Service, or (2) The amplifier can be easily modified to operate on frequencies between 26 MHz and 28 MHz."

So, how / what limits the gain of the linear amplifier?
Fred

[WQ9E]

Fred,

Thanks for posting the info.  Design points to hit a specific gain target can include:

1.  Choice of circuit (grid driven, cathode driven, etc.)
2.  Choice of tube and specific operating conditions including overall class of operation (B, AB1, AB2) and specific element potentials used within that class.  Once you get into AB2 and B then significant grid current is being drawn so the driving source needs to provide significant driving power, the older class AB1 amps had very high gain unlike most modern amplifiers.
3.  Input attenuation
4.  probably other methods but those are what most commonly used.

I guess if the FCC were in charge of gun control they would only allow the sale of guns that used ammunition which is not available.  As is typical of most government regulation the regulations came out far too late to address the problem until it was much too late and they continue in place long after the reason for implementation is long gone.

[w1vtp]

That ruling is for the certification of and legal sale of amateur radio amplifiers.  It does not apply to anything you might build.

There is nothing wrong with building an amplifier that will bring a FT817 or the like from qrp status to full qro status.

The usual and rules of good engineering practice always applies to anything we build and put on the air

BTW, my Courier could not be legally manufactured and sold with the current ruling.

Al

[K1JJ]

Hmmm... Why would the FCC limit a commercial linear to only 15 dB of gain?  Might be a shotgun approach  to discourage CB use with a 5 watt driver.  

They already require 0 dB gain on 11 meters, so it makes sense.

This will eliminate new commercial amps using AB1 grid driven tetrodes like the 4CX-1000, etc., which can easily do 22dB of gain cleanly.

I use a homebrew class A  4CX-350J, good for a super clean -50db 3rd order IMD, that does 22 dB of gain as an IPA driver. They do have their place.  

There are a number of 100mW in, 1500 watt out, commercial Harris and Collins amps that would also fall into this category.   As already said, it's like a knee jerk gun control law.

T

[W3RSW]

Interesting.

But at the 5 and ten watt input level, high power GG amps may run almost 18 db gain.

Following from K3, LP100a, Alpha 89, actual k3 gen. 2 tone pep measurements, 238vac line.
7200 kHz, 13 April, 2014, w3rsw

K3 nominal,    Lp100a in,  Alpha out,   Gain db, 10 log (out/in).
 ( non lab precision, Rounding and sig. fig. overstatement but indicative anyway )

         Both Lp100a
W.      W        W.       db

5,      5.87,    361,    17.9
10.   10.9.     654.     17.8
15.   17.6.     856.     16.9
20.   22.1.    1102.    17.0
25.   26.3.    1215.    16.7
30.   33.4.    1365.    16.1
35.   37.6.    1489.    16.0
40.   42.5.    1560.    15.7

The nominal 15.0 db gain pt. Isn't matched until the 55/60 watt input point on the Alpha.
Output is a point best left to the student.  

Tune and load were optimized at each power output point., (as if the output power was all that was desired or on hand for a given input.) -- looks like the 15 watt input point was a little off gain curve but all rug merchants have a flaw to show the data is real.  . Like an oriental carpet from Persia.
Couldn't resist editing in this remark after rereading the thread.

[WQ9E]

Rick,

I hope you haven't set the events in motion where a group of FCC agents backed by Homeland security will soon raid Alpha headquarters. 

The oddest power rating I recall reading about was one of the EU countries, maybe Germany, limited power in the amateur service by requiring the use of final tubes whose combined plate dissipation rating (not actual dissipation) could not exceed a certain spec.  The article compared a number of tubes and found one setup that could run at 5 to 6 times their rated dissipation without jumping out of their sockets.

[W3RSW]

Yeah, that is really odd.  We hams run everything over spec.  
Sounds like your typical 811a lash up.

As to raids, no danger in that unless they raid everyone and utilize real lab grade equipment, constant voltage power supplies, etc.

....no suck down in an Alpha at low power..

Far cry from the 6jb6 feeding a quad of 6lf6's on 11 using the ol' TV transformer.

Oh and meant to add that if single tone was used, ave. plate current would be greater for 1500 watts output, hence more voltage draw down on real world supply, etc. and 15 db gain might be a closer derived figure for 1.5 kW.
I'll Have to think about it a sec. ..

[WQ9E]

Yeah, that is really odd.  We hams run everything over spec.  
Sounds like your typical 811a lash up.

Far cry from the 6jb6 feeding a quad of 6lf6's on 11 using the ol' TV transformer.

Shortly after I got my novice ticket in 1975 I saw a homebrew quad 811A amp at a hamfest and the little fins on the plates had gotten too hot and looked like they needed a dose of Viagra to return to their proper erect state

A couple of those multi-sweep tube amps showed up on ebay a few months ago.  One of the technical training schools that advertised in Popular Electronics featured one of their graduates testing these at the factory.   I bought a Galaxy V (WRL not the new CB stuff) setup a few years ago and it came complete with the optional console and their Galaxy 2000+  amp that uses 10 6HF5 tubes in parallel.  The power supply is separate and has a plug that is used to switch the HV output from 800 volts for SSB down to 600 for CW/RTTY by choosing whether the 200 volt supply is connected totem pole style with the 600V supply.  I ordered replacement filter caps for the supply but it hasn't made it on the bench yet.  Even with its linearity enhancing circuit that WRL was very proud of I have a feeling at anywhere near rated power it is about as dirty as some of the CB amps of the 70s.

[n2bc]

Lots of Collins and Harris military stuff only need 100mW input.  I have a Harris RF-110A, 100mW driving a pair of 8122s in class A, driving a pair of 4CX1500Bs in class AB1.  Rated 1000W output 24/7, 40dB gain.  At 1500W output it barely gets warm.  This is the low power end of the mil stuff.

[Opcom]

Respectfully I think the gain limit created by the FCC is intended to prevent a 12W PEP CB radio from pushing a storebought amplifier very hard.

Perhaps the low gain, along with the '10-11-12M bandwidth blackout', is an impediment to the typical plug and play CB operator.

[W3RSW]

Hey ya all I just discovered I can link pix files off'n this here Ipad.  You can actually take an instant shot after clicking the AMfone browse button.
Uh oh, yer in fer it now!

Here's one of the Guilty Gain party.

[flintstone mop]

When toooobs have handles on them; they become dangerous!!
Fred

[W3RSW]

There once was an old tube from Chantilly
  Pulled from an old amp in low Philly.
Her handles were warped; her innards were scorched,
  But that didn't bother ol' Billy.

He scoured her bright, crimped her pins tight,
  Ran ten thousand Volts through her Stripes.
Now you musn't stray beyond ten they say,
  And that's why he's king o' the night.

[n3lrx]

100 watt minimum drive doesn't stop CB'ers from running QRO, they'll just buy a 100 watt leenyar and use it as a 'modulator' [sic] Or they'll buy a Dave Made and drop a 1,500 watt maul. Where there's a will there's a way. CB'ers have more access to high power than us hams do. Some of it up to 10,000 watts. Granted no self-respecting ham would use any of the amplifiers themselves since most are just wideband amplifiers without any filtering. Garbage in, garbage out.

[WQ9E]

Granted no self-respecting ham would use any of the amplifiers themselves since most are just wideband amplifiers without any filtering. Garbage in, garbage out.

I believe in this instance that GIGO begins well ahead of the RF drive to the amp and this phenomenon is easily measured without test instrumentation at the sound wave to microphone element interface.

[n3lrx]

I believe in this instance that GIGO begins well ahead of the RF drive to the amp and this phenomenon is easily measured without test instrumentation at the sound wave to microphone element interface.

Foe-Ten Rojah right on

Thats what they call PEBCAK (Problem Exists Between Chair And Keyboard), but in this case PEBCAM.

[W3RSW]

Really the problem exists in the microphone addressing
wetware before "sounds" are even generated by the vox populii.

Sorry, couldn't help digging this up again.

Thronestone. --Rereading dead threads for fun and profit while ensconced.

[KD0HUX]

100 watt minimum drive doesn't stop CB'ers from running QRO, they'll just buy a 100 watt leenyar and use it as a 'modulator' [sic] Or they'll buy a Dave Made and drop a 1,500 watt maul. Where there's a will there's a way. CB'ers have more access to high power than us hams do. Some of it up to 10,000 watts. Granted no self-respecting ham would use any of the amplifiers themselves since most are just wideband amplifiers without any filtering. Garbage in, garbage out.