Tube Linear Amp design for basic K2

classic Classic list List threaded Threaded
7 messages Options
Reply | Threaded
Open this post in threaded view
|

Tube Linear Amp design for basic K2

Wyn Hughes
Greetings from a newbie K2 builder. I am currently building #5390. Into phase 3 now, somewhere in the early stage of the rf board. Don't get too much time for this detailed work and my xyl has made me clear out the spare bedroom in our apartment (aka construction area), as she does not seem to like the lead droppings and other bits and pieces on the floor, while one of our daughters and young toddler are visiting.

While the build is only a work in progress, and although it passed the stage 1 and stage 2 tests it's not over till it's over, my mind is turning to the question how to exploit the K2 when it finally gets launched. Let me say, I have nothing against qrp. The main reason I bought the kit (instead of an Argo V), was the desire to have a low current drain rig, to take on /P. I don't have much of an antenna sytem at this VR2 location. I have a Icom IC-781 which serves as the main rig at this station, but if the K2 works, I would like to try it as a main station rig with decent power as well as a /P rig.

Here's the question. My only amp at this location is a Drale L4. It is just a 10db gain amp. So it would only produce at most, 150 watts driven by the K2. In practice, probably less. I have never driven a grounded grid triode amp sccessfully with such low power with any success. How does one drive a 10-15 watt output radio to serious output without an intermediate amp?

I have designed and built many amps and repaired many others but not before an amp with a projected power gain of 20db. They were all either grounded grid triode affairs or a G2DAF (pair of 813), which is pretty similar. My basic thinking is a high mu tetrode amp, grid driven, preferably with a swamp resistor as input but, if needed, a tuned input and neutralized. I love glass tubes and am shy of using a 4cX400a or 4cx800a type tube, as I hated my ex-Alpha 01b, although I think the tubes would probably work ok but uninspiringly with the right design. I would really like to try a 4-1000a or even a 4-400a in that configuration. I have no issues with working on high voltage and plan to use the 2.5kv supply from the L4, though the 4-1000a does benefit from a higher anode voltage if available.

I would be interested to hear from anyone else who has gone down this road and of their experiences.

Meanwhile, best 73s and GL to you all from Hong Kong.

Wyn
VR2AX (ex VS6UK)
GW3YGH
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

Reply | Threaded
Open this post in threaded view
|

Re: Tube Linear Amp design for basic K2

Charles Greene-2
Wyn,

I have an antique Hallicrafters HT33A with a PL172 tube, and  I
modified it so it will accept lower drive.  Right now when I drive it
with about 20 watts from my K2/100 in PSK31, I get about 400 watts
out.  That's a gain of 20 times, or 13 dB.  The tube input is swamped
with a 50 ohm resistor to help in its stability.  The PL172 is a
power pentode with 1000 watts plate dissipation, operated in the
class AB1 mode (no control grid current) when amplifying SSB.  On CW,
it operates in the class AB2 mode, with a little higher
efficiency,   A 4-1000A is the nearest new(er) design tube.  I
thought about how I could make the old guy put out something exciting
when driven by my K2.  I could swamp the input with a 200 ohm
resistor and use a 4:1 balun and the impedance to the K2 would still
be 50 ohms.  Actually, with the K2 automatic antenna tuner, you could
tune the 400  ohms to 50 ohms.  Same difference.  The tube itself
does not take any power in class AB1, but the input voltage needs to
be so much for a given output.  A 4:1 balun would increase the drive
voltage by a factor of 4 times which would increase the output of the
amp to 1600 watts if the drive remained the same, 800 watts with 10
watts drive, or 400 watts PSK31 with the K2 maxed out at 5 watts,
which is respectful.  From 20 watts to 1600 watts is a power gain of
19 dB which in my opinion is a little high.  I would be more
comfortable with a power gain of 16 dB, but you can try it.

The point is, yes you can find a tube that will take a 10 watt
transmitter to 800 watts or so, but be careful.  Those tubes are expensive.

73,  Chas W1CG


At 09:17 AM 6/9/2006, Wyn wrote:

>Greetings from a newbie K2 builder. I am currently building #5390.
>Into phase 3 now, somewhere in the early stage of the rf board.
>Don't get too much time for this detailed work and my xyl has made
>me clear out the spare bedroom in our apartment (aka construction
>area), as she does not seem to like the lead droppings and other
>bits and pieces on the floor, while one of our daughters and young
>toddler are visiting.
>
>While the build is only a work in progress, and although it passed
>the stage 1 and stage 2 tests it's not over till it's over, my mind
>is turning to the question how to exploit the K2 when it finally
>gets launched. Let me say, I have nothing against qrp. The main
>reason I bought the kit (instead of an Argo V), was the desire to
>have a low current drain rig, to take on /P. I don't have much of an
>antenna sytem at this VR2 location. I have a Icom IC-781 which
>serves as the main rig at this station, but if the K2 works, I would
>like to try it as a main station rig with decent power as well as a /P rig.
>
>Here's the question. My only amp at this location is a Drale L4. It
>is just a 10db gain amp. So it would only produce at most, 150 watts
>driven by the K2. In practice, probably less. I have never driven a
>grounded grid triode amp sccessfully with such low power with any
>success. How does one drive a 10-15 watt output radio to serious
>output without an intermediate amp?
>
>I have designed and built many amps and repaired many others but not
>before an amp with a projected power gain of 20db. They were all
>either grounded grid triode affairs or a G2DAF (pair of 813), which
>is pretty similar. My basic thinking is a high mu tetrode amp, grid
>driven, preferably with a swamp resistor as input but, if needed, a
>tuned input and neutralized. I love glass tubes and am shy of using
>a 4cX400a or 4cx800a type tube, as I hated my ex-Alpha 01b, although
>I think the tubes would probably work ok but uninspiringly with the
>right design. I would really like to try a 4-1000a or even a 4-400a
>in that configuration. I have no issues with working on high voltage
>and plan to use the 2.5kv supply from the L4, though the 4-1000a
>does benefit from a higher anode voltage if available.
>
>I would be interested to hear from anyone else who has gone down
>this road and of their experiences.
>
>Meanwhile, best 73s and GL to you all from Hong Kong.
>
>Wyn
>VR2AX (ex VS6UK)
>GW3YGH

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

Reply | Threaded
Open this post in threaded view
|

Re: Tube Linear Amp design for basic K2

Robert Allbright
In reply to this post by Wyn Hughes
Hi Chas

Thank you for your interesting info.

I was surprised that you mention using 400w on PSK31

I've never seen anyone mention using more than about 70w on PSK31 as  
high powers are not supposed to be necessary?

Just my comment as when I experimentally tuned up my Acom 1010 amp  
for 400w output I had a very irate neighbour knocking on door!

That's another story and I'm still alive hi!

Since then I've been using 5 - 20 watts cw and psk31 and having fun

73 Robert G3RCE (from the future Islamic Republic of the UK - God  
help us!)

------------------------------------------------------------------------
---------------------

From: Charles Greene <[hidden email]>
Subject: Re: [Elecraft] Tube Linear Amp design for basic K2
To: "Wyn" <[hidden email]>, <[hidden email]>
Message-ID: <[hidden email]>
Content-Type: text/plain; charset="us-ascii"; format=flowed

Wyn,

I have an antique Hallicrafters HT33A with a PL172 tube, and  I
modified it so it will accept lower drive.  Right now when I drive it
with about 20 watts from my K2/100 in PSK31, I get about 400 watts
out.  That's a gain of 20 times, or 13 dB.  The tube input is swamped
with a 50 ohm resistor to help in its stability.  The PL172 is a
power pentode with 1000 watts plate dissipation, operated in the
class AB1 mode (no control grid current) when amplifying SSB.  On CW,
it operates in the class AB2 mode, with a little higher
efficiency,   A 4-1000A is the nearest new(er) design tube.  I
thought about how I could make the old guy put out something exciting
when driven by my K2.  I could swamp the input with a 200 ohm
resistor and use a 4:1 balun and the impedance to the K2 would still
be 50 ohms.  Actually, with the K2 automatic antenna tuner, you could
tune the 400  ohms to 50 ohms.  Same difference.  The tube itself
does not take any power in class AB1, but the input voltage needs to
be so much for a given output.  A 4:1 balun would increase the drive
voltage by a factor of 4 times which would increase the output of the
amp to 1600 watts if the drive remained the same, 800 watts with 10
watts drive, or 400 watts PSK31 with the K2 maxed out at 5 watts,
which is respectful.  From 20 watts to 1600 watts is a power gain of
19 dB which in my opinion is a little high.  I would be more
comfortable with a power gain of 16 dB, but you can try it.

The point is, yes you can find a tube that will take a 10 watt
transmitter to 800 watts or so, but be careful.  Those tubes are  
expensive.

73,  Chas W1CG
------------------------------------
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

Reply | Threaded
Open this post in threaded view
|

QRO/QRP (WAS: Tube Linear Amp design for basic K2)

Ron D'Eau Claire-2
Robert G3RCE wrote:
I was surprised that you mention using 400w on PSK31

I've never seen anyone mention using more than about 70w on PSK31 as  
high powers are not supposed to be necessary?

-------------------------------

Truth be told, one can work the world using 5 watts (or less) on CW too, but
what percentage of our fraternity do that?

Why use QRO? Because it's there! It needs no further justification by most.
And, on occasion, it'll pull through copy where copy isn't possible at lower
power.

Whether QRO is really "needed" or not is a whole different issue that
changes from moment to moment. For better or worse, virtually every Ham I've
met runs max power almost all the time. The only exception are some with
kilowatt amps they do occasionally turn off. Still, how often do you hear
someone say, "PSE QRO" or, if sig reports run 599, "PSE QRP"??

Actually I'm a bit of a stealth QRP operator. If I get a 599 report, I'm
very likely to turn the power down by half, no matter what I'm running. Then
I'll just ask if my signal is still FB. If so, I might cut it by half again.
Each halving is about 1/2 S-unit so if I do it twice my signal is down an
S-unit on  most receivers. Most of the time that's barely noticeable. It
works the same if the initial power is a kilowatt, 100 watts or 10 watts.
One of the interesting operating gambits many QRPers enjoy is to QRP
together during a QSO so see "how low they can go". The results are often
astounding, getting down  into the milliwatt range with perfect copy.

One of the issues with PSK is that most software includes an AGC function to
keep the audio levels in the soundcard under control. A really loud signal
will then suppress the gain and mask others since most PSK operators use
wide-open I.F. bandwidths so they can see all the nearby frequencies on the
waterfall display.

That's how some of us like to operate CW using wide bandwidths and letting
our ears to do the filtering. And we've learned that one can't run AGC doing
that if there's a very strong signal in the passband. More and more PSK
operators are discovering the same thing, whether it's a 5 watt signal from
next door or the kilowatt farther away.

As more QRO shows up on PSK, PSK techniques will have to change too: either
filter out all but the one signal being received in the I.F. or by using
limiting or some similar process like those of us who like wide-bandwidth on
CW use.

Ron AC7AC

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

Reply | Threaded
Open this post in threaded view
|

Re: Tube Linear Amp design for basic K2

Wyn Hughes
In reply to this post by Charles Greene-2
Chas,

Thanks for your comments. It's interesting to see that someone else has come
to some very similar fairly similar basic conclusions as oneself.

I have emailed RF Parts for the current price of the 4-1000a. At this stage,
it may remain a drwing board exercise..

73s,
Wyn


----- Original Message -----
From: "Charles Greene" <[hidden email]>
To: "Wyn" <[hidden email]>; <[hidden email]>
Sent: Saturday, June 10, 2006 3:46 AM
Subject: Re: [Elecraft] Tube Linear Amp design for basic K2


> Wyn,
>
....   A 4-1000A is the nearest new(er) design tube.  I

> thought about how I could make the old guy put out something exciting when
> driven by my K2.  I could swamp the input with a 200 ohm resistor and use
> a 4:1 balun and the impedance to the K2 would still be 50 ohms.  ......
> The tube itself does not take any power in class AB1, but the input
> voltage needs to be so much for a given output.  A 4:1 balun would
> increase the drive voltage by a factor of 4 times which would increase the
> output of the amp to 1600 watts if the drive remained the same, 800 watts
> with 10 watts drive......  From 20 watts to 1600 watts is a power gain of
> 19 dB which in my opinion is a little high.  I would be more comfortable
> with a power gain of 16 dB, but you can try it.
>
>.......  Those tubes are expensive.
>
> 73,  Chas W1CG

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

Reply | Threaded
Open this post in threaded view
|

Re: Tube Linear Amp design for basic K2

Charles Greene-2
In reply to this post by Robert Allbright
Robert,

I worked a guy the other day using 400 watts, but it is very
unusual.  I usually use my K2 /100 at 25 watts and my SDR-1000 at 35 watts.

I made an error on the calculation of power output of the linear
amplifier.  If you double the input voltage, that is a multiply of 4
times the input power, as power = E^2 / R.  That also produces 4
times the output of the linear amp.  Bottom line, if I use an input
swamping resistor of 100 ohms, I can drive the amp to 400 watts
output with 5 watts.  Useful addition for a plain ordinary K2.  I
haven't tried it, just calculations.

73,  Chas, W1CG

At 02:20 PM 6/10/2006, Robert Allbright wrote:

>Hi Chas
>
>Thank you for your interesting info.
>
>I was surprised that you mention using 400w on PSK31
>
>I've never seen anyone mention using more than about 70w on PSK31 as
>high powers are not supposed to be necessary?
>
>Just my comment as when I experimentally tuned up my Acom 1010 amp
>for 400w output I had a very irate neighbour knocking on door!
>
>That's another story and I'm still alive hi!
>
>Since then I've been using 5 - 20 watts cw and psk31 and having fun
>
>73 Robert G3RCE (from the future Islamic Republic of the UK - God
>help us!)
>
>------------------------------------------------------------------------
>---------------------
>
>From: Charles Greene <[hidden email]>
>Subject: Re: [Elecraft] Tube Linear Amp design for basic K2
>To: "Wyn" <[hidden email]>, <[hidden email]>
>Message-ID: <[hidden email]>
>Content-Type: text/plain; charset="us-ascii"; format=flowed
>
>Wyn,
>
>I have an antique Hallicrafters HT33A with a PL172 tube, and  I
>modified it so it will accept lower drive.  Right now when I drive it
>with about 20 watts from my K2/100 in PSK31, I get about 400 watts
>out.  That's a gain of 20 times, or 13 dB.  The tube input is swamped
>with a 50 ohm resistor to help in its stability.  The PL172 is a
>power pentode with 1000 watts plate dissipation, operated in the
>class AB1 mode (no control grid current) when amplifying SSB.  On CW,
>it operates in the class AB2 mode, with a little higher
>efficiency,   A 4-1000A is the nearest new(er) design tube.  I
>thought about how I could make the old guy put out something exciting
>when driven by my K2.  I could swamp the input with a 200 ohm
>resistor and use a 4:1 balun and the impedance to the K2 would still
>be 50 ohms.  Actually, with the K2 automatic antenna tuner, you could
>tune the 400  ohms to 50 ohms.  Same difference.  The tube itself
>does not take any power in class AB1, but the input voltage needs to
>be so much for a given output.  A 4:1 balun would increase the drive
>voltage by a factor of 4 times which would increase the output of the
>amp to 1600 watts if the drive remained the same, 800 watts with 10
>watts drive, or 400 watts PSK31 with the K2 maxed out at 5 watts,
>which is respectful.  From 20 watts to 1600 watts is a power gain of
>19 dB which in my opinion is a little high.  I would be more
>comfortable with a power gain of 16 dB, but you can try it.
>
>The point is, yes you can find a tube that will take a 10 watt
>transmitter to 800 watts or so, but be careful.  Those tubes are
>expensive.
>
>73,  Chas W1CG
>------------------------------------
>_______________________________________________
>Elecraft mailing list
>Post to: [hidden email]
>You must be a subscriber to post to the list.
>Subscriber Info (Addr. Change, sub, unsub etc.):
>http://mailman.qth.net/mailman/listinfo/elecraft
>
>Help: http://mailman.qth.net/subscribers.htm
>Elecraft web page: http://www.elecraft.com

_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com

Reply | Threaded
Open this post in threaded view
|

Re: Tube Linear Amp design for basic K2

William Tipton
In reply to this post by Wyn Hughes

   Eimac  introduced the 8166/4-1000A in 1945 as part of its postwar line
   of  glass-envelope, radial  beam   tetrodes  which  also  included the
   4-65A,  4-125A, and 4-250A; and later in 1947, the 4-400A, essentially
   a 4-240A upgraded with a finned anode (plate) radiator.

   External  anode,  forced-air  cooled  tetrodes and pentodes were later
   developments.  Penta  Labs  developed the 8295/PL-172 pentode--a later
   version,  the  8295A  had ceramic instead of glass insulators, but the
   same   1000W  plate  dissipation  rating.  Penta,  Eimac,  and  others
   manufactured this tube. Comparable EIMAC external anode tubes were the
   8168/4CX-1000A  radial  beam tetrode and the 4CX-1500B, both specially
   designed for AB1 linear amplifier service.

   The  Russian  4CX-800  and 4CX-1600 tetrodes marked under Svetlana and
   other  brand  names  are  as  far  as  I know the most recent designs,
   incorporated striped cathode structure to reduce grid interception and
   differential cathode-to-anode radius to reduce secondary emission.



   William L. Tipton, Jr.

   1332 Pinewood Road

   Jacksonville Beach, FL 32250-2941

   904-241-0134


       ______________________________________________________________

     From: "Wyn" <[hidden email]>
     To: <[hidden email]>, "Charles Greene" <[hidden email]>
     Subject: Re: [Elecraft] Tube Linear Amp design for basic K2
     Date: Sun, 11 Jun 2006 15:51:52 +0800
     >Chas,
     >
     >Thanks  for  your  comments.  It's interesting to see that someone
     else
     >has come to some very similar fairly similar basic conclusions as
     >oneself.
     >
     >I have emailed RF Parts for the current price of the 4-1000a. At
     >this stage, it may remain a drwing board exercise..
     >
     >73s,
     >Wyn
     >
     >
     >----- Original Message ----- From: "Charles Greene"
     ><[hidden email]>
     >To: "Wyn" <[hidden email]>; <[hidden email]>
     >Sent: Saturday, June 10, 2006 3:46 AM
     >Subject: Re: [Elecraft] Tube Linear Amp design for basic K2
     >
     >
     >>Wyn,
     >>
     >.... A 4-1000A is the nearest new(er) design tube. I
     >>thought about how I could make the old guy put out something
     >>exciting when driven by my K2. I could swamp the input with a 200
     >>ohm  resistor  and  use  a  4:1 balun and the impedance to the K2
     would
     >>still be 50 ohms. ...... The tube itself does not take any power
     >>in  class  AB1,  but  the input voltage needs to be so much for a
     given
     >>output. A 4:1 balun would increase the drive voltage by a factor
     >>of  4  times  which  would increase the output of the amp to 1600
     watts
     >>if  the  drive  remained  the  same,  800  watts  with  10  watts
     drive......
     >> From 20 watts to 1600 watts is a power gain of 19 dB which in my
     >>opinion  is  a  little  high.  I would be more comfortable with a
     power
     >>gain of 16 dB, but you can try it.
     >>
     >>....... Those tubes are expensive.
     >>
     >>73, Chas W1CG
     >
     >_______________________________________________
     >Elecraft mailing list
     >Post to: [hidden email]
     >You must be a subscriber to post to the list.
     >Subscriber Info (Addr. Change, sub, unsub etc.):
     >http://mailman.qth.net/mailman/listinfo/elecraft
     >
     >Help: http://mailman.qth.net/subscribers.htm
     >Elecraft web page: http://www.elecraft.com
_______________________________________________
Elecraft mailing list
Post to: [hidden email]
You must be a subscriber to post to the list.
Subscriber Info (Addr. Change, sub, unsub etc.):
 http://mailman.qth.net/mailman/listinfo/elecraft   

Help: http://mailman.qth.net/subscribers.htm
Elecraft web page: http://www.elecraft.com