Elecraft

K3: 630m amplifiers

Classic

List

Threaded

30 messages Options

Alan Bloom

Re: K3: 630m amplifiers

In the "good old" tube days, typically the transmitter's final amplifier
was the keyed stage, or at least one of them. So the key shaping was
not degraded by a subsequent amplifier. It's OK if the PA is a class C
amplifier as long as the key shaping circuit takes that into account.

One way to get away with using a separate class C amplifier it to use a
two-stage grid bias circuit. A fixed voltage biases the tube so that it
draws a little resting current (like a linear class B amplifier) and the
rest of the bias is developed from a grid-leak resistor, which only
kicks in when RF is applied to the input. So as the RF input ramps up
at the beginning of a dit, there is no abrupt transition from off to on.
It still increases the slope of the rise and fall, so it does increase
key clicks, but not as bad as with pure fixed bias. You can compensate
by slowing down the rise/fall times of the key shaping circuit.

Alan N1AL

On 07/27/2015 04:57 PM, Tom Azlin W7SUA wrote:
> Interesting discussion, but I'm now confused.
>
> In all the old handbooks the discussion of amplifier design says ( page
> 76, 1941) "In amateur transmitters, and r.f. amplifier is invariably
> operated Class C ( see Chaptr 3)."
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Jim Brown-10

Re: K3: 630m amplifiers

In reply to this post by Tom Azlin W7SUA-2

On Mon,7/27/2015 4:57 PM, Tom Azlin W7SUA wrote:
> In all the old handbooks the discussion of amplifier design says (
> page 76, 1941) "In amateur transmitters, and r.f. amplifier is
> invariably operated Class C ( see Chaptr 3)." I though the final tuned
> circuit was simply "pulsed" at the RF freq and then the output
> filtering knocked off harmonics leaving a clean CW note.

Yes and no. Yes, the tank of a class C amp is pulsed and the output
network kills the harmonics. But it does NOT kill the intermodulation
distortion that is produced by the non-linearity, and that intermod is
heard as clicks.

BTW -- you need a MUCH newer Handbook. :)

73, Jim K9YC
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Phil Wheeler-2

Re: K3: 630m amplifiers

In reply to this post by Alan Bloom

Clearly you are not among those who reminisce
about the rotary spark gap, Alan :-)

73, Phil W7OX

On 7/27/15 1:53 PM, Alan wrote:

> The need for linear amplifiers with CW has been
> known for many years. Quoting from the 1964 ARRL
> Handbook:
>
> When key clicks introduced by the addition of
> an amplifier stage
> are found only near the transmitter
> frequency, amplifier "clipping"
> is indicated. It is quite common when fixed
> bias is used on the
> amplifier and the bias is well past the
> "cut-off" value.
>
> [In other words, a class C amplifier. Similar
> to a bipolar transistor amplifier with no base
> bias.]
>
> A linear amplifier (Class AB1, AB2 or B) will
> amplify the
> excitation without adding any clicks, and if
> clicks show up, a
> low-frequency parasitic oscillation is
> probably the reason.
>
> Alan N1AL

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Alan Bloom

Re: K3: 630m amplifiers

On 07/27/2015 05:32 PM, Phil Wheeler wrote:
> Clearly you are not among those who reminisce about the rotary spark
> gap, Alan :-)

By crackey them vacuum tubes are just too durn delicate for serious ham
radio work. Old reliable spark is much more rugged.

Quoting from "The World of Ham Radio, 1901-1950, A Social History" by
Richard A. Bartlett talking about the first issue of QST after World War I:

In the two years that had elapsed since amateur radio was silenced,
impressive progress had been made. Most significant was the
development of the vacuum tube. At first known as audions, bulbs
or lamps, the originals lacked uniformity, ... needed careful
adjustments of the filaments and plate voltage, and were simply
too delicate for "the Service."

You could abuse a spark transmitter, but the tubes would die if you
weren't gentle with them.

Reminds me of comments I heard hams make about transistors during the
tube-to-transistor transition 50 years ago.

Alan N1AL

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Don Wilhelm-4

Re: K3: 630m amplifiers

Those first point contact transistors such as the CK722 were rather
'finicky'. Not much margin for error, and they were expensive.
Now we can buy 2N2222As for pennies.

73,
Don W3FPR

On 7/27/2015 8:49 PM, Alan wrote:
>
> You could abuse a spark transmitter, but the tubes would die if you
> weren't gentle with them.
>
> Reminds me of comments I heard hams make about transistors during the
> tube-to-transistor transition 50 years ago.
>
> Alan N1AL
>

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Barry K3NDM

Re: K3: 630m amplifiers

Don,
You are making feel old. I remember the CK722. I had one as an
audio amp behind a crystal set I built using a 1N34, not sure it was
even an A version. I had a 210 foot piece of wire strung from the radio,
through my bedroom window to a post in the back of our yard. I could
hear, almost all, the AM stations in the Washington, D.C. area, all at
once. And, that was in the 1950's. The really nice thing about the set
up is that a D cell battery could power the thing to drive high
impedance headphones nicely. Coupling to the CK722 was directly to the
base with no bias. It worked. It may not have been the best design, but
it worked. For a young kid that was all that mattered.

73,
Barry
K3NDM

On 7/27/2015 9:00 PM, Don Wilhelm wrote:

> Those first point contact transistors such as the CK722 were rather
> 'finicky'. Not much margin for error, and they were expensive.
> Now we can buy 2N2222As for pennies.
>
> 73,
> Don W3FPR
>
> On 7/27/2015 8:49 PM, Alan wrote:
>>
>> You could abuse a spark transmitter, but the tubes would die if you
>> weren't gentle with them.
>>
>> Reminds me of comments I heard hams make about transistors during the
>> tube-to-transistor transition 50 years ago.
>>
>> Alan N1AL
>>
>
> ______________________________________________________________
> Elecraft mailing list
> Home: http://mailman.qth.net/mailman/listinfo/elecraft
> Help: http://mailman.qth.net/mmfaq.htm
> Post: mailto:[hidden email]
>
> This list hosted by: http://www.qsl.net
> Please help support this email list: http://www.qsl.net/donate.html
> Message delivered to [hidden email]
>

Tom Azlin W7SUA

Re: K3: 630m amplifiers

In reply to this post by Edward R Cole

OK on all the thread.

Ed, Yes, I am interested in following your linear amplifier project.

73, tom w7sua

On 7/27/2015 10:43 AM, Edward R Cole wrote:

> To catch up on the comments I address all that I have so far:
>
> Re: linear vs class-C is really a moot point as its easy enough to build
> a linear at 500-KHz. CW, JT65, FSK all do fine with class-C, of
> course. And its true that any modulation of a continuous carrier will
> impart some bandwidth. Its not likely that wide-band digital modes will
> be approved in a band that is only 7-KHz wide so probably those approved
> would work with a class-C amp. I think its just simpler to provide
> class-AB then its a non-issue.
>
> Re: 15-dB HF linear amp gain limit. If that is applied to 630m amps
> then either I sell a kit or partial kit.
> 1mw in to 1w out is 30-dB and 100w out is 50-dB. I would have to
> research the regs to see if the limit is for a certain level of input
> (like 5w). Also the regs allow hams to sell amps to other hams without
> this restriction (but at what quantity per year?).
>
> First off I am not proposing to go into competition with MFJ. This is
> just a home business building one at a time. If the FCC is going to
> make this a small (tiny) business killer than its just not worth the
> bother. If I say more than this it will get political and that is OT.
>
> 73, Ed - KL7UW

Tom Azlin W7SUA

Re: K3: 630m amplifiers

In reply to this post by Jim Brown-10

Hi Jim,

I do have much more modern handbooks. Current in fact. :) Just
remembered more Class C referenced in the older ones.

Will have to go read articles back then on key clicks and distortion.
And continue to look at what amplifiers I could use on the two soon to
be lowest bands.

73, tom w7sua

On 7/27/2015 5:13 PM, Jim Brown wrote:

> On Mon,7/27/2015 4:57 PM, Tom Azlin W7SUA wrote:
>> In all the old handbooks the discussion of amplifier design says (
>> page 76, 1941) "In amateur transmitters, and r.f. amplifier is
>> invariably operated Class C ( see Chaptr 3)." I though the final tuned
>> circuit was simply "pulsed" at the RF freq and then the output
>> filtering knocked off harmonics leaving a clean CW note.
>
> Yes and no. Yes, the tank of a class C amp is pulsed and the output
> network kills the harmonics. But it does NOT kill the intermodulation
> distortion that is produced by the non-linearity, and that intermod is
> heard as clicks.
>
> BTW -- you need a MUCH newer Handbook. :)
>
> 73, Jim K9YC

Mike Morrow-3

Re: K3: 630m amplifiers

In reply to this post by Edward R Cole

Don wrote:

> Those first point contact transistors such as the CK722 were rather
>'finicky'.

Point-contact transistors could be variable in terms of electrical characteristics, but mechanically they were a lot better than mythology presents. Regardless, the famous 1953 Raytheon CK722 was *not* a point-contact transistor. It was a Germanium junction transistor. It is very unlikely that anyone here ever worked with point-contact transistors...they disappeared quickly after the junction transistor appeared in 1951.

Alan wrote:

>> Reminds me of comments I heard hams make about transistors during the
>> tube-to-transistor transition 50 years ago.

They were late talking about it then. I don't recall that argument having much strength left by 1965. The Raytheon SBE-33 HF ham transceiver came out 52 years ago, followed by the SBE-34 two years later. They were all transistor except for driver and finals and were very successful ham rigs. And the most successful military tactical VHF-FM sets of all time...the AN/VRC-12 series and AN/PRC-25, all transistor except for finals, were fully developed by 1961...and remained in some US service until 2008!!!

Mike / KK5F

______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]

Edward R Cole

Re: K3: 630m amplifiers

In reply to this post by Edward R Cole

Jim,

I was thinking of back in the 1950's when class-C was used with
amplifiers for CW. I guess the finals were direct-keyed and not
driven by a keyed stage. Its been awhile. My DX-35 used grid-block
keying and ran screen-grid AM on the 6146. But then you youngsters
probably don't go that far back so would not know this. Just yanking
your chain a little, Jim, so don't get all riled up by that statement
(you and I both were licensed in the 50's; ex-K8MWA).

I was thinking that my classic W6PO-8877 was running class-C but in
fact is running linear (probably class-AB). Class-C was used in the
past for higher efficiency when gain was expensive. Now even on
microwaves gain is cheap.

But you missed the point that I am not planning to build a class-C
amp for 630m: "I think its just
simpler to provide class-AB then its a non-issue"; a direct quote of
what I said.

If you look at my original post on this topic: "I am wondering what
kind of interest there would be for a linear amp that could be driven
by the K3s/K3?" No where did I suggest anyone use a class-C
amp. Other's brought up using class-C, D, E.

On 630m spectrum is dear so clean signals are more important than
other bands. Likewise, frequency stability is also required to be a
good citizen on the band. The ARRL Group specifies 1-Hz for both
accuracy and stability. The K3 with TCXO-3 (1ppm) can do that.

I don't have a P3. Guess I would have to use my old frumpy HP141T.

73, Ed - KL7UW
Jim, nice pics on your QRZ.com

----------------------------------
On Mon,7/27/2015 10:43 AM, Edward R Cole wrote:
> CW, JT65, FSK all do fine with class-C, of course.

Ed,

I suggest that you drive a Class C amp with a K3 transmitting CW and
look at the spectra with a P3 both at the output of the rig and at the
output of the amp. Set the P3 display for 5 kHz.

73, Jim K9YC

73, Ed - KL7UW
http://www.kl7uw.com
"Kits made by KL7UW"
Dubus Mag business:
[hidden email]
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[hidden email]

This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [hidden email]