Amps
[Top] [All Lists]

Re: [Amps] Good amp to buy (or something like that)

To: garyschafer@comcast.net
Subject: Re: [Amps] Good amp to buy (or something like that)
From: Rob Atkinson <ranchorobbo@gmail.com>
Date: Sun, 31 Jan 2010 17:23:45 -0600
List-post: <amps@contesting.com">mailto:amps@contesting.com>
Gary,

You are correct but your knowledge is out of date.  What you say does
indeed apply to old methods of analog audio limiting such as diode
clipping (eg CBS Labs Volumax) which should not be used in SSB
service.  Newer methods of processing employing DSP are different.

I should also add that when I wrote about audio peak limiting earlier,
I was not referring to your common garden variety audio limiter of the
type sold in music stores but rather a special category that is
extremely fast and aggressive that does not allow for any delay and
lets nothing past that is over a set level.  These usually employ
multi-band processing to eliminate distortion products (Optimods,
Inovonics MAP, Dorrough DAP for example) and/or pulse width modulation
methods (Inovonics 222) and are normally only found in broadcast and
high quality professional analog products (for example the old analog
Orban compressor/limiters that employed proprietary timing and
look-ahead VCA circuits).

Unfortunately, information in old ARRL handbooks on this subject plus
a certain company making and selling an RF Processor have everyone
believing audio frequency processing and limiting for SSB cannot be
done.  It can be, if the right methods are employed.  This is really
my last post on this as this is far from RF linear amplifiers as a
subject.

73

Rob
K5UJ


On Sun, Jan 31, 2010 at 2:57 PM, Gary Schafer <garyschafer@comcast.net> wrote:
> All you need to do is feed a square wave audio into an SSB transmitter and
> look at the RF output waveform. You will not see a square wave but will see
> sharp spikes that represent the rise and fall of the square wave corners. If
> you do the same with an AM transmitter you will get a nice square wave
> modulation envelope out at RF provided the modulator can pass the
> frequencies involved.
>
> Yes audio peak limiting helps some on SSB but because of all the phase shift
> involved in SSB generation the RF envelope does not closely resemble the
> audio input. Look at what happens as soon as you start to limit audio peaks.
> You are creating square waves from the otherwise sin wave audio.
>
> 73
> Gary K4FMX
>
>> -----Original Message-----
>> From: Rob Atkinson [mailto:ranchorobbo@gmail.com]
>> Sent: Sunday, January 31, 2010 2:17 PM
>> To: garyschafer@comcast.net
>> Cc: amps@contesting.com
>> Subject: Re: [Amps] Good amp to buy (or something like that)
>>
>> I have performed tight peak limiting on audio operating SSB and find
>> it works well and the ALC is not needed.  There is a relationship
>> between input audio amplitude and SSB RF power IF your SSB transmit
>> power is flat i.e. the same level for a given input audio amplitude
>> across your transmit frequency response and your SSB transmitter can
>> pass all frequencies in your baseband audio source, i.e. the
>> transmitter is not being used to limit the transmitted audio range.
>> These are both big ifs for the typical ham SSB station.  Perhaps in
>> the newer all DSP rigs with everything coded in firmware the ALC
>> operates differently from the method I described in an earlier post.
>> I hope this is true provided it works better than the scheme I wrote
>> about.
>>
>> As this is the amps reflector I do not wish to digress farther off
>> topic however I stand by my earlier statement and wished to add a few
>> key points I neglected to provide earlier.
>>
>> 73
>>
>> rob
>> K5UJ
>>
>> On Sun, Jan 31, 2010 at 11:19 AM, Gary Schafer <garyschafer@comcast.net>
>> wrote:
>> > Audio reduction doesn't work either but for a different reason. The
>> audio
>> > waveform only has a slight resemblance of the RF waveform in an SSB
>> > transmitter. Audio limiting works well on AM with both side bands but
>> does
>> > not work well on SSB.
>> >
>> > 73
>> > Gary  K4FMX
>> >
>> >> -----Original Message-----
>> >> From: amps-bounces@contesting.com [mailto:amps-
>> bounces@contesting.com]
>> >> On Behalf Of Rob Atkinson
>> >> Sent: Sunday, January 31, 2010 10:16 AM
>> >> To: amps@contesting.com
>> >> Subject: Re: [Amps] Good amp to buy (or something like that)
>> >>
>> >> I'm afraid I must disagree in some cases.  I know nothing about the
>> >> Icom ALC design but I'll be willing to bet it is typical of modern
>> ham
>> >> SSB transceiver design attempts at a cheap way of preventing power
>> >> output excursions above some front panel set limit.
>> >>
>> >> With modern SSB transmitters what is driving all the stages up to and
>> >> including the PA is the AF current from the audio source coming into
>> >> the rig.  Designers usually put some kind of high vswr protection on
>> >> the PA output so the transistors are not damaged by high vswr.  This
>> >> is often done with a reflected power coupler on the PA output that
>> >> produces some control voltage that is fed back to a low level voltage
>> >> controlled amplifier at some IF stage, 455 KHz perhaps, that performs
>> >> a gain reduction to fold back drive to the PA and reduce output to
>> >> protect the final transistors.  This design works pretty well.  It
>> >> doesn't cost much to add a forward power coupler engineered to
>> produce
>> >> a certain voltage level such that with a certain setting of the power
>> >> out pot (PA collector current pot) that voltage fed back to the VCA
>> at
>> >> a previous IF stage, performs the needed gain reduction to hold the
>> PA
>> >> to some set forward power limit.
>> >>
>> >> Obviously, while this is great for VSWR protection, it is a really
>> bad
>> >> way to limit forward power output peaks and spikes for the simple
>> >> reason that once that horse is out the barnyard gate, it's too late
>> to
>> >> close it.   Most ALC circuits employ some delay time constants so
>> once
>> >> a power spike is sensed and a gain reduction voltage fed back to the
>> >> VCA it holds for a few seconds and continues to be effective if there
>> >> is high average audio driving the RF but don't count on placing your
>> >> SSB rig in transmit, turning your PA power pot to 100 w. and expect
>> >> that not very well thought out design to trap every initial breath
>> >> blast or sudden impulsive sound force at 100 w.   Once that spike is
>> >> out, it's on the way to the amp and everyone in radio land and your
>> >> ALC can say "oopsie" and send a gain reduction v. back to the VCA and
>> >> deal with any other pops and blasts for a few seconds but that first
>> >> one has left the station.  And far too many get out the gate in
>> >> typical operation to make ALC something to depend on for power output
>> >> limiting.
>> >>
>> >> So, since ALC is a bad scheme to rely on for peak power limiting.
>> >> What to do?  A far better solution is to employ a device that
>> >> completely controls audio level peaks being fed into the ham rig as
>> is
>> >> done in broadcasting, so ALC is not even necessary and does not have
>> >> to intervene.   Why isn't this done?  Because a zero attack time
>> >> extremely fast gain reduction limiter on the audio input (i.e. mic
>> >> jack) isn't cheap.  And hams who do not understand how ALC works and
>> >> realize that it is inferior as far as what it is supposed to do is
>> >> concerned, won't want to pay for something that gets the job done.
>> >> So, we're left with a sort of half assed method and too many hams
>> >> overdriving their amps (those cheap s.s. ones are the worst)
>> >> unfortunately.
>> >>
>> >> 73
>> >>
>> >> Rob
>> >> K5UJ
>> >>
>> >>
>> >> <<<<There are two kinds of ALC and I think Roger is talking about the
>> >> kind
>> >> of ALC that is fed back from the amplifier to the transceiver. With
>> >> that kind, I fully agree with the above.
>> >>
>> >> But there is also ALC that is generated entirely within the
>> >> transceiver (such as my IC-756Pro3), and is completely independent of
>> >> the amplifier. With that kind you can safely use ALC to limit the
>> >> drive and it does an excellent job, I might add.
>> >>
>> >> And if you use the internal type of ALC, you don't need the fed-back
>> >> kind at all.
>> >>
>> >> 73, Bill W6WRT>>>>
>> >> _______________________________________________
>> >> Amps mailing list
>> >> Amps@contesting.com
>> >> http://lists.contesting.com/mailman/listinfo/amps
>> >
>> >
>
>
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps

<Prev in Thread] Current Thread [Next in Thread>