[Date Prev][Date Next][Thread Prev][Thread Next] - [Date Index][Thread Index][Author Index]

Re: Geosynchronous Satellites



on 12/25/00 10:57 AM, Jose Mihotel at jmihotek@gte.net wrote:

OK, Now with a little more time, let's do indeed get some things straight.
This post is a little long, but bear with me, I think we reach common ground
here.......


> Lets get a few things straight
> 
> When you say FM, lets agree you mean class C.

No I do not.  FM is a modulation form.  Class C is a type of amplifier.
Totally different.  Please, do not confuse things.

> When you say cellular you mean class A

No, I do not.  Analog Cellular can use either a class A amplifier or a class
C amplifier.  And digital cellular, while generally using Class A can use a
less linear amplifier for modulation forms such as GSM (because it is pretty
close to being a constant envelope signal.).

Also, no cellular amplifier uses totally Class A PAs.  Most are a
combination of A and AB.

> When you say SSB or AM or FM  you mean same bandwidth.

No I do not.  In ham radio FM channels have a 25 KHz bandwidth.  AM is about
6 KHz.  SSB is about 2 KHz.  I do not mean the same bandwidth.


> 
> Presently I design class A PAs for wideband  MMDS wireless-cable  around the
> world. I know how innefficient class A PAs can be. ( Cellular uses class A
> power amplifiers) I'm not arguing that.
> But please, when you say FM lets agree we are talking about FM repeaters,
> and the like . (class C)

My point in my original posting that you picked on and claimed that I was
incorrect about stated the following:

"Also, Phil, there are some very valid forms of digital modulation that do
use AM.  They work quite well, IN THE PROPER application.  FM is used in
subscriber units, etc. because of the efficiencies involved in the
modulation (and for some of the reasons you state in your post).  But the
same cannot be said for that base stations that use amplifiers with
efficiencies of 7% (and do NOT argue with me on that one as cellular
amplifiers is my background)."

My comments here were a discussion of an AM modulated signal vs. an FM
modulated signal.  Phil Karn feels that FM signals, particularly digital
ones, perform better than AM signals.  My statement was that an AM modulated
digital signal could be used.  However, digital FM modulation is used in
subscriber handsets for cellular because you don't need a PA as linear as
you would for a digital AM signal.  For analog FM, you can use a handset PA
with typical Class C efficiencies.  For digital signals, you need more
linear amplifiers that are typically in the 30% to 40% efficient range.
Now, perhaps my statement was not quite clear enough as I am not sure how
linear a PA you would need to support an AM digital signal.  In handsets
efficiency is the name of the game since they run on batteries.
Manufacturers are doing all sorts of interesting PA topologies including
such things as Dougherty Amplifiers to improve efficiencies.

My statement about the basestations was that they do NOT typically use
efficient amplifiers these days even for FM modulated signals.  I stated a
fact, that efficiencies are typically about 7% to 8%.  You and Phil both
jumped on me and ridiculed me saying effectively, "You're nuts! Are you
saying that an amplifier with greater than 7% efficiency can't be made?"  In
fact, I quote you:

"your statements about PAs that have 7% efficiency is plain not true"

I would have quoted the rest to get some context, but I deleted your message
and only have what I responded to in a previous post.

But you said my statement was just not true.  But it is true and it is true
particularly for the context that I was talking about.

> 
> Now you  tell me , how efficient is the PA in the local club FM repeater?
> Most likely a little better than  ~50%
> That is Power in = power out.

A PA in an FM repeater is a completely different animal.  If the designer or
manufacturer has any sense, it's a class C amplifier with probably at least
70% efficiency.

> 
> The point I'm making is that FM PAs are much better than 7% efficient.

No.  Class C PAs are better than 7% efficient.  There is no such thing as an
"FM" PA.  FM is a form of modulation.  I don't care about the terms people
use.  If they call a Class C PA, an FM PA, then technically, that is
incorrect!  A class A, AB, or B amplifier will amplify an FM signal just as
well as a Class C PA except it will not be as efficient and you'll waste
unnecessary power.  And in fact, a class C amplifier can be used to amplify
any constant envelope signal including CW.

If we are going to talk engineering and design here, let us talk in the
technically correct terms.  If a linearized amplifier with about 7%
efficiency is not capable of amplifying an FM signal then how did Lucent
ever sell thousands of their Linear Amplifier Cabinets (LACs) over the last
12 years or so?  In fact, the funny thing about that linearized amplifier is
that it was really designed with only analog FM in mind (digital was not
really around when the LAC was first designed).  For CDMA signals it doesn't
have the ability to properly handle the high peak to average ratio of a CDMA
carrier.  The PA has to be derated by up to 3 dB depending on which model it
is.  And adding multiple CDMA carriers is even worse.  Again, it was
designed to handle analog FM signals and it is not a class C amplifier.

> 
> Another thing to set straight.
> Back in the days of  Mayor Armstrong, the inventor of FM radio, he proved
> mathematically that for narrow band FM,  if you make the occupied bandwidth
> the same, FM will talk practically as far as AM. The noise power is the same
> for both cases.

OK.  Now we get to my original point and you proved me correct.  Thank you.
My point was that in a standard ham radio set up that an FM channel has a
bandwidth of approximately 25 KHz.  AM or SSB signals use narrow bandwidth
(typically).  So the power is effectively concentrated over a narrow
bandwidth.  In a standard FM signal it is spread out farther.  You've got
say 50 Watts spread out over 2 KHz or 50 Watts over 25 KHz.  This is why you
typically have farther propagation with an AM or SSB signal.

But you are correct, if you narrow the bandwidth of an FM signal you will
get the same noise power and pretty much the same propagation distance.  But
that is IF you narrow the bandwidth.  However, as you narrow the occupied
bandwidth of your FM signal you begin to lose audio fidelity and I would bet
by the time you take it down to 6 KHz or less that an AM signal would give
better fidelity.  That's an unqualified guess, I'll admit, but I know that
most people who are looking at making land mobile FM radios with a 6.25 KHz
bandwidth are looking at making that a digital signal as analog FM would
just not sound as good with that narrow a deviation.

I would agree with you on this point totally.  You backed up my original
contention.


> 
> So there, you can make your FM occupied bandwith 2 KHz if you like.

Well, that's true.  But how would it sound?


> 
> The main point here I think is that for LEO sats. at VHF or UHF either
> system will talk just as far,  but for  geosycchronous satellites, AM or SSB
> systems have an edge.

Oh I agree too.  In fact, that is one of the things I was trying to say.
Phil and I were arguing about that and he disagreed with me on it.  Perhaps
you misunderstood my post.  And the original, original question of why I
even started this post was about a guy asking about doing an FM AO27 type
geosynchronous bird.

> 
> As for the reason EME and other modes don't use FM is because by nature FM
> detectors (discriminators, ratio detectors, phase locked loops,etc...) need
> better than 8dB carrier to noise ratios to provide the signal to noise
> improvement that you get from FM de-emphasis. With C/Ns lower than that the
> FM detector falls appart very quickly. That is to say for 10 dB C / N you
> get 10 dB S / N but for 5 dB Carrier / Noise you  only get say 1 dB Signal
> / Noise. So in situations where your signal might be burried in the noise,
> say 1 dB carrier to noise, an FM detector does not work.
> 
> AM / SSB detectors, (diode, synchronousdetector, product detector) on the
> other hand, don't have that problem. They hold their own all the way down
> into the mud.

Thanks for answering that question clearly and stating it in a way that I
was having difficulty finding words for (that's one problem not being
involved in engineering on a day to day basis any more!).

> 
> So the moral of the story is to be fair and compare Apples and Apples.

Why not compare an Apple and a Windows based product?  :-)

Seriously, I think we agree far more than we disagree!  And probably Phil
and I would as well........

> 
> Merry Chritmass !!!

Merry Christmas to you too.

73,

Jon
NA9D

-------------------------------------
Jon Ogden
NA9D (ex: KE9NA)

Member:  ARRL, AMSAT, DXCC, NRA

http://www.qsl.net/ke9na

"A life lived in fear is a life half lived."

----
Via the amsat-bb mailing list at AMSAT.ORG courtesy of AMSAT-NA.
To unsubscribe, send "unsubscribe amsat-bb" to Majordomo@amsat.org



AMSAT Top AMSAT Home