Jim,
I don't want to engage in a war about this matter, so I won't reply to
all your statements, even when I feel different than you about them. But
a few things you wrote, like direct questions, need to be answered.
> ### Aside from all the typ problems of SS amps, there is even more
> problems. None of em will run full bore rtty, etc into a 3:1 swr !
That's simply because solid state amps usually don't have adjustable
impedance matching. It's the convenience of using a no-tune amplifier,
that requires low SWR antennas, and not the use of transistors per se.
It's perfectly possible to make a solid state amp that needs to be tuned
up on each band, like a tube amp, and then it would also be able to
operate over the same SWR range as the tube amp does. That no company
makes them, is probably because nobody would buy them. It's a fact that
most people prefer no-tune equipment, and are willing to set up good,
low SWR antennas for that. It's indeed my case: I don't enjoy tuning up
an amp, even if it takes just 10 seconds. And I have an antenna system
that has an SWR below 1:1.5 over most of 80 meters, almost all of 40
meters, and all of 20, 17, 15, 12 and 10 meters. So I have no need at
all to run an amp into 3:1 SWR. Like me, there must be many hams.
Fix your antennas, Jim! ;-)
> So now you require the mating automatic 2 kw CCS rated tuner.
Yes, for antennas with the occasional high SWR an autotuner is good. But
why 2kW CCS??? I'm a ham, not a broadcaster. The legal limit here in
Chile is 1200 watts, in the US it's 1500 watts if I'm right. And hams
normally do listen a good part of the time. At least good hams do! So,
what you need is 1.5kW ICAS, not 2kW CCS, unless you are fully
intending to infringe both FCC rules and the good ham practice of listening!
> ## With these SS amps, you also require a monoband 2kw CCS
> rated LP filter for each band !
Again, 1.5kW ICAS is all you need.
> At least a min of 5-6 required to cover 160-10m + warc.
Depends on how clean or dirty the amp is. But yes, usually it's a good
idea to have 5-6.
> Plus you have to be able to switch them in and out.
Normally these filters are assembled on a printed circuit board, along
with relays to switch them. The relays can be controlled from a manual
bandswitch, or an interface can be made to have the radio switch them
when you change bands. That's simple. It's also possible to make a
circuit that senses the frequency as soon as you transmit, and selects
the proper low band filter automatically. That's a little bit more
involved, but not terribly so.
> ### ham radio is in its dying years.
It's painful to admit it, but I tend to agree with that. At least, ham
radio as I understand it, as a hobby practised by people who like to
build radios, improve radios, and experiment with those radios they have
built, seems to be almost completely gone. The heartening news are those
from a very few hams spread out over the world, who are still developing
technology, building equipment from scratch to their own designs, and
using it on the air. But those are few, and getting fewer.
> Empty QSO?s,
> that?s at least one step better than 10,000 brain dead contesters
> on every 2nd weekend.
Probably you are right. I sometimes visit the XR6T contest station, in
mid action, and then I get a strange feeling. On one side, it's good to
see all those hams obviously having a lot of fun. On the other hand, for
the life of me I cannot understand where's the fun in running several
thousand QSO's within 48 hours, powered by coffee and adrenaline, each
lasting perhaps 5 to 10 seconds, each (and I mean each!) with a 5-9
given and a 5-9 received, even when the other station was almost
inaudible, and none of these QSOs having been about anything more than
the callsign, the 5-9, and maybe some serial number.
Not for me. But it's one side among many in our hobby, so I accept it.
We are all a little bit crazy, aren't we?
> You have a better chance of getting mashed in your car by a soccer mom,
> trying to do some texting while driving vs HV accidents.
Right. Even so, a lot of people are terrible scared of HV, and keep
driving in a reckless way. Humans just aren't logical all the time. If ever.
> #### How am I supposed to get 750 watts AM CXR, and 3750 watts
> pep out.... and measured at the feed-point of the antenna..with
> SS.
That's actually a good application for those pulse-rated transistors,
which can electrically deliver far higher power than their dissipation
rating allows to produce continuously. But the best way to get that
power, of course, is with a high efficiency, class E or class F
amplifier, modulated by a switching audio amplifier (class D).
> Almost forgot.... that?s into a high swr. Repeat for all bands
> 160-10m, including warc.
Use a tuner, of course. Could be a beefy homebuilt autotuner, or a plain
old fashioned manual T-type tuner, built with a hefty variable inductor
and two nice big wide spaced variable capacitors. My homebuilt tuner
handles RTTY at 1200W into 12:1 SWR. It gets warm, but not hot. At my
former QTH (an appartment) I was antenna-challenged, and on 160m I had
to tune up the 40m loop. That gave a lousy low radiation resistance
along with very high reactance. Now instead I have room for a 160m full
size rhombic! Haven't put it up though, and probably won't ever do it,
because 160m here is very narrow, just 1800 to 1850kHz, and there is a
huge second harmonic of a BC transmitter on 1840, booming in at 9+40dB
and splattering over everything from 1810 to 1870.
> ### What is the TX IMD on these SS amps you have built.
> That Elecraft K3 xcvr is crap at -29db IMD-3. I don?t think
> their KPA-500 is much better.
I have some trouble measuring the IMD, precisely because the radio I
have to drive it, a Kenwood TS-450, also has such por IMD. So, when I
drive the amp into saturation, the IMD shoots up and I can see it, but
otherwise I see the same IMD that the radio has alone, without amp, and
so I don't really know how much IMD the amp is contributing. The typical
values I see are between around -35dB to -28dB on 160 through 17 meters,
and then it gets worse, down to a horrible -20dB on ten meters! I
repeat, that's the output of the TS-450, and the output of the amp looks
just the same, so it isn't contributing enough IMD to make it visible.
The TS-450 contains the same design mistake in the output stage that I
mentioned in the post before this. That's probably a good part of the
reason for that poor IMD.
My amp is powered from a rather high voltage, and has strong source
degeneration. This should make it pretty clean, but I really would need
a cleaner transceiver to be able to measure the actual IMD of the amp.
For now I only can say that it won't degrade the spectral purity of my
Kenwood radio.
> ### Using 3-4 x drake L4B amps, all wired nose to tail in series,
> (since they will handle 2 kw on bypass mode), I just tune each
> one up on a different band ! One rotary switch ensures only
> one key line is active at any one time. Simple.
Well, that's certainly a truly cost-effective and compact setup!
I hope you are joking...?
> ## it takes exactly 45 secs to qsy from 15m to 160m. Just dial up
> using pre-sets.
And with a solid state, autoswitched amp it takes a small fraction of
one second to change bands. CLICK. Ready.
> ## My other problem with SS is eff. Your typ SS amp operates at a
> lousy 50% max eff..and that?s when run at max output !
When using the wrong version of the Granberg designs, yes, indeed, even
45% can often be seen! But use a correct design, and you suddenly will
see 62 or 65% in class AB.
For best efficiency, use tuned matching, just like with tubes. That will
get you pretty close to the theoretical limit.
> Reduce
> the PO by half....and eff will drop down to 35%. At that point,
> you have to reduce the vdc a bit to get the eff back up. The
> problem with reduced vdc is.... the TX imd drops off !
That's a problem with all linear amps, tube or SS.
Speaking about efficiency, what's your average TX/RX duty cycle? A very
serious ragchewer might get close to 50%, but most hams are way lower.
I'm below 10%. That means, 9 times as much time spent listens, as
transmitting. And this changes the efficiency equation quite
drastically! How much power does your tube type amp draw while standing
by? Maybe a hundred watts or so? You have filaments to keep hot, fans to
keep running, power supply capacitors to bleed off... My SS amp instead
will use about 3 watts while in standby. Most of that will be for the
LEDs on the front panel. The water pump will be shut off a few seconds
after going to RX. At present, consumption duirng RX is zero, of course,
since there isn't even a front panel yet... far from it!
And yes, the switching power supply too gets shut down during RX, and
restarts in a matter of 10-20 milliseconds after pressing the PTT.
In this light, even an 45% efficient SS amp is more efficient over the
whole TX/RX cycle, than a tube amp with 65% plate efficiency is. But 45%
efficiency is lousy, and a decently designed SS amp is a lot better.
> ## a 1.5 to 2kw output CCS SS amp, that covers all 9 x bands, and will
> handle
> a 3:1 swr and LOW IMD, is no simple task. Hats off to anybody who can
> pull it off at any price.
I'm working on mine, but I'm aiming only at 1.2kW, the legal limit here.
Results so far are promising. ICAS or CCS with my amp depends mainly on
the cooling of the passive components. It's planned to be fan-less. By
adding a small fan, it could be changed from ICAS into CCS, and of
course a radiator would have to be used instead of my water bucket. I'm
not intending to handle 3:1 SWR inside the amp, because I don't need
that. It would require some sort of tuning network, either manual or
automatic. The 9 bands will be covered, but 30m won't get its own low
pass filter, because that band isn't open here. I had some struggles
getting my output transformer to work over the full range, but now it
does. Power output on 10 meters will be "best effort". At this time it's
around 900W only, in linear service. From 15 meters down it's 1200W.
That's a result of using switching MOSFETs costing under 75 dollars for
the whole set. For that price, I find it good enough. Six meters will
NOT be covered in my amp. The FETs I'm using would work on six meters
only with a tuned matching network.
Phew, I ended up commenting most points anyway...! ;-)
And if I keep doing so, instead of going back to the workbench, my amp
will never get ready!
Manfred
========================
Visit my hobby homepage!
http://ludens.cl
========================
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|