Jim, and all,
SPE is delusional if they think they are building state of the art amps.
While I tend to smile about a manufacturer that brags about having 35,000 lines
of code in the controller of his amplifier, which if true indicates inefficient
code rather than advanced technology, I still think that the SPE amps have
several things in their favor. I have no practical hands-on experience with SPE
amps, but from looking them up on different web sites, I notice several things.
One is that the buyers and users of these hams seem to be quite happy with them.
And another is that SPE offers several different models, which definitely are
not all alike.
There is the SPE 1.3K-FA, using a single large gemini MOSFET, run at a power
level that requires either a clever variable bias scheme, or lots of courage.
All the IMD information I have seen about it seems to indicate that it's barely
adequate at 1kW, and probably not adequate at its rated 1300W output. For a ham
looking for a lightweight, automatic, 1kW PEP amplifier for SSB use, this should
be a pretty good choice. It's certainly not an acceptable choice for a ham
wanting to transmit RTTY bulletins at 1500W output.
But then there is the SPE 2K-FA. It uses a combination of six modules, each of
which can produce 300W output with some comfort. If this amplifier is run at
1500W, it should be quite clean. There are spectrum graphs published showing the
performance of this amplifier, and at 1.6kW on 15 meters the 3rd IMD products
are 36dB below each tone, 42dB below PEP. That's much cleaner than most radios,
and thus it's 100% satisfactory for nearly any ham, excepting only those who
pursue ultra clean signals, using SDRs optimized for signal purity.
The power dissipation rating of the FETs used in the 2K-FA is also generous
enough that brick-on-the-key operation at the legal limit should not be a
problem, assuming that the heat sinking has been correctly done for that task.
My main complaint about that amplifier is its price - but I have the same
complaint about many other amps! For any ham willing to spend a lot of money on
an amp, I can't see any reason right now why the 2K-FA should be a bad choice.
And hams who are on a somewhat tighter budget, and want a lightweight amp,
should do well with the 1.3K-FA, as long as they don't abuse it.
No schematics published - yes, that might be a reason not to buy one, but more
out of principle than out of practical need. These amps aren't rocket science.
They are simple things, following very well known basic standard circuits, so if
repair becomes necessary, I don't think the lack of a published schematic will
be much of an obstacle.
Jim, I tend to agree that these amps aren't state of the art. Certainly not the
2K-FA, using six modules in class AB with a splitter and a combiner. The 1.3K-FA
at least uses a more modern approach in that a single package, relatively
recent, high power MOSFET is used. But to be really state of the art, an
amplifier would need to use a high efficiency design, like broadcasters,
military, satellites, etc have been using for quite some time, and at the same
time produce a very clean signal. In this regard, it seems that there is no true
state-of-the-art ham amplifier on the market right now, and so the SPE amps
certainly aren't any less state-of-the-art than the other ones you can buy.
Let's be fair! Let's consider the actual, real, performance and value of each
amp. I suspect that many list members are picking on the SPE amps simply because
they aren't made in the USA, and that's very unfair.
While being fair, of course we can still have an innocent little laugh. 35,000
lines of code equals roughly 600 printed pages. No human being could possibly
keep full track of how such a large program works! I suspect that SPE means the
compiled machine language code of their high-level language program, including
all libraries included with the compiler, whether they are actually used or not.
The firmware of a 15 band receiver I built some time ago, which controls a
frequency synthesizer, display, filter banks, implements a digital S-meter, has
100 frequency memories, automatic band switching, step optimizing, and
everything else such a receiver needs, has a total of roughly 800 lines of code,
if I include all the comment lines, empty lines added for readability, etc. But
the full machine language listing produced by the compiler has a whopping 52,800
lines. Hooray, I'm "higher tech" than SPE! :-)
Manfred
========================
Visit my hobby homepage!
http://ludens.cl
========================
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|