Amps
[Top] [All Lists]

[AMPS] Operating comments using the Emtron DX-2a

To: <amps@contesting.com>
Subject: [AMPS] Operating comments using the Emtron DX-2a
From: K1WPO@aol.com (K1WPO@aol.com)
Date: Sun, 31 Jan 1999 10:57:58 EST
General Technical Comments of Emtron DX-2a
(http://www.emtron.com.au/AmpDX2D.html)

I received one of the first DX-2a shipped to the U.S. a few months ago, and
have been pleased with the performance,
value and performance. Several friends have asked about
the amplifier, so without a lot of technical measurements,
etc, the following are my general, non-technical comments.

Tube
This amplifier uses a Svetlana 4CX1600B
(http://www.svetlana.com/docs/tubeframe.html) running in class AB1.I prefer
the use of a single, 1600 watt plate dissipation tube over two (or even three)
tubes in parallel. This tube is the tetrode equivalent of an 8877, which has
always been an expensive tube. The lower cost allows a better 2KW amplifier at
significantly lower cost.

P/S
The power supply is robust, with a very large and well
constructed transformer. The unit I received was an export (50Hz) model. Later
versions should provide some weight savings when a 60Hz transformer is
available. The rectifier diodes are typical series string as a F/W bridge, and
use the large 6 ampere variety.
B(-) fault diodes are used to GND as well as B(+) glitch
resistors, all located on a very compact HV P/S module.
The entire P/S is removable with two screws and six push on terminals, no
soldering - a very neat, compact design.
Operation is silent, with no transformer hum or lamination
noise. Idle voltage (bleeder load only) is around 3,200Vdc
(at 245Vac in) and full load at 1.0 ampere is about 2,900 Vdc.

AC Input
The six foot cord set is a little light by American
standards, and has a vinyl sleeve. I looks like the
equivalent of 15-16 AWG. Australia rates it at 25 amperes,
which is adequate, but I prefer a real, American cordset.
I replaced mine with a longer (8ft), more flexible, rubber> 14-3 SJOW cordset.
The amp ships without an end plug, so
You will need to supply what ever you're using at your shack for your 240Vac
mains. Since there are no 120Vac components used internally, and the cordset
is three-wire, the GND is tied directly to chassis at the input. The two HOT
leads go first to 20 ampere fuses accessible at the rear panel. HOT then
proceeds to the cabinet interlock, and front panel switch which carries only
relay solenoid current. Both sides are then switched through a large, 277Vac
25-ampere contactor to a proprietary soft-start thyristor circuit, which
generates a true AC ramp function. The soft start takes about 2 seconds to
complete, and operates the entire amplifier. Everything from the meter lamps,
to the tube filament (12.6V) and the HV p/s are ramped. The AC input is wired
with heavy, short runs, and is done better than most AC input circuits I have
seen. Attention seems to have been paid to minimizing AC input losses through
adequately sized conductors and terminations.

RF Deck
The RF deck components are large by today's standards. All
wiring components, straps, taps, coils, etc are silvered.
The L1 tank coils are entirely wire-wound, toroids are not
used for the lower bands. Coils for the upper HF bands are
heavy, 1/4 inch silvered stock, and for the lower bands
are silvered #8-10 wound and mounted on a heavy ceramic form.
Ceramic/porcelain standoffs are used throughout, something you don't see too
often in today's amplifiers. 
Workmanship around the tank wiring and taps is excellent. The size and quality
of the components and workmanship are similar to the old ETO 77dx.The band
switch is a five deck silvered ceramic switch. All nine amateur bands are
supported. Three decks are close spaced, and are wired in parallel for current
handling capability. The two additional decks are used to add C1 and C2
padding capacitance for 160m. The padders are the largest I have ever seen,
measuring 1.5 and 2 inches in diameter for C1 and C2, respectively. The output
tank circuit is a Pi-L with the output L an untapped 4:1 coaxial UNUN. This
transforms the image impedance of about 200 ohms to the 50 ohm output
impedance.
The RF choke is the largest I have seen on any amateur
amplifier, measuring about 7/8" diameter by 7" long on a
ceramic form. The plate blocking capacitor is neatly
fitted directly to the end of the RF choke, and is a 2700pF, 15KV component.
Plate tune and Load capacitors are large air variable, and driven with high
quality ball-bearing reduction drives from the front panel.
The 4CX1600B is housed in a huge ceramic chimney with a
Wall thickness of a half-inch. A parasitic suppressor connects the anode to
the plate choke/blocking capacitor. The tube is similar in shape and size to
an 8877.

RF output Wiring
Teflon coax (RG-303) is used on all output wiring to the
QSK board which uses a Jennings RJ-1a vacuum relay for the
output. Full break-in operation is supported at full power
on all bands. A shielded and enclosed RF sensor is used
for the forward, reflected and SWR Fault detector.

Operation
There is a 2-3 minute wait for warm up. The Ready LED does
not flash during warm-up, which would be a simple feature
to add, and one that is needed. As received, all LEDs are out during warm-up
until the timer has expired, then the ready LED comes on. The factory presets
for all nine bands were right on the mark. The amplifier puts out 1,750 -
1,875 watts on all bands with 60 watts of drive. With 50 watts of drive, the
amplifier is at 1,500 watts out.
If an incorrect antenna is chosen, the amplifier faults
immediately to bypass. I was not able to get the amplifier
to flash over in any circumstance with the wrong antenna
selected. Anything over about 2.5:1 kicks in the
protection circuit, a very nice feature. This feature will protect against
potentially significant RF deck damage in the event an antenna comes down, is
covered with ice, or is simply the wrong antenna for the band. Tuning is
simple, and similar to tuning a triode amplifier. After pre-sets are chosen,
apply 25 watts and peak the Load and tune. Slowly increase power, while
peaking the output, and watching the Ig2 LED's for screen grid current.
Slightly heavier loading keeps the Ig2 LEDs to one or two RED segments, which
represents about 20-25mA.
The amplifier will immediately fault to bypass of too much
input drive power is used. Both the SWR and Overdrive
faults are reset automatically in two seconds. If the fault
remains, the amplifier will not re-key. Observation on a scope of RF output
during QSK operation was perfect with both full and semi-break-in.

Cooling
A high quality blower is used, and it moves a lot of air.
It appears to be a 220Vac 50Hz motor. At 60 Hz, it runs a bit too fast and has
a bit more high frequency white noise
Than I would prefer. It is run at about 120Vac in the normal mode, using a
primary tap on the transformer as an auto transformer (no neutral current on
the mains). There are temperature detectors in the tube chimney air flow that
will switch to 220Vac operation at about 100 Deg C. I never heard the blower
increase in speed. The tube runs very cool, with efficiencies in the 65% range
on all bands, and cools very quickly after CW or tune operations.
Air flow in this amplifier is excellent for the tube, but
not as good for the other two items that need attention.
First, the tank components should have some air
Circulating by them, and the p/s diodes and bleeders should also be in the
intake air path. There is some intake air that flows by the p/s, but none
flows through the tank.
In this amplifier the tank components are so large, not
Much heat is generated in any of the tank components, except the #10 ceramic
coil form, which does get warm. The P/S diodes are so large they also run
cool, but the bleeders and B(+)glitch resistors do get warm.
This is about the only area in the entire amp that I can
complain about, but realistically, even with RTTY operation
at 1,500 watts continuous, it would probably never be a
problem.

PC Boards
Glass epoxy boards are used throughout, for the HV p/s,
Control board, Display board(s) and QSK. The HV p/s,
Control and Display boards are all replaceable with push pin connectors.
Should a replacement or upgrade be necessary, use of connector housings
(rather than individual pins)would greatly simplify that exchange.

Metering
I chose the DX-2a version because I am partial to analog
meters. The digital version has the advantage of seeing
all five parameters at once, with no switching.
The meters uses are very nice, large bright white meters
with easy to read back-lit scales. One is fixed for plate
current, the other is switched for Forward Power,
Reflected Power and Plate Voltage.

Ip = 1.5A f/s
Vp = 4KV f/s
Pf = 2KW f/s
Pr = 200W f/s

There are several spare positions on the meter function
switch. I would like to see filament voltage added, since
it may vary with line voltage. Since this is a tetrode with a 350Vdc regulated
screen supply, it might be nice to read
that too. Screen Grid current is also metered with a 6 segment LED, reading
from neg 10mA to pos 35mA. The Ig2 LEDs are a great tuning aid, and make it a
very fast chore to peak the tune and load, while keeping the loading
sufficient to keep Ig2 below 25mA. At 35mA all four segments light, at greater
than 35mA they flash. This is one of the better tuning aids I have seen, and
is very intuitive to use.

Documentation
The users manual is one of the most extensive of any
amplifier I have used. There are diagrams for linearity,
overdrive conditions, QSK timing, etc... Partial
Schematics are provided and the operation section is well written and
thorough. A pre-set chart is provided with factory settings included.

Summary
While most US amplifiers have concentrated on cost
Reduction in recent years, it's a pleasant surprise to see quality improving
at a very fair price. This amplifier is a great value for the money (about
$3200 delivered with QSK). It is built with more expensive components than
most US amps selling for more money. Workmanship is excellent, and
Layout is modular and easily serviceable.
The performance is flawless, and the protection circuits
work well. You can not easily damage this amplifier, so
the user is protected from simple and common mistakes.

A few recommendations to future revisions:
1. Flash the Ready LED during warm-up;
2. Duct the intake air by the tank components and P/S;
3. Add a few more metering functions to the meter switch;
4. Use AWG 14-3 SWOJ cordset;
5. Add complete schematics and calibration/alignment
 procedures;
6. Add connector housings to the wiring harness for PC
 board connectors.




--
FAQ on WWW:               http://www.contesting.com/ampfaq.html
Submissions:              amps@contesting.com
Administrative requests:  amps-REQUEST@contesting.com
Problems:                 owner-amps@contesting.com
Search:                   http://www.contesting.com/km9p/search.htm


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