A few words in defense of Henry is in order here....
> >-henry doesn't change the bias on cw/ssb. The bias consists of
> >15-20 diodes in series in one block like a W2AW module. The
> >idle current zsac goes up and down cuz of plate V difference
> >between SSB V and lower cw V. Manual sez no load HV on ssb
> >should be 5400V . Henry uses a 6kv plate xfmr with a choke
> >input..... .9 x 6000= 5400V. They also suck a continuous
> >300W of bleeder power all the time to make the tuned choke,
> >look like a choke.
The 3CX3000A7 runs just fine with zero bias. Henry elected to add a little
bias to cut down a bit on heating of the operating area, which Texans
appreciate. With a 3X3 throttled back to 3500 watts PEP, several things
can be done that might not look kosher to someone designing for 6000
watts.
> > If any of the 5 x 100 watt bleeders ever
> >opened up, the HV would skyrocket to 8485V. And that's if you
> >have the correct tap to start with.
> >
> // However, the filter capacitor could probably take the surge. When
> the amplifier was keyed, the ZSAC acts as a bleeder current.
Amen, Rich...if a bleeder failed during key-down, you still have at least an
800 watt load on the HVPS from the ZSAC from the 3X3. The filter caps
are 2X-16 mfd @ 7.5 kv oils. An 8400 volt surge during bleeder failure
is no big deal here.
>
> > The fil V should be below 7.5 V measured right at the
> >socket. Life of tube is dependent on fil V. If the tube is
> >lit, the clock is ticking. The tube won't last any longer cuz
> >you are running it at 1500w out. If you had a 2nd amp. with
> >just the tube fil lit, and no drive applied, and no HV, it
> >would last just as long as your driven amp.
Henry takes care of two problems in the filament circuit with one
simple part. A power resistor (R617) in the primary of the filament
transformer limits inrush current and allows setting of the filament
voltage to the correct value for whatever line voltage that may be
encountered. A filament voltmeter is included, which can be calibrated
to a precision measuring source, so filament monitoring is available
at the flick of a switch.
> >The stock blower on the 8K ultra will not provide enough air
> >for anywhere near 4kw of anode diss, more like 2500w tops. The
> >stock straight up and down chimney puts quite a restriction on
> >the airflow.
You get what you pay for. The 8-K is rated for 3500 watts PEP MAX.
No where near 4 kw dissipation......
I restate...look at the manufacturer's design targets and specs before
making judgements of component selection. The cooling system is,
IMO, very inovative, a la the Heath SB-220. A fiberglass cylinder is fitted
from the tube anode cooler and runs up to the top cover where the hot
air is exhausted through holes in the cover. Air enters from below deck
through holes around the tube. There is enough air available to circulate
throughout the tank circuit components before entering the anode cooler.
The ambient heat build-up inside the RF deck is many degrees cooler
than with conventional methods. Heat generated by the tube is segregated
from the other components on the deck.
> // A high-pressure blower from Kooltronics or Rotron would cure this
> problem. Henry engineering is apparently clueless about manometers and
> the effect of filament potential.
Not in the case of the 8-K. Those wishing to exceed Henry's specs and
operating instructions will need to look at Rotron et al to beef up for QRO.
> >
> >The henry back to back un-uns are interesting though I don't
> >understand the theory behind them. They operate from 1.8-30mhz
> >continuous.
>
> // But the SWR is somewhat less than satisfactory.
The target for the 8-K was an input circuit that will guarantee <2.0-1
SWR on any frequency from 1.8-30 mhz. with a MAX drive power
of only 150 watts. Also the target for feculance was >-60 db second
harmonic and >-35 db IMD @ 3500 watts PEP output. By throttling
the 3X3 back 2500 watts and adding a gang-tuned Pi-L tank circuit,
these targets were easily met with the components Henry selected
for the job.
My guess is that if the intent was to produce a 6 kw amp with the
same specs, you would see pi network input circuits, (it would take
18-20 of these to cover 1.8-30 mhz with no gaps) a bigger blower,
and a 2.5 amp Ip meter. Also included would be a much larger price
tag!
Sometimes the end justifies the means. Amateurs today are not
used to seeing over-engineering when they look "under the hood"
of an amp. I have to admit, when Henry attempted to be competitive
in the amateur market by "dumbing down" their fabulous commercial
designs, some really bad mistakes were made. The 8-K is not one
of these "mistakes!"
(((73)))
Phil, K5PC
--
FAQ on WWW: http://www.contesting.com/FAQ/amps
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
|