At 02:19 PM 5/17/97 +0000, you wrote:
>Hi Phil,
>
>The Ten Tec Centaur measured -39dB (3rd) and -45 dB (5th) in the ARRL
>review, and uses no "special length" cable or tuned input! The
>AL-811H measures about -41and -48 when operated at the same power
>level of the Collins (it uses neutralization and tuned input).
Too much mixing apples and oranges here. No tube exciter was used in either
test, and they both have pi-l tank circuits. The 30L1 and 30S1 do not.
>I've tested several dozen amps for IMD over the years, and find the
>feedline length makes no difference at all in IMD results on lower
>frequencies. Feedline length does affect stability and IMD at
>the high frequency end in PA's without or with poor neutralization.
This would never show up if you did not use a tube exciter with a pi
network in its tank circuit.
>
>> How can furnishing a 20.5 piece of coax with each radio require an excuse?
>> Was there an over-stock of RG-8 at Collins, or a conspiracy with Belden?
>> (I don't recall any other manufacturer of amps giving me any coax of any
>> length.) I am sure the price of the coax was justified in improvement of
>> specs.
>
>Of course there was a reason. It helps terminate the input in one
>of the few four 811A tube PA's manufactured without neutralization.
The instructions in the 30S1 manual state the same thing...20.5 inches
for the S-Line, and 23.0 inches for the KWM-2.
>
>If a PA is properly designed and constructed, input cable length
>or exciter type has no effect at all on PA performance. Failing
>to meet this criteria is NOT always a fault with engineering or
>construction. Sometimes cost, time, and component availability limits
>the results.
Again, we agree because this problem apparently does not show up anymore
with modern transceivers. The 30L1 had many time constraints during its
development. Orders were coming in, and Collins was waiting on Eimac
for the 3-500Z tubes to put in it. The 811A's were an afterthought when
Eimac failed to produce. Maybe the "free coax" with the 30S1 was from
"force of habit," or if it really makes no difference, 20.5/23.0 ft of
coax is a good as any. They did not remove the caveat from the 30S1
manual, however.
>
>The 811A was an offspring of a tube originally designed for audio
>service. Initially the tubes were low-mu 810 triodes designed for
>class B modulator applications. RCA added a hi-mu grid and made a
>zero bias triode, and eventually the tube got pressed into RF service
>despite the long thin grid lead and poor internal shielding.
It supposedly got pressed into service at Collins by a ham who worked
there and built up the circuit at home for his personal use. When Eimac
left Caooins up a creek with no paddle, the ham showed the HB amp to
Mr. Collins and the rest is history.
>
>When a less than optimum designed tube is the heart of ANY PA, the
>design will almost always be full of contradictions and compromises.
>A pretty round emblem and good reputation can't correct all the tube
>problems.
>
>If you look at the Heathkit Warrior and Gonset 811A PA, you'll see
>they are neutralized. So is the AL-811H. In ALL three of these PA's,
>you can unterminate the input and rotate the tuning through all
>positions with no instability. In all three of these PA's, input
>cable length has little or no affect on IMD.
I agree totally. This had to be a problem unique to Collins TUBE type
equipment. The AL-811H is far superior to the 30L1 in many respects,
especially the power supply. I get a good laugh at seeing these old
30L1's being resurrected at exorbitant prices just to run 500w or so
output, with the power transformer running red-hot on the digital modes.
>
>If you try this test with the Dentron and Collins, most of them will
>oscillate on ten and 15 meters. They will all oscillate if you select
>the correct input cable length.
It makes me wonder how different the new breed of "off-shore" 811A's
are from their predecessors. Some types say "do not install in a
horizontal position." The only fair test to measure the effects of the
20.5 ft. cable IMO would be with original Collins OEM tubes installed
in both the amp and the exciter.
>
>I'm not saying the 30L1 was poorly constructed, but it certainly
>would have been a better PA if they neutralized the PA. In that case,
>any length cable could have been used.
At least they had the presence of mind to install a tuned input. Many
of the amps of that era had no tuned input and strait-pi tanks.
>
>Let's just agree to disagree. From an engineering standpoint, I view
>the problem differently than a consumer.
I am definitely on the consumer end of the spectrum. I will always
think that Collins had a very valid reason for the coax with the 30L1/
30S1 and collins exciters. Mr. Bruene lives about 10 miles from me,
and no longer is employed by Collins. I will see if I can glean some
historic lore out of him. I think he will say that all references to
coax length can now be deleted from the manuals, if you are driving
with rice boxes or Ten Tecs!
>
>73, Tom W8JI
>
>--
>FAQ on WWW: http://www.contesting.com/ampfaq.html
>Submissions: amps@contesting.com
>Administrative requests: amps-REQUEST@contesting.com
>Problems: owner-amps@contesting.com
>
>
--
FAQ on WWW: http://www.contesting.com/ampfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
|