Amps
[Top] [All Lists]

SV: [AMPS] SB-220 conundrum

To: <amps@contesting.com>
Subject: SV: [AMPS] SB-220 conundrum
From: sm5ki@algonet.se (sm5ki)
Date: Wed, 19 Jan 2000 16:44:48 +0000
Hello Pete:

I usually start my parasitics hunt with the old Millen grid dip meter and
the amp disconnected from the mains. You will be surprised how many odd
resonces there may be.

If you find a resonant frequency in the grid or cathode circuitry try to
rotate your C1. If there is a change in the dip you may have a tgtp
oscillator at that frequency.

Also I try to parasitic test my amplifiers with higher then normal resting
current. The small signal gain will be higher and any tendency for
parasitics will be increased.

I never tried this on a SB220 but could there be some new resonant
frequencies in the 3-500Z grid circuitry?  The paralell circuit of the
negative feedback condensers and grid chokes maybe are resonant somewhere?
And the plate choke maybe has the same resonant frequency?  Maybe you tried
earlier to run the amp on one of the  bands not avaible the SB 220 and one
of the grid chokes was hurt by a parasitic and changed its self resonant
frequency? If I remember right, the new bands where not avaible in 1968-69
when the SB 220 came out and the plate choke nay very well be resonant on
the newer bands?

Lets take an example how I fixed a Swedish copy of the old COLLINS 30L-1:
the amp was completely stable on all bands except on 80 meters. There the
811As quickly turned red when trying to tune up and many a tube quickly went
kaputt with broken anodes. They had on 80 meters connected an additional
capacitor section  to the input C1 pi variable cap and the inductance of the
connecting wire and C1 became a resonant circuit somewhere at 40 -50 MHz as
indicated on the grid dip meter. The same resonant frequency was found in
the bifilar filament choke. When rotating the C1 there was a change in the
filament choke resonant frequency as indicated on he grid dip meter. The fix
was simple: I moved the resonant frequency of the added cap section down by
inserting a small coil in series with the connecting wire.

Hope you fix your problem. 73s de Hans SM5KI
----------
>Från: Mike Baker <bakerhouse@uswest.net>
>Till: Pete Smith <n4zr@contesting.com>, AMPS reflector <amps@contesting.com>
>Ämne: Re: [AMPS] SB-220 conundrum
>Datum: ons 19 jan 2000 08.32
>

>
>Hello Pete,
>    Yep, sounds like something isn't doing what its supposed too but, here is
>an idea or two you might try to figure out what freq your power output is on.
>    Best Idea.
>    I see by the map maker on Buckmaster that you are not too far from the
>Baltimore area.  Hopefully if you can't find some ham in your area that has
>access to a spectrum analyzer perhaps you could get to someone in the Baltimore
>area who would be willing to get together at your place and make some accurate
>measurements of your system to determine just what freq your signal is on.
>    Next best Idea.
>    Find someone in your area who has either an absorption wave meter or a dip
>meter that will work as an absorption wave meter and sample the output.  To do
>this without having to maintain full key down power output try this.  Tune
>everything up into a dummy load with full drive using your electronic keyer as
>a tuning pulser and set the dots to max (about 40+WPM) preferably using a PEP
>reading watt meter.  See if reducing your drive follows in a linear fashion.
>I.E.:  if 125 watts of drive produces 1450 watts output then 62.5 watts should
>be somewhere near 725 watts output and 31 watts of drive should be about 365
>watts output etc.  Your original post said that your put out about 1100 watts
>tuning up using the dip method and when tuned up for full max output you showed
>about 1450 watts.  In my mind that means that approx. 350 watts is going
>somewhere it shouldn't and represents a possible spur or parasitic that is
>approx. -6 DB below the fundamental output.  With the transmitter keyed as I
>said above and running at reduced output of about half of full tilt, the
>spur/parisitic should be easy to find with the wave meter.  Measure as closely
>as you can its freq.  If you don't have an accurately calibrated signal
>generator to send into the wave meter to check the exact freq it is set too,
>then use a frequency counter to check any old signal generator into the meter.
>The end result is to end up measuring the freq of the unwanted signal, IF THERE
>IS ONE.
>An assumption.  (Yea, I know what they say about assuming.)
>From what I have read in the info presented by others on the Amps reflector, a
>pair of 3-500Zs tends to have a self resonate or parasitic that is at approx.
>40 MHz. Normally your PC suppressers would take care of this and if they were
>tuned to this freq they would go up in smoke trying to dissipate 350 watts so I
>am left to presume that the extra power is not on that freq but is somewhere
>loosely within the pass band of your tank assembly. A Pi network is a low pass
>filter so I would presume that at 350 watts additional output the mystery
>signal is not much higher/lower in freq. than the fundamental.  Most amps have
>someplace in the -35 to -40 (or more) Db range of suppression of harmonics so
>that is what I base my assessment on in terms of where this mystery signal
>might be.
>    A second thought.
>    If you have found some other weird output freq from the above test on your
>wave meter or spectrum analyzer don't forget to check the exciter output to see
>that your are not generating the signal there in the first place.  OK, perhaps
>a bit of a stretch but it could happen.  I have seen it before when just the
>right/wrong combination of transmitter tuning, coax length, final matching
>network, input matching network, etc. would produce some weird response in an
>amp.
>    Where you go from here is anyone's guess.  I don't have the skills or
>experience to tell you where to go from here but if you can make the test I
>mentioned above and then post your results to the group, perhaps one of the
>resident experts could be more helpful.  The first test should be to make sure
>your test equipment/watt meter measurements are accurate and then proceed.
>I am sure some of the AMPS experts will have some pearls of wisdom to impart to
>you with respect to your problem.
>    Good luck and let us all know what your find.
>    OK guys, fire proof skivvies inplace, let her rip. ;>}
>    Best 73 and happy hunting.
>    Mike Baker  K7DD
>
>Pete Smith wrote:
>
>> Here's a weird one for the group.
>>
>> I replaced the T/R relay in my SB-220 with a pair of vacuum relays -- an
>> HC-1 on the output side and a tiny Hi-G d'Italia on the input side.  It
>> works great, but there is one anomaly.  The input SWR on 40 meters (only)
>> is high (~2.2:1) and this seems to be paired with abnormally high plate
>> current and output power.  All other bands are fine.  Just for fun, I tried
>> bypassing the input relay entirely, but that had no effect on the input
>> SWR, so I'm inclined to think the relays aren't the problem.
>>
>> Running into a dummy load at ~ 1.1:1 SWR and using my HM-102 wattmeter (no
>> Bird, but power levels presumably indicative, at least relative to each
>> other), I find the following:
>>
>> 1.  power output is ~1100w on 80, declining to ~950w on 10, all with the
>> exciter showing 125 watts output and the SB-220 grid current ~200 ma.
>> Plate currents are ~750 ma.
>>
>> 2.  But on 40m, power output (PLATE and LOAD tuned for maximum) is ~1450w,
>> grid current ~250 ma. and Plate current ~900 ma.  There is also a fairly
>> striking difference between the PLATE setting in this condition and the
>> setting for minimum plate current.  If I tune the PLATE for minimum plate
>> current and LOAD for maximum output, the figures look much more like those
>> on the other bands.
>>
>> Any idea what's happening?  I find it hard to equate higher input SWR with
>> apparently higher gain.
>>
>> K9MA has suggested that in any case it would be a good idea to re-do the
>> input tanks on all bands to raise the Q and improve the SWR, and has kindly
>> given me values that he has tested in his amp, but before I embark on all
>> that I'm curious what people think about this specific situation.
>>
>> 73,  Pete N4ZR
>>
>> Don't forget to update your entry in the World Contest Station Database
>> http://206.102.70.3/search.htm
>>
>> --
>> FAQ on WWW:               http://www.contesting.com/ampsfaq.html
>> Submissions:              amps@contesting.com
>> Administrative requests:  amps-REQUEST@contesting.com
>> Problems:                 owner-amps@contesting.com
>> Search:                   http://www.contesting.com/km9p/search.htm
>
>
>--
>FAQ on WWW:               http://www.contesting.com/ampsfaq.html
>Submissions:              amps@contesting.com
>Administrative requests:  amps-REQUEST@contesting.com
>Problems:                 owner-amps@contesting.com
>Search:                   http://www.contesting.com/km9p/search.htm
>
>

--
FAQ on WWW:               http://www.contesting.com/ampsfaq.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>