Amps
[Top] [All Lists]

[AMPS] TL-922 Filament Transformer Protection

To: <amps@contesting.com>
Subject: [AMPS] TL-922 Filament Transformer Protection
From: measures@vc.net (Rich Measures)
Date: Sat, 5 Jun 1999 12:34:09 -0700


>
>> CT.  .  .  The Kenwood 922 has a tendency to intermittently oscillate at
>> its anode-resonance freq. of c. 130MHz.  Such oscillations may cause
>> intermittent spitting at the Tune-C and or bandswith, large changes in the
>> R of the vhf suppressor resistors, and bursts of high grid current which
>> place lateral electromagnetic force on the hot filament-helices.  With
>> repeated oscillations, the filament may eventually be bent far enough to
>> touch the grid cage.  When the filament touches the grounded-grid, the
>> +110v power supply shorts to ground.  If the amplifier is not switched
>> off, the filament transformer will melt down.  One fix is to cut the wire
>> on the bias relay that connects from the coil to the bias contacts.  For
>> purists, the existing 100k resistor can be connected across the NO bias
>> contacts on the relay.  .  - later, Court
>
>
>Rich gave good advice, except for the parasitic stuff. He has a 
>fixation and pet theory that almost any failure is do to a "parasitic". 
>Sorta like ethnic cleansing applied to PA's, where those nasty 
>conventional suppressors have to go.
>
?  A photograph of a grid/filament shorted 3-500Z appears on p. 15 of the 
September, 1990 *QST*.  The title of the article is "Parasitics 
Revisited".  

>The bulk of arcing failures in the 922 (and other PA's) are caused 
>by improper relay sequencing, 

?  The 922 was designed for the TS-820S transciever.  The RF relay 
switches so slow that hotswitching takes place with modern transceivers. 

> improper loading, 

?  I have neither seen nor heard of this problem. 

> driver transients,  

?    For fundamental freq. arcing to occur at the 922's 6kV Tune-Cap, the 
3-500Zs would need to develop over 6000w pep.  Does this seem possible?

>or load faults.
>
?  true enough.  An arc in the antenna tuner can arc the 922's tank.  
Same thing for an arc in a damp coax connector or an arc in a defective 
lightning arrestor.  

>Anytime you  drive a PA without proper loading, the tank voltage 
>soars and can reach several times the normal operating voltage. 

?  My SB-220 will not arc the tank when it is fully driven and fully 
mistuned.  The Tune-C in a SB-220 typically has a withstanding potential 
of approx. 4kV.  During the Grate Debate, someone actually measured 
maximum Tune-C potential in a mistuned sb-220 at 3600v-peak.  Normal 
Tune-C potential is approx. 2600v-peak on SSB. 

>That's an effect we are all familiar with in SB-220's, Viking Valiants, 
>Rangers, DX-100's and other tube equipment. In MOSFET PA's, it 
>just wipes out the FET's from drain to gate punch-through.
>
>A common problem in the 922 is they never bent the relay contacts 
>slightly, to be sure the antenna connected before the drive could be 
>applied. 

?  I agree that current transients brought about by hotswitching 
stimulate parasites.  .  However, bending contact supports will not 
prevent hotswitching if RF comes down the pipe before the relay contacts 
have stopped bouncing.   The fix for hotswitching is either to buy a 
TS820S, or to install high speed switching  (approx. $60).  

> Another problem is rigs like the TS950, or IC775, that slam 
>the PA with a few hundred watts of drive. That extra drive switches 
>the tube hard, and tank voltages momentarily build up to many kV 
>because the tube is misloaded for that drive level.
>
? ..... if you happen to have a pair of 3-500Zs that can deliver over 
6000w output -  and you remember to disconnect the load.  .  

>Another common cause is operator error, where the operator tunes 
>the amp at lower power and never loads the amp at full drive.
>
?  It is true that this malpractice temporarily increases the potential 
across the Tune-C.  A maximal increase of perhaps 25% seems possible.  In 
a 922, this is more than 2000v shy of what is needed to arc the Tune-C.  

>Also, lightning arrestor sometimes arc, and that disconnects the 
>load allowing the tank to build up high voltages.
>
?   The tank typically arcs.   The load is still there.   The arc in the 
arrestor limits the potential across the coax.  .  

>While I can't say for certain an amp could never oscillate and 
>cause arcing, it is probably one of the least likely causes of 
>failures. I've never been able to get one to oscillate and arc, 
>however, even when I intentionally introduced VHF oscillations.
>
?  As I recall, during the grate parasitics debate you stated that you 
were able to make an AL-80 oscillate at c. 155MHz.  
.  Dick Erhorn claims that no Alpha has never sustained a 
parasitic-oscillation --  even though he has reportedly seen plenty of 
gold-sputtered tubes.    

- cheers, Tom.  


The long excursion on the wide river in the Land of the Pharoahs 
continues.  


Rich...

R. L. Measures, 805-386-3734, AG6K, www.vcnet.com/measures  


--
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>