[AMPS] Power Handling of Resistors

[Tom Rauch]

Personal attacks ignored, I have a few comments:

From:           	Wt8r@aol.com
Date sent:      	Tue, 7 Mar 2000 10:18:06 EST

> Wow....this is rocket science or is this from the General class question
> pool?

 >  >  So the question is: >  >  Rich claims that suppressor resistor
> are damaged by a nearly instantaneous >  output of VHF energy due to an
> "oscillation."  This "oscillation" occurs so >  fast that the panel meters
> on the PA don't have time to respond.  This, as >  W8JI said, is probably
> on the order of nano-seconds or even a few >  milliseconds at MOST. >  > 
> If these same resistors can handle 160 Watts of VHF energy for 2 seconds,
> >  how much could they really handle in less than a thousandth of the
> time?  I >  doubt enough to be damaged as Rich claims.
> 
> This is the point where ignorance overtakes (overshadows) common sense. 

> (1) Apparently Jon hasn't the slightest idea of the magnitudes of
> difference between the amount of energy stored in the power supply of a
> multi-kilowatt power supply hooked to 220 VAC mains and a toy desk top 160
> watt amp.

It's all watt-seconds Dave, as long as two parameters aren't ignored.

1.) The applied power must be high enough to destroy the parts, 
and 160 watts could certainly do that. 

2.) The applied power can cause destruction from mechanisms 
other than heating, so you have to be careful the part does not arc 
internally and explode from expanding gasses or have magnetic 
fields that damage the part.

I think Jon was within those rules, and I'm sure a parasitic would be 
also since the current can only be driven by the tubes time-varying 
resistance and maximum saturated current.

 (2)  Neither does he have any concept of the DIFFERENCE  in the
> application of that small amount of power for 500ms, the length of which
> is controlled by the pushing of a microphone button, as opposed to the
> UNCONTROLLED avalanch of the dumping of ca. several joules of energy in a
> very short time which usually ends in the catatrsopic destruction of a
> bandswitch or other tank component.

You are mistaken Dave. A conventional PA can not dump "several 
joules" of energy into the bandswitch in a short period of time.

The PA excites the tank by charging and discharging the blocking 
capacitor, and moving tiny amounts of energy at a time in-phase 
with the tank resonance until the current and voltage product 
equals something below the dc power supplied to then output 
device.

Even at the resonant frequency of the tank, where everything is 
perfect, you simply can not move many joules of energy in the type 
of  amplifier we all use into the tank system, and that includes the 
bandswitch. It takes many cycles of RF to even build the energy to 
a kilowatt.
    
> What follows is pure sophistry and, Yes, he has not thought it through.
> 
> >  For that matter, I've not thought this through, but how many RF cycles
> >  at VHF could take place in a matter of a few milliseconds?  I guess
> >  you'd need to calculate the period, which is 1/F.  At 144 MHz, this is
> >  6.9 nano-seconds.  So if our event is on the order of nanoseconds you
> >  could get just a couple of RF cycles going which ain't gonna hurt
> >  anything.  The 
> point
> >  is that a short event as Rich claims is what happens isn't going to
> >  stress components beyond a breaking point especially when it is so
> >  short it 
> doesn't
> >  even show up on the meter movements.
> >  
> >  At least that's the way I see it.


Sorry Dave. It looks like Jon has considered the problem.

Like those who think RF currents can explode tungsten control 
grids in a nanosecond, I suggest you look at the saturated 
emission current of some power grid tubes and the thermal and 
mechanical "inertia" or mass of the elements and calculate the 
amount of energy required to damage them from heating or 
magnetic stress.

We are assuming the cause of the failure is actually the victim of 
some esoteric uncontrolled act of fate. Fitting for modern society, 
but technically incorrect.

I can't make a 3--500Z grid explode no matter how much current I 
drive from the filament emission, because filament emission is 
limited to somewhat over ten amperes under ANY condition of bias.

Bandswitches generally fail from over-voltage, they almost certainly 
can't fail from VHF parasitics. The VHF parasitic would have to 
drive the tank capacitor to a much higher voltage swing than the 
normal HF energy, and its impedance is very low at VHF. If the 
tank is doing its job suppressing harmonics, it won't be able to 
develop high VHF voltages.

Its nice to have one simple answer that makes everyone an expert, 
but there are multiple reasons that PA's fail. The unfortunate thing 
is one of the least likely causes has been touted as the main 
failure mechanism.


73, Tom W8JI
w8ji@contesting.com

--
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