[AMPS] Re: Parasitics

[Rich Measures]

>> Date:          Wed, 13 May 1998 21:08:41 -0500
>> From:          Jon Ogden <jono@webspun.com>
>> Subject:       Re: [AMPS] Re: Parasitics
>> To:            w8ji.tom@mcione.com, amps@contesting.com
>
>Hello Jon,
>
>
>I said:
>> >It is possible, by unloading a tank in a conventional PA, to generate 
>> >voltages of many times normal operating voltages. Underloading of a 
>> >tank is the single largest cause of tank circuit arcs, and is a 
>> >repeatable and measurable cause source of damage.
>
>You said: 
>> If the HF tank voltage can vary wildly and do what you say, which I 
>> belive it can, what would prevent the VHF voltages in an oscillatory 
>> condition from doing the same thing?
> 
>Very good Jon. Nothing at all, except the load presented by the 
>tank. The reason the HF signal can swing so freely is the tank has 
>high HF Q and is resonant. 

Just for fun, the next time you are doing a Spring-cleaning/oiling on 
your HF amplifier, pull the mains-plug and couple a dipmeter to the tank 
L in order to find the resonance.   Is the tank resonant on the frequency 
where you tuned up the amplifier?    Is a pi-network tank resonant, or is 
just two L-networks in series? 

>For VHF signals, the anode can swing quite 
>freely but the tuning cap looks like a very low impedance.

Unless the Tune-C is near one of its self-resonances.  
>
>This is easily measurable.
>  
>> If the tube has gain at VHF (which it does, we are all in agreement on 
>> that) then what would prevent the tube from generating voltages high 
>> enough to arc or discharge across the bandswitch in a runaway oscillation?
>
>The tube, the tuning capacitor, and the series impedances of the 
>leads in the PA. This is why I suggested exciting the anode (of a 
>cold PA) with a sweep or single frequency swept source with a source 
>impedance about average for the anode impedance.
>
>Consider this point. How much power can you get from a 3-500Z at 180 
>MHz or so if you INTENTIONALLY try to amplify or build and 
>oscillator? Now imagine the likelihood that would occur by accident 
>with small wire leads..

Most 2x 3-500Z amplifiers appear to have an anode-resonance of 95 - 130 
MHz.  
>
>It is easy to track voltage through the tank system, and watch it 
>change. The problem at VHF is often  one of transmission line 
>transformer effects. 
>
good point

>> Why is it possible at HF and not at VHF?
> 
>No one said it isn't "possible", but it is "impossible" t generate 
>such voltages in a tank that presents a very low impedance at VHF, or 
>one that has a considerable ratio of impedance transformation in a 
>downward direction.
>
>Virtually ALL amplifiers that have acceptable VHF harmonic levels 
>have very good VHF attenuation starting at the anode terminal and 
>moving through the tank.

... except near frequencies where the Tune-C is self-resonant.  
>
>> To me, it doesn't make sense to claim that in one operating frequency 
>> region an amp will do one thing, yet in another region where it can 
>> easily operate, it will do something completely different.
>
>Exactly right. Think about this conundrum. 
>
>Some people claim VHF signals, at frequencies where tank impedance is 
>VERY unfavorable for both generation and transmission of signals (and 
>this can be measured and reliably confirmed to be such), can cause 
>the tube to flash over, 

Based on autopsy-evidence, the only case of tube flashovers appears to be 
a secondary effect of VHF oscillation -- i.e., in 8877s and the like, 
during VHF/UHF oscillation, thin layers of gold boil off the grid, which 
may cause anode to grounded-grid flashovers.  In a healthy 8877 the anode 
withstanding potential is over 20kV, which means that if something runs 
amok, the Tune-C will be the first thing to flashover.   

>the tank capacitor to flash over, and the 
>band switch to flash over.
>
agreed, however, the Tune-C would seemingly not be able to do so where it 
presents a low Z.

>At the same time, they claim such voltages can never be reached at 
>the main operating frequency, where tank Q is highest and impedances 
>are optimized!

I have seen it happen when an antenna tuner cap. arcs.  
>
>Pretty weird science, isn't it?
>
.*Weird Science* was an awesome flick.  

>> Please explain why HF energy can cause arcs and huge voltages while VHF 
>> energy cannot?
>
>Did that above. Impedances than can be measured are wrong. 

.    HELLO

>Now you do the same, and explain why very high HF voltages can not be 
>generated.
>
>> I am ignoring the presense of the VHF parasitic suppressor.  I am 
>> assuming that it is either not working properly or is blown.  IMHO, a 
>> properly constructed supressor will effectively limit that VHF voltage.
>
>Not much. It's job is to load the system with series resistance. 

It seems to me that series resistance unloads and that shunt/parallel 
resistance loads.  
>
>The optimum suppressor would have an equivalent parallel resistance 
>(remember it is two components in parallel) that, when connected in 
> series with the anode lead, is greater than the impedance of that 
>path.

If the Rs/Ls combination has a high VHF impedance, then the VHF voltage 
amplification of the tube would be high.  Is more VHF amplification in an 
HF amp. mo' betta?  
>
>That insures the anode is loaded by a lossy resistance, rather than a 
>non-lossy reactance that mostly shifts phase.

makes sense
>
>That's why, if you look at PA's that have long thin leads in the 
>tubes (look at 811A's in this example), tank to tuning cap 
>connections, and so on you will see these amps need a lot of 
>suppressor inductance and higher suppressor resistance.

agreed, which means that VHF-Rp decreases, however, the trade-off is more 
28mHz dissipation in the suppressor R.  
>
>On the opposite side of the coin you have external anode tubes with 
>grid cones that terminate in a flange at the base. These tubes have a 
>very short wide low impedance connection and often require nothing at 
>all for suppression to obtain absolute stability.

Is Mr. Rauch defending the design of what seems to be the only HF amp. on 
the market that uses no VHF suppressor whatsoever? (the AL-1500)
>
>Look carefully and you will find  PA's using tubes that provide the 
>best VHF operation (tubes with the highest VHF gain and performance) 
>are the most stable, while tubes with long thin leads and poor VHF 
>performance are the most unstable tube and require the highest Rp 
>suppressor for stability.
>
>None of this is all that mysterious.

It looks sorta mysterious to me.  For an 811A, he said more suppressor L 
and R are needed, which makes sense because doing so is known to decrease 
VHF-Rp, which undoubtedly decreases VHF gain.  .  .  Next, he tells us 
that more VHF-Rp means more stability.  .  Which way is up?
>
>73, Tom W8JI



Rich...

R. L. Measures, 805-386-3734, AG6K   


--
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