[Amps] Alpha 87 (not 87A)

[Jim Thomson]

Date: Mon, 6 Nov 2017 20:22:46 -0500
From: Ken Boasi <n2zn at rochester.rr.com>
To: amps at contesting.com
Subject: [Amps] Alpha 87 (not 87A)

<Hi guys,

<I was recently going through some back issues of NCJ from the late 80's. One of the ads that caught my attention was Alpha announcing their new Alpha 87 linear that was soon to come out. 

<The picture showed a manual tune amp that had the similar passband tuning bandswitch that the 78 and 374A had, but with the LED bar graphs of the 87. 

<I'm assuming this amp was never produced, since the 87A came out around this timeframe. 

<Has anyone ever seen one of these? My guess is that it never really existed. 

<Just curious (and no, I don't have one!)

<73, Ken N2ZN

##  My guess is.. it was never released.   The alpha 78 and  374A had that passband tuning scheme... + manual tuning.  If I remember correctly,  the passband tuning scheme  consisted of a mess of series L networks. 
While the passband tuning  worked well on the 78 + 374A,  I believe it was  dumped, since the harmonic suppression did not meet the  new FCC specs, which is like –42 db for a 1500 w out amp. The motor 
driven PI-L used in the 87-A would easily meet those specs.   Another way to make an amp broadbanded, or more broadbanded, is to use a manual tuned  PI-L.... but with a lot lower Q.   Its a trade off though,
lowering the loaded Q  reduces harmonic suppression, but the amp is a lot broader, once tuned.   On my hb amps,  I dont get wound up about harmonic suppression, its low on my priority list.   I avoid the PI-L
like the plague, then the  peak V across the load cap is way down, since its at a 50 ohm point, and not the typ 300+ ohm point.   Also having to  switch taps on a 2nd coil is a major pita.   The PI-L  would have
some merit if used on something like a 6M  mono band amp, or any other monoband amp, where neither of the 2  coils are tapped.  But even then, harmonics are not a major issue in my book, at least not for a 
single op station.  Monoband ants are not resonant on their 2nd harmonic, and my 40m yagi doesnt resonate on its 3rd harmonic, but higher,  like 24 mhz, besides, the 3rd harmonic is WAY down on any plane jane 
PI network. IMO,  30-36 db of 2nd harmonic attenuation is ample.
Ameritron achieves and meets the –42 db FCC spec..using a simple PI on each band.... except 160m... but their amps are all manual tune.   On 160M, Ameritron uses a PI-L... simply for the reason that a smaller load
cap can then be used, and or, less padding of a load cap required.  

Amps with a lower loaded Q, whether a PI  or a PI-L used, also means less circulating current through the tune + load caps, main tank coil,  2nd smaller tank coil, if a PI-L used, and also less current through the
poor bandswitch.   On the upper HF bands, like  12-10-6m, typ a L-PI is used,  or a L-PI-L, which  consists of a small, typ .5 to .6 uh heavy duty coil, inserted between plate block cap and   main  PI  or  PI-L
input.  The tube or tubes  anode to grounded grid C.... and the extra small coil, form a step down L network.  The plate load Z of the tube, or tubes, is greatly reduced,   reduced enough so that a conventional 
PI or PI-L  tank can then be built,  without the typ sky high loaded Q   typ seen on amps on the upper HF bands.  

Having said all that,  I think it was hank, who emailed me that harmonics could also be greatly reduced, simply by increasing the loading on an amp.  IE: increasing the loading means Decreasing the value of the load cap. 
Typ, you pulse tune the amp up for max power out.... then increase the loading slightly,  till power output drops 2%.   He found that if the loading was further increased, such that power out dropped  5 to 10%,
spurs and harmonics dropped  another 20 to 40 db.   He measured it all on his spectrum analyzer.  Point here is, the actual  2-3-4-5-6-7 th harmonic suppression was greatly influenced   by the loading control. 
Ok, so you tune your amp up to 1550-1650 pep out, then increase the loading, till po drops to 1500w.    You can then still achieve plenty of harmonic suppression, and still use a simple low Q,  PI net.    

Jim  VE7RF