[Amps] Alpha PA-77 (not D or S) Questions

[k6xx at juno.com]

Amplifier Experts:

Perhaps you could lend me a bit of your experience? I acquired an old Alpha PA-77 this past weekend. This is the 1973 version of the 77: it uses a single 8877 and is a bit smaller than the newer 77D or 77S. The bad news is that it arrived in worse shape than I hoped. It has been modified several times, with several things removed (such as the QSK feature). The good news is that 160m and BYPASS has been added--evidently by the factory.

I have a few problems with it, and would like some advice so I can fix it quickly without killing myself (preferrably). When I first powered it up, I tuned up on 20m. It is great there: full 1500W is easily made.

1. 160m operation. While the amp has been enabled on 160m, it doesn't like the band much. The two times I hit 1200W of output, I blew the plate overcurrent sense relay shunt resistor (the resistor across the plate current overload relay). The second time, I also damaged the RFC in series with the plate choke. Obviously these components though there was way too much plate current. But the front panel meter didn't indicate this, and it is directly in series with them both.

I replaced the blown-open 5-watt sense resistor with a 10W, and then a 20W unit, and replaced the small-winding 6uH encapsulated RFC with a huskier 2A, 15uH version. I can now run the amp at 1100W without issue, and the efficiency was measured at 59% to 63%, depending upon frequency. Of course, I'd like to run a few hundred more watts, but figured I'd get your input on this amp's suitability for 1500W on 160m before I pushed it again. FWIW, the 77D uses a 15uH inductor in this position. I do not know if ETO used the same plate choke in the 77 as in the 77D/S amps.

2. Efficiency on 15m is poor, only about 43% at Alpha's test frequency of 21.3MHz. It rises to 51% at the bottom of the band, but this is still unsatisfactory. (For a comparison, the efficiency on 10m is 55.4%, and it is 60% to 65% on the other bands). Grid current is OK for the power output, just the efficiency is low.

80m Issues: 
3. Amplifier draws a small amount of grid current without drive applied. Zero plate current (this amp 'features' Electronic Biasing), but several mA of Ig. It doesn't draw any Ig (without drive) on any other band.

Other than that, it seems to work great at Alpha's test frequency of 3.80MHz, with the TUNE and LOAD capacitor settings within one count of the book value. Efficiency is 60%.

4. However, at 3.505MHz, LOAD is at maximum capacitance and the amp really wants more. If the LOAD capacitance setting at 3.8MHz was off a bit, I'd have guessed that the fixed capacitance added into the Pi network on 80m simply drifted low over the years. But LOAD almost exactly matches the book at 3.8MHz, either tuned for 1000W DC Input (at B+ = 2.4kV, no load) or 1500W RF output (B+ = 3700V) ... virtually the identical setting at both power levels, by the way. 

Efficiency at 1500W output on 3505kHz was measured at 66%, even though the loading is not optimum (I admit this measurement looks questionable!)

5. Should I assume that this amp, with its simple unswitched plate choke, will not operate on 12m?

I appreciate any comments or suggestions. TU.

73 de Bob, K6XX
_____________________________________________________________
Keep your home running smoothly with great home automation solutions! Click now!
http://thirdpartyoffers.juno.com/TGL2111/fc/Ioyw6iifZqi03IuYi3taamKjpamJSxvtY0GyltRaRuiOwInBRqvKxs/?count=1234567890