Tuning is a reiterative process, and assuming the amp's components are all intact, it "sounds" to me like you're simply underloading it.

I've had an AL-80B for over 12 years and for it to run full power, the Ip is always well over 400mA; it can get up to 600mA, and for intermittent service, that's fine.  I keep Ig down to about 150mA max using the ALC.

But the first peak you hit while tuning is never the "final" peak; after the first peak in output is achieved (using both controls), increase LOAD more (clockwise) and repeak PLATE.  Then increase LOAD more, again, and repeak PLATE, etc.

The "final" adjustment should achieve 150mA Ig and at least 500mA Ip, and when both controls are re-peaked for MAX, turn the LOAD control slightly more clockwise to reduce Ig about 10%, and then leave it that way.

Regardless of 3-500Z tube ratings, I've checked linearity and IM3, 5, 7 and 9 with the amp loaded to almost 600mA Ip (takes about 75-85 drive power, usually) and Ig = 150mA, and it's still linear (delta input vs. delta output remains the same as at lower power levels, with about 10-11 dB uniform gain; it starts to flatten out beginning at about 800W carrier power output, and is in compression at 1kW output) and worst case IM3 is about -33 dB.  IM5, 7 and 9 are lower, in the -39 to -45 dB range.

Might be something else, but first guess is you're just not loading it heavily enough.

Is the amp generating any measurable RF output power when it's doing this, or is the "hash" just something you observe on a receiver?

That's an oldie with tube rectifiers and such, but they were good amps and well made.

After a thorough cleaning and inspection, I'd request the assistance of a local ham who is highly experienced with amplifiers (best would be someone who has built amplifiers himself) to help you bring it to life.

If the 3-400Zs were stored unused for a long time, pretty good chance they'll be "bad;" although we get lucky sometimes, and maybe they won't be.  3-400Zs haven't been made in over 40 years, but 3-500Zs can be used with small modifications.

That model had a HV choke and oil-filled capacitors that can and sometimes do go bad with age; a failure in operation can be quite exciting, which is why I recommended you do the first power-up with some assistance if possible.

Watch the last 15 mins or so of the film Independence Day (1996, with Will Smith, Jeff Goldblum, et al).

What is used to "save the world" from the alien invaders who wish to destroy the planet?

On the Shure 450 cable:  Tie the black, white and shield wires together to form a common ground.

That "common ground" (B/W/shield) go to Pin 2 on the Ten Tec.

The green wire from the Shure 450 goes to pin 1 on the Ten Tec.

The red wire from the Shure 450 goes to pin 3 on the Ten Tec.

Pin 4 on the Ten Tec would not be used.

It also depends what the house siding is made of.

Stucco, cement, bricks...very bad news.  Obviously, aluminum siding would be terrible.

Just plain dry wood (like cedar shakes) isn't ideal, but not nearly as bad (unless there is foil-backed insulation in the walls).

Also, a "stapled to the wall" antenna is much more likely to both generate and receive interference from electrical and electronic appliances inside the house.  Almost a disaster waiting to happen.

Just I guess, but I suspect the Q52 might be better in this regard because it's bandswitched and only tuned to one band at a time using selective circuits within the two "matchbox" units.

However a lot of this may be closely related to how high above ground you install either antenna, and also what the noise source(s) really is(are).  There's not a lot of atmospheric noise on 20 through 10 meters unless there are lightning storms in the area and these bands can be pretty "quiet" (my noise level at home using an 8 element LPDA is usually in the S1 area on these bands), so it's been my experience that most "high noise level" issues are locally created and man-made.  Either eliminating those sources or moving the antenna farther away from them helps a lot!

Sure, that's just a 1/2-wavelength 40m center-fed dipole.

If you move the feedpoint away from the center, you shift the feedpoint impedance that can be a more reasonable compromise on more bands.  A common approach with a 65' doublet is to feed it 21-1/2' from one end (making the "other" side of the doublet 43' long).  At the feedpoint, do not connect coax directly, but use a 6:1 (or even 4:1 can work) balun as a matching transformer.

Some easy-to-understand theory behind this is here: http://www.w8ji.com/windom_off_center_fed.htm

A power point presentation that's kind of fun is here: http://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=3&sqi=2&ved=0CEMQFjAC&url=http%3A%2F%2Fwww.yccc.org%2FArticles%2FAntennas%2FN1IW%2FOCFD_basic.ppt&ei=s7NtUd6TG8jXigL16oD4CA&usg=AFQjCNHfWn-0DE-cK-bztlDbeKcro0ArDw&sig2=Epu261hkS_6pZnJk0FtCtw&bvm=bv.45218183,d.cGE

More ideas here: http://www.hcra.org/2012/09/06/home-made-off-center-fed-dipole/

More fun reading here: http://w0hc.com/ocf-dipole/

It's a very old idea and well used in amateur circles for a long time.

I wouldn't do what you suggest.  If you use a Variac to power the SB-220, when you have the voltage setting below normal, you're decreasing not only the plate voltage but also the filament voltage.  Trying to draw cathode current (and anode current) with filament run below normal voltage can hurt the tubes.

Salt water dummy loads are tricky.  Why not use a normal dummy load (carbon/ceramic resistor) in a one-gallon container of mineral oil?  A 100W Globar resistor immersed in mineral oil will handle a kilowatt for a minute or two.  To reduce the stress on the load, if you have coax laying around, you can connect the load at the end of a lot of coaxial cable to add some loss and dissipate half the power in the cable, so the load only needs to dissipate the other half.

 

You didn't say "how" it's actually fed.  How is the RG8 coax connected to the 65' wire, and where is it connected?

Eimac hasn't built a 3-500Z in decades.  The tubes used in the AL-80B aren't Eimacs, they're 3-500ZGs (graphite anode) built elsewhere.

Philips (now Richardson) specs the Amperex 3-500ZG quite differently.  If you refer to the Amperex data sheet, (lower section of page 3) they list OPERATING CONDITIONS as 3000Vdc Ep, and 770mA peak anode current, for 1420W peak output power (single tube): http://www.mhzelectronics.com/ebay/manuals/amperex_3-500z_tube_datasheet.pdf

I have no idea why people still refer to Eimac data sheets that are 40 years old, when Eimac doesn't manufacture this tube.

For contesting I use computer logging/duping so of course the computer also sends the code (although it can't copy any).  I set it for 40-45 wpm almost always for this purpose, and rely on "wet matter" for decoding.

But for casual QSOs, which I make almost every day, the "norm" seems to be 22-25 wpm.  If someone's going slower than this, what I usually do is keep my dit/dah speed at 22-25 wpm but just increase spacing between characters and words, doing that manually with the paddles.  So each character might be 22 wpm, but the actual "code speed" might be 10 or 13 or whatever the other guy's doing.  This doesn't seem to "lose" anybody, and it's easier for me.

If the other guy is obviously getting "solid copy" at my adjusted 13 wpm, I'll space the characters and words closer together to ramp it up to 16 or 18 and see what happens.  It's obvious is somebody's really copying well or not.

I notice that most who start each transmission with "R R R SOLID COPY" probably aren't really.   I never send anything like that, and copy everyone 100% unless they take a big fade or QRM/noise burst takes them out.  I think it should be understood that we copy everything unless we ask for a repeat.

When all is optimized, with my AL-80B Ip is 150mA maximum and anywhere from 450 to 600mA Ip, depending on the band and the load.

I use the ALC connection to limit Ig to 150mA, so even operator errors won't let it exceed that.

Ip can run 600mA and even above that for short duty-cycle modes like SSB and CW.  Don't try it on digital modes or FM.  For AM work, I limit carrier power to 150W (600W PEP output) and average plate current is pretty low, around 200mA but peak Ip can run easily 400-500mA.

For "most" antenna towers, I use NMFC Class 85, which is durable goods weighing 12 to 13.5 lbs/cubic foot.

They do drift out of calibration over time, and the Bird elements are calibrated at 25C; when used at much higher or lower temperatures, they're all off and that's to be expected since they're calibrated at 25C.

Calibration isn't difficult; I use a heat gun (for just a few seconds) to warm up the adhesive to remove the label and an X-Acto knife to lift one edge, and it comes off pretty easily.  Clean off the remaining adhesive, and then "hot melt" works well to re-attach.  Done this dozens of times.