If its a defect why elevate the heater and not just replace the defective tube?
How to know? Does this leakage show in a tube test?
Or is there some other way to assess, like swapping tubes?
> why elevate the heater and not just replace the defective tube?
You may not want to replace the tube if a buck's worth (or less) of parts will make it satisfactory.
The excuse I heard was that in *factory* work this trick greatly cut-down the number of amps which failed final play-test and had to be re-worked (tube-juggled) by higher-paid workers.
would this imply that we should always elevate the heaters?
A little anecdote:
A boutique amp-maker was building an amp based on the
5F6-A Bassman, but wanted to use a cathode bypass cap smaller than the 250µF found on the original input stage in order to lose some of the mud. But he encountered a hum he couldn't resolve. He thought he ruled out the preamp tubes as the cause of the hum, because he tried 5, 10, 20 different tubes and the hum persisted. But tacking a 220µF cap across the cathode resistor killed the hum dead.
The problem was all 100+ tubes in the case-lots he got had the same heater-to-cathode leakage defect, so the build hummed unless an over-size cap was used on the input stage. I don't recall (this convo happened ~2005), but that builder may have used heater-elevation to avoid the problem in future amps while still being able to use the small cathode bypass cap to shave bass (this might have even been a switched voicing-option for the model). Further, the builder was buying tubes by the hundreds, and somewhat at the mercy of whatever his vendor shipped (gotta get those sold-amps out the door).
_______________________________________
This experience with that builder convinced me that the reason Fender used
12AY7 for the "low noise ... hum" that the manufacturer bragged about on the data sheet, and then went overboard with the Bassman's cathode bypass cap to be doubly sure to kill hum in the input stage (in a "belt & suspenders" sort of approach).
Where possible, the very-large bypass cap will do the job.
When the very-large bypass cap cannot be used, heater-elevation may be a solution.
Not in my world. Everything I have built (or worked on) has had the filaments referenced to chassis ground ... ***EXCEPT*** my November project from AX84. ... I was experimenting with a lot of the web voodoo. Years later when I built a very similar 6V6 Plexi I just referenced heaters to ground.
... I do have a cabinet full of NOS mil-spec tubes. 
I've got a couple-hundred 6BC4 tubes made by RCA. They weren't "a dime a dozen" but maybe only $1.20/dozen.
I used them in an amp build that has a switchable bypass cap for a bass roll-off. Wouldn't you know, about half of them hum when the small bypass cap is used. Enough so that the amp & cap-switch is a strong qualitative test to sort 6BC4s into "leakage" and "no-leakage" piles.
If the primary application for the tube didn't force the manufacturer to control some characteristic (like heater-to-cathode leakage) then they didn't focus effort on it (unless perhaps the manufacturing techniques to control the issue were already being applied elsewhere in their products).
Found that was true for microphonics in "computer tubes" like the
5965 (because the computer has no speaker, and the tube normally switches between cutoff & saturation). If anyone thinks, "that Mu of 47 might make this a good 12AY7 sub"... just
don't.
Topic: Ground loop purgatory with outboard reverb tank (Read 3571 times)