![]() ![]() |
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
i've been using using 2 GE 6350 as drivers and
2 Miniwatt ecc82/12au7 as outputs in my VTL 2.5
preamp for about 10 months 24/7, i.e. about 7,000 hours;
VTL recommends to replace those tubes every 5,000 hours;
i'm not aware of any signs of death, but am curious about
which symptoms are they supposed to show first? TIAbest, paco
Follow Ups:
My career was in the Hollywood broadcast and recording studios, beginning in the '50s. "In the early days" there were hundreds of tubes in most installations.Of course Hickock testers were a necessity. We routinely checked each amplifying tube for mutual conductance (the main parameter relating to performance) every few months, and most pro equipment is left on 24/7.
The main thing that determines tube life is how hard it is used. That is, how much plate dissipation is used compared with the rated value of the tube. In general, that translates to a tube that runs very hot will have a short life. What poops out is cathode emission; the tube just gets "soft." Usually no audible effect if they are caught before mutual conductance falls to 1/2 normal value. Amperex and some Telefunkens used to worry me because the heaters were SO bright it looked unhealthy. But they seemed to last as well as others.
Once in a broadcast station they had ordered a large stock of "6L6," not specifying the glass version, 6L6GC, for the last version. Well, a "6L6" is a metal version, used in RF applications where controlling RF radiation is important. Problem is, the metal tubes didn't get rid of their heat as efficiently as the glass, and those 6L6's had about half the life of the glass ones in audio amplfier ouputs. The devil is in the details.
I have been talking about normal lifespan. Then there were the abnormal problems - microphonics (preamp tubes only), gas in power tubes (that blue glow), sudden shorts etc. Some are not going to like hearing this, but it was my experience that RCA tubes usually lived out their lives quietly, eventually just pooping out when cathode emission fell too low. On the other hand, GE tubes usually went out with a bang of some kind. Rarely did I find a GE tube with a long, peaceful life.
I don't know how you can check tubes for emission other than with a tester, except by measuring the DC voltage across the cathode resistor in self-biased tubes. Some pro equipment had that feature built-in.
I turn off tube equipment if it is not going to be used again for several hours. But I don't leave it on 24/7; a tube is using up cathode coating (which is the source of the emission) every minute it is operating. The strain (which is only on the heaters) of firing them up has to be balanced against "emission time."
I hope this is useful information for those who didn't grow up with tubes by the hundreds in pro installations.
I'm suffering from many of those symptoms with my preamp & yes I did run 24x7 in the past. If I don't see a complete restoration of sound with new valves - what else after the power caps should I look at?
Typically the first signs of tube death (no matter how many hours are on them) are when you read about some rare NOS tube that might be fractionally better than what you are using.....dont ask me how I know :)
![]()
That has happened to me. I thought my RFT 6BQ5s were the best after four other sets of tubes. Along came a rather 'run of mill' GE 6BQ5 grey oval plates and they handily unseated the RFTs- go figure!
![]()
Some tubes will start reproducing crackling sounds through your speakers as they age; this is a sure sign that your tubes need changing. More subtle signs is loss of dynamic range, with closed in highs and truncated bass. This is harder to notice since it's a slow progression and people usually get used to it; when they change their tubes out, however, they'll immeidately realize what they've been missing.
![]()
Tubes with a lot of hours become lifeless & dull sounding. 5000 hours are not a lot of time for some tubes as many variables exist.
![]()
... gut starts to tell me 'something has changed'; it's usually a tube failing. A tube becoming a little dimmer isn't a very useful predictor; there could be several changes in the circuit causing that.
![]()
Some recommend leaving gear on 24/7. I don't know but to me it really seems like a bad idea. I've PERSONALLY had a Sansui blow up like a firecracker and fizzly and smoke until I could pull the plug. My tube gear gets hotter than I would feel comfortable to leave. Something about leaving a piece of electronic gear with 550 volts kicking around doesn't seem smart to leave alone. To me it's like leaving the iron on and going out. Sure it's UL approved and how many Irons catch fire. But do you leave it on? NOJust turn on the unit 10 minutes before you go listen. That's more than enough and save the tubes, capacitors and maybe your house.
Leaving a tubed power amp on 24/7 is not reasonable. the tubes will heat up to a steady temp in 5-10 minutes. there is no reason that i know of to leave a tubed amp on all the time unless he means that it is on standby. If just the heaters are on, that should prolong the life of his tubes, i understand that it is the on/off sequence that strips the cathode.
![]()
I've had a few large output tubes flare up, which obviously says its time to change them. Sometimes before they fail, they'll cause observable vibrations in either the tube or the transformers.
![]()
This is a question, not a reply, but...Why don't you turn the pre off when it's not in use? Maybe you might even leave it on while you are awake and home, but why let it wear out the tubes all night long? I'll bet you could easily double tube life.
BTW, they may not be pricey, but 6350s aren't real easy to come by anymore.
The power supply caps have a limited life too, as do many of the components in the pre.
My advice - re-think the "24/7 on" approach.
One last thought - the tube performance often has dropped off considerably before you can hear it. It often happens gradually which makes it very hard to detect any changes in the sound.
![]()
This post is made possible by the generous support of people like you and our sponsors: