[jamato8]

Quote:

Originally Posted by Ragonix /img/forum/go_quote.gif Does anybody ever see a blue spark inside a tube when you turn on your amp? What could this mean?

A blue spark can happen in rectifiers when there is too much current draw for the tube. The bright color at the inner base of the tube was characteristic of some Mullards and a few Yugo tube that glowed on turn on. For rectifiers, there will be a much shortened tube life for the others, as mentioned above, it is normal and very different.

 

[Ragonix]

Quote:

Originally Posted by jamato8 /img/forum/go_quote.gif A blue spark can happen in rectifiers when there is too much current draw for the tube. The bright color at the inner base of the tube was characteristic of some Mullards and a few Yugo tube that glowed on turn on. For rectifiers, there will be a much shortened tube life for the others, as mentioned above, it is normal and very different.

So, for rectifiers it's not normal and will much shorten the life of the tube?

 

[jamato8]

Yes, very much so. It is Never normal for a tube rectifier to do this. It may be too large a capacitor for the first cap or just too much current draw for the rectifier.

 

[Eisenhower]

Quote:

Originally Posted by manofmathematics /img/forum/go_quote.gif Funny. I've never lost a tube leaving it on all day, or even weeks at a time. I have, however, lost a couple during power on before. Not really speculation, just experience. Heck, come to think of it, the majority of my light bulbs seem to go when turning the light on too. Now, I wonder why that is? But what do I know?

And I bet you lost the tubes during power up after they had seen prolonged use. The failure might have occurred during thermionic shock, but the overall weakening of the filament is from exposure from heat.

Quote:

Originally Posted by atbglenn I have to disagree. 9 times out of 10, a light bulb will burn out when powered up. When a tube is cold and power is applied, the internal parts expand due to the heat rise. That has to be stressful and probably is a factor on tube life. I think it would be better to leave a tube amp on for the day instead of turning on and off several times. The heating and cooling can't be all that good. Just my 2 cents

Again, its the prolonged exposure to heat that weakens the filament, not the process of heating up / cooling down. If a rusty bridge collapses when a car passes over it, is it the car that caused the bridge to break, or the oxidation? Which one is of concern?

 

[Eisenhower]

Quote:

Originally Posted by Uncle Erik /img/forum/go_quote.gif Light bulbs get AC right off the house line. Tube filaments are usually fed off a 6.3V or 12.6V transformer and power supply. Depending on your amp, you might have a soft start from a special circuit or a slow startup thanks to tube rectifiers. You might also have a DC supply for your tubes, too, and sometimes the supplies are regulated. Turning on a tube amp isn't the same as turning on a light bulb.

For the purposes of discussing heated filaments, they are exaclty the same. Most filaments are AC heated, and most don't have a ramp up circuit. Even if they were DC powered and had a ramp up circuit, it wouldn't matter, its still ohmic heating, except with less noise and less thermionic shock (which would prove my point even more).

Filament/vacuum failure aside, prolonged use leads to cathode poisoning and evaporation of the emissive layer on the cathode, lowering the work function of the cathode. This plainly means tubes lose the ability to amplify the more they're used, which to most means they would sound WORSE.

 

[cooperpwc]

One thing that I have noticed is that a couple of my driver tubes light up like Christmas trees when I turn my EMP on. They then settle down. That observation may not translate into anything tangible wear-wise but it makes sense to me that constant on and off is a bad idea. That also accords with the instructions that come with the EMP. I will trust Stefan Brocksieper on this; he sure designed a good amp.

Quote:

Originally Posted by Skylab /img/forum/go_quote.gif turning tubes on and off repeatedly is far more likely to lead to premature death than turning them on in the morning on a day you plan to listen, and then turning off at the end of the day.

I agree with this as above. In my situation, I listen to my amp for an hour or so in the morning then I go to work for about ten hours before returning to listen in the evening. So I am turning it off and then back on after work. I am thinking that two on cycles a day is better than the extra wasted 10 hours of burn and also having the amp and tubes running every day for 17 hours straight. Any thoughts?

 

[jamato8]

Quote:

Originally Posted by cooperpwc /img/forum/go_quote.gif One thing that I have noticed is that a couple of my driver tubes light up like Christmas trees when I turn my EMP on. They then settle down. That observation may not translate into anything tangible wear-wise but it makes sense to me that constant on and off is a bad idea. That also accords with the instructions that come with the EMP. I will trust Stefan Brocksieper on this; he sure designed a good amp. I agree with this as above. In my situation, I listen to my amp for an hour or so in the morning then I go to work for about ten hours before returning to listen in the evening. So I am turning it off and then back on after work. I am thinking that two on cycles a day is better than the extra wasted 10 hours of burn and also having the amp and tubes running every day for 17 hours straight. Any thoughts?

Well I wouldn't leave it on all day either. Not only tube where but if a tube does go bad and your not around it can be hard on the amp, to say the least.

 

[Skylab]

Yes, I also agree that it makes no sense to leave an amp on for 10 hours while you do not use it. It's a waste of electricity if nothing else. I also never leave tube amps on if I am not home.

 

[xxbaker]

i think everyone is right in a sense. if you leave it on too long it does cause stress to the filament/tube but also turning it on and off has a higher stress/time ratio. so its really a matter of how big the gap in between uses is.