AVS Forum banner

Aging Tube Sound - What Does it Sound Like?

7K views 42 replies 12 participants last post by  Gary McCoy 
#1 ·
I got into tube amplification. My ARC power amp is on 10 hours a day, so it burns thru 2,000 - 4,000 hours pretty quickly.
How do I know (apart from the time lapse) THAT THE TUBES NEED REPLACEMENT? Is it just the loss of power or aging tubes have a specific sound?
Thanks
 
#2 ·
#3 ·
That certainly helped, thanks, LOL
 
#5 ·
Don't move the amp when the tubes are hot, mechanical shock to the tubes when hot will shorten their life several-fold.
Let the tubes/filaments heat for at least 30 seconds before playing music.
Bias colder for your day-to-day listening volume.
The elements that can affect the sound is the aging of the cathode coating and loss of vacuum thus cathode deterioration, or the filament may just pop; the plate and grids do not actually deteriorate.

I'm a believer that if the tube works just keep it in, some tubes will last you a life time, yes even power tubes.
 
#6 ·
Don't move the amp when the tubes are hot, mechanical shock to the tubes when hot will shorten their life several-fold.
Let the tubes/filaments heat for at least 30 seconds before playing music.
Bias colder for your day-to-day listening volume.
The elements that can affect the sound is the aging of the cathode coating and loss of vacuum thus cathode deterioration, or the filament may just pop; the plate and grids do not actually deteriorate.

I'm a believer that if the tube works just keep it in, some tubes will last you a life time, yes even power tubes.

Agree. Some folks will leave power tubes on all of the time. I don't because I'm cheap, and you'll see this in your electric bill if you have enough tubes. I do turn them back on well in advance of a listening session.


I've replace maybe two tubes in all of my life so they can last a life time. When we restored and modded my dad's old Dynaco stuff from 1959 the Mullard EL-34s tested fine so they're still in there.
 
#7 ·
Don't the tubes just lose power over time? If the amp is self biasing you shouldn't note any significant difference. If not, bias manually ever 6 months. I just biased the tubes on my amp after 6 months and the change was nominal, maybe from noon to 12:05.
 
#8 · (Edited)
"Don't the tubes just lose power over time?"


I presumed they did, but I just haven't had issues.

My VTL 300s and 450s used 6550s originally, which I swapped for KT-90s when they came out because they were better sounding tubes. In those amps you had to manually set the bias, but I don't recall much change over time. I can't remember in the ARC Classic 150 whether I had to manually set bias or not. But man did those amps sound different on ESLs ..... ML CLS 1s and Soundlabs.
 
#10 ·
The book, "Getting the Most Out Of Vacuum Tubes", by Robert Tomer, Chapter 1 covers catastrophic failures like arcing, cracks and heater failures. Improper biasing is mentioned in the context of fixed vs adjustable.

Chapter 2, "Degenerative Failures" points out the things that reduce tube life. A big one is heat. Active cooling will definitely improve tube life. That's "active", not "passive" convection. Transfer of heat out of the tube through the envelope is the only real way to reduce internal heat.

Running filaments below rated voltage by as much as 20% can extend tube life, but has to be weighed against performance. Gas is a big one too, but not typically well understood by users. It's not a loss of vacuum, which would be catastrophic, not degenerative. Cathode depletion is actually pretty far down the list as a cause.

There can actually be damage to plates and grids from excessive heat/current. They are not immune.

Since biasing affects linearity as well as current, it does affect life, but more importantly, it affects distortion. Under biasing for day-to-day listening may not be the best choice, though deliberately raising distortion is part of that "warm tube sound". If I may editorialize....yuck.

In a class AB circuit with negative feedback it would take considerable loss of emissions for the amp to lose power. Measureable perhaps, not likely audible until tube performance is really poor. Distortion would be the indicator.

Everyone rolling tubes and re-biasing measures the resulting distortion, right?
 
#25 · (Edited)
I am all for science and unbiased hearing test, but after many years of experiences with guitar tube amps, I can say that tubes DO sound different depending on the circuitry they are part of.
In a linear Hi-Fi amp (I own a small Class A amp) the difference is practically zero, but in a guitar amp (buffers, filters, EQ stack, Phase Inverter) the difference is very much obvious.
In other words, if the circuit around the tube interacts with the tube in some way (miller capacitance, resistance plate, etc..), there WILL be a difference in sound, but if the tube is properly biased and working in the linear zone (hi-fi) the difference will be small.

Power tubes EL34, 6L6GC will sound practically the same if designed to work in the 20-20k and linear zone like in a hi-fi amp (in my experience).
But in a Guitar tube amp, different tubes make audible difference because the circuitry around the tube is designed to be FAR from linear.

In the name of peace, love and science; lets agree to disagree without fighting.

Everyone rolling tubes and re-biasing measures the resulting distortion, right?
Not me.
1. Pull old tubes,
2. Put new tubes.
3. Bias (cathode current over its resistor, plate dissipation, etc..), I like COLD biasing in guitar amps (50% dissipation in push-pull) for greater power and longer cooler life.
4. Rock the house!
 
#12 ·
IMO, Tube rolling: Swapping tubes and listening for a change without measuring anything, and without any means of comparing to anything. A completely sighted, subjective, biased, and nonsensical method of wasting time and money. You pay up for an exotic, expensive and "matched" set of tubes...yes, you will hear a difference, even if there is none. Might as well pour gasoline on $50 bills and light them up. At least there would be some visual effect.
 
  • Like
Reactions: Ratman
#33 · (Edited)
With my Audio Research LS26 preamp, I find that the tubes last 1000-1500 hours and then the sound quality starts to deteriorate.

I usually notice a loss of bass definition and less bass in general, and then a general grunginess of the overall sound will start to set in and get gradually worse.

I do not have a tube power amp any more, but when I had them I found that power tubes usually completely fail when they go bad; sudden total failure.
 
#34 · (Edited)
I had both a SP-9 and SP-14 and they had tubes only in the phone stage. I played with tubes there too so none of them got old enough to start to go out on me.

My tube failures as best I recall were the new, at that time, KT90s in my VTL 300s. They were known to have a bunch of quality issues. Two of them failed immediately on first power up. When you got a set that worked, I thought they were a noticeable improvement over the 6550s. As with most things audio YMMV.

Still though what a stark difference between the VTLs and the ARC Classic 150s. I liked both but in very different ways. I swapped back and forth and used them interchangeably on the Maggies, Soundlabs and Martin Logans. Funny though that I didn't like the ARCs with the Maggies as well as the MLs and Soundlabs in spite of the typical synergy that you generally had between ARC products and Maggies. Maggies weren't ribbons in those days and mask some of the differences that you'd hear with the ELSs. Still though, there was sort of a grain structure that you'd hear with both the ARCs coupled with the bit of midrange grain you'd get with the Maggies and the result wasn't as pleasant as it should have been. But the ARCs were great on both the Soundlabs and MLs and the VTLs noticeably not as good in many ways. That backwards of the conventional wisdom at the time.
 
#35 ·
Thanks, guys. Unfortunately, the ARC decided to go belly up on me yesterday. A puff of smoke, one of the tubes lit up like a 100W light bulb. It was not even playing, I was listening thru headphones. I turned it off immediately. I thought a tube was old and needs replacing. I have 6 spare tubes. I checked bias level, three were out of wack. Replaced them. Still the same problem. So, it is the amp part that burned, not the tubes.
Called Audio Research about fixing the VS110. They said "ship it to us, we will fix it, $120 an hour plus parts, two hours minimum". Didnt give me a ballpark figure. Called a few local tube shops. One said most likely the amp had a tube arc on a 6550C, and took out a small portion of the board in addition to the two resistors that blow when a tube arcs. A common problem, apparently. He said ARC will charge on average $810 plus shipping both ways. He quoted $250. He has been both fixing and selling amps and speakers for 30 years, ARC's including. Should I?
 
#43 ·
In answer to your original question, when the tubes age they weaken, the internal resistence of the tube increases with an aging cathode coating, and gain declines. If you reach an average volume setting that is more than 20% above the setting when the tubes were new, the gain has degraded and distortion is increasing as you turn up the volume. The volume control is not typically a linear taper potentiometer, this is to some degree subjective.

Your ears are not good enough. I play a reference 1000Hz test tone from an audio oscillator and measure the speaker output with a old Radio Shack sound meter. This requires keeping records in a notebook, as I found out that sticky notes perish even when stuck to the amplifier bottom cover. The note has a date and a drawing of the volume setting that achieved the 0db reference level in the main listening position.

I never "roll" tubes, they are too expensive. I use them until they weaken, then replace them. Some brands last longer than others. I've had the longest service from Russian military tubes.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top