In recent years, fueled by social media and a certain tendency to mythologize tubes as if they were sacred relics, a rather dangerous habit has spread: replacing tubes “by feeling.” This article was born after seeing a video on Facebook of an “expert” teaching how to tell if a tube was good or to be discarded… by tapping it with fingers, like checking watermelons. Yes, you read that right.

Over the years I have witnessed the most absurd situations: people who replace tubes “because some time has passed,” crafty gurus who charge thousands of euros to change perfectly good tubes and resell them as NOS, or improvised technicians who replace the tubes of an old radio while leaving all the leaky capacitors in place… It’s time to set the record straight!
Tubes: they’re not fruit, don’t “knock” them
Let’s start with a basic concept: tubes do not wear out with time. They do not lose performance simply by being turned off or unused.
A tube device—whether it’s a vintage radio, a hi-fi amplifier, or a guitar amp—contains many other parts subject to aging: capacitors, sockets, oxidized contacts, tuned circuits, wire insulation that becomes brittle. All these parts deteriorate over time. In vintage radios, based on my experience, out of ten devices only one or two tubes actually need replacement, and only if truly necessary.
Visible signs of a tube to discard
There are unmistakable signs that a tube must be replaced, even without instruments…
White getter:
indicates air ingress; the tube is completely non-functional.
Partially whitened getter:
a dangerous situation, the tube might still work but risks internal arcing that can damage transformers!!!
Brown spots or internal smudges (soot):
sign that the tube has many hours of use, but it doesn’t always mean it must be discarded. It should be measured.
(in the photo an extremely sooted EL42; rimlock tubes are particularly prone to this issue).
Broken glass: nothing more needs to be said…
Microphonics: when a tube becomes a microphone
Another issue is microphonics, the tube’s tendency to vibrate and generate noises or whistles if tapped or exposed to environmental vibrations. All tubes are naturally slightly microphonic, but excessive microphonics can cause problems, especially in phono preamps or very sensitive stages. It can happen that a tube is too microphonic for a preamp but perfectly usable as a phase splitter in a push-pull circuit.
Moreover, many audiophiles today buy tubes originally intended for computer use (for example the 6350), which were not designed to minimize microphonics since they served as logic switches. That explains certain mysterious noises…
When is replacement really needed?
Beyond visible signs, the real judgment is made with instruments. You need a serious tube tester (not the SRE “Scuola Radio Elettra” style, totally unreliable even if restored).
Even better, if available, a computerized curve tracer: it allows in-depth analysis of the tube, observing emission, linearity, leakage, and making precise selections essential in hi-fi applications.

Real comparison: new tube vs worn-out tube
For those who still think a tube can be evaluated “by eye” or “by ear,” here is a concrete example from my lab. In the graphs below you can see the comparison between two tubes of the same type and origin:
| tube new or otherwise perfectly efficient | tube worn out, with greatly reduced emission |
![]() |
![]() |
In the first graph we see a family of well-distributed, regular curves with stable gain and anode current that increases predictably with grid voltage. This indicates a tube in excellent operating condition. In the second graph, however, the curves are “flat,” and the anode current grows little even with an open grid, indicating that the tube has lost emission. In an audio circuit this tube could cause distortion, loss of gain, a muffled sound, or even fail to work at all.
Case study: two ECC82 compared, what a tube tester won’t tell you
Another interesting example involves two ECC82 tubes, one brand new and one used, apparently still in good condition. On a classic emission-type tube tester (or even a transconductance tester like the TV-7), the used tube might appear “within spec” or at least acceptable, with values still within the manufacturer’s limits. But the curve tracer tells quite another story.
![]() |
![]() |
In the first graph (left, new tube) the curves of the two triodes are regular, parallel, well overlapping, and show stable gain across the entire bias range. This is what one expects from a tube selected for hi-fi use.
In the second graph, relating to the used tube:
- The curves start out well but tend to drop in the upper part
- The two triode sections are no longer well matched, with a clear divergence in the curves
This type of degradation is not detected by traditional instruments, which read only a few static parameters. In a general-purpose application (e.g., an instrument amplifier or small circuit), this tube might still work. But in hi-fi, especially in linear stages, driver circuits, or preamplifiers, this degradation can lead to audible distortion, channel imbalance, loss of detail and dynamics. This is why in my restorations and selections I don’t settle for a “passed test” on a TV-7 or similar: an ECC82 may seem “good,” but in a quality system… the difference is clearly audible.
Only with a serious instrument like a curve tracer is it possible to clearly see these differences.
“I think it’s time to change them…”
This phrase should be banned by law. There is no “I think” when it comes to tubes. If you don’t have an instrument to measure them, you cannot judge. And if a technician proposes replacing all the tubes without giving you an instrument-based report or returning the replaced ones, be wary. Always ask to get your old tubes back. It is not uncommon for them to be resold (perhaps to someone else who will pay a high price). In this article I tell the story of a client whose valuable NOS tubes were stolen to be replaced with current production ones.
Tube radios and automatic gain circuits
Almost all vintage radios have automatic gain control circuits (AGC), which compensate for the varying efficiency of the tubes. In practice, even if the tube is only 60% efficient or less, the circuit compensates and you won’t notice significant differences in listening.
The myth that “changing tubes improves the sound” is often due to the fact that in a misaligned circuit or with worn components, any change is perceived as an improvement. But it is just a placebo effect masking more serious problems.
Getter on the side? No, it’s not burnt!
A very common misunderstanding concerns the position of the getter, that shiny metallic deposit (usually silver, mirror-like, or lead gray) inside the tube. In many tubes the getter is positioned at the top, and everyone has learned to recognize it. But in many others, especially certain miniature types or tubes produced in Europe in the ’60s–’70s, the getter is located on the side, sometimes on both sides, and can form wider, asymmetrical shiny “tongues.”

I often hear comments like: “This tube is burnt, look at that strange spot on the side!” No, it’s not burnt!!! What you are seeing is the getter deposited in a different area, perfectly normal and intended by the manufacturer. It has nothing to do with damage or burning.
The getter serves to absorb residual gases inside the tube, and its appearance can vary greatly depending on the type of getter used (flash or sputtering), the production method, and the point where it was “fired” during evacuation. So: do not judge a tube as “worn out” just because the getter isn’t where you expect it. The only case for concern is when the getter turns white, a sure sign that the tube has leaked air.
Further practical tips
- Don’t waste money on complete tube sets, except when they are physically missing or visibly damaged.
- Don’t trust those who propose total replacements without measurements.
- Don’t throw away old tubes, keep them for possible future checks.
- Be wary of those who measure filament resistance with a tester and talk about “matched tubes”, a ridiculous practice unfortunately still widespread.
Conclusion
Tubes are not candies to be changed every so often. They are robust; NOS ones often outlive their users. An original, well-preserved tube, properly measured, has priceless technical and historical value. Rely only
on those who really know how to measure them, always ask to see the instrument data, and have your old tubes returned. Only in this way will you preserve the value (and the sound) of your equipment, avoiding enriching those who prey on others’ ignorance.
. . .
. . .
. . .











