Tube matching? How anal should one be? Can you hear the difference (outside of extreme circumstances)?
- Thread starter caesar
- Start date
All IME/IMO. Bear in mind I do not have any tube gear in my current systems but have run many tube circuits over the years including several of my own design.
Tube designs tend to be single-ended with relatively low feedback. That makes them more sensitive to matched devices than typical SS circuits. Matching of both sides within a dual tube as well as tube-to-tube can be important to provide the same gain and frequency response from each channel. Matching power tubes in a push-pull design can also reduce distortion.
The tolerable difference depends upon the circuit and how sensitive you are to mismatch in gain (mainly, again IME/IMO). I usually strove for 1% matching but that was not always possible.
With new tubes, I checked matching after a few minutes to check for gross problems, then usually again after they had been running a few hours, and again around the 50~100 hour mark. That meant basically when I plugged them and and they had warmed up, at the end of a day or two of listening, and again after about a month. After that it depends but roughly every six months if using them daily. That said once the tubes had stabilized (which IME happened after 10~100 hours) they would often run for a 1000+ hours (power tubes) to 10,000+ hours (low-level tubes) before nearing EOL. At that point they would start to vary significantly but other things would indicate they were going bad as well (lower gain, higher noise, etc.) My experience is outdated at this point, but 1-5 years for a pair of power tubes and 5-10 years for a preamp/low-level tubes was not out of line, depending upon the design. Some designs run the tubes hotter than normal so they fail more quickly; others run them under-biased and they will last longer. I tended to use DC filament bias and there were claims that shortened their life; I did not observe that but my sample size was small. Running higher B+ voltage and greater current would drastically reduce their life (as some audiophiles and quite a few CB radio fans discovered).
FWIWFM - Don
Tube designs tend to be single-ended with relatively low feedback. That makes them more sensitive to matched devices than typical SS circuits. Matching of both sides within a dual tube as well as tube-to-tube can be important to provide the same gain and frequency response from each channel. Matching power tubes in a push-pull design can also reduce distortion.
The tolerable difference depends upon the circuit and how sensitive you are to mismatch in gain (mainly, again IME/IMO). I usually strove for 1% matching but that was not always possible.
With new tubes, I checked matching after a few minutes to check for gross problems, then usually again after they had been running a few hours, and again around the 50~100 hour mark. That meant basically when I plugged them and and they had warmed up, at the end of a day or two of listening, and again after about a month. After that it depends but roughly every six months if using them daily. That said once the tubes had stabilized (which IME happened after 10~100 hours) they would often run for a 1000+ hours (power tubes) to 10,000+ hours (low-level tubes) before nearing EOL. At that point they would start to vary significantly but other things would indicate they were going bad as well (lower gain, higher noise, etc.) My experience is outdated at this point, but 1-5 years for a pair of power tubes and 5-10 years for a preamp/low-level tubes was not out of line, depending upon the design. Some designs run the tubes hotter than normal so they fail more quickly; others run them under-biased and they will last longer. I tended to use DC filament bias and there were claims that shortened their life; I did not observe that but my sample size was small. Running higher B+ voltage and greater current would drastically reduce their life (as some audiophiles and quite a few CB radio fans discovered).
FWIWFM - Don
My pleasure!
In a line stage I would just buy another tube. In an output stage I would get a matched pair as matching is more critical for low distortion, and like headlights, if one went out the next is probably on its way.
One caution is that most generic tube testers do not load the tubes anywhere near what the actual circuit does so it may be hard to see how well they are really matched, again and especially power tubes. Back in the day I built a custom tube tester for output tubes (audio and RF).
Finally, I do not know how good tubes are these days, but I found a significant variance in performance (gain, noise, etc.) even in low-level tubes. That is one thing that drove me away from tubes; I was having to buy two or three to find one good one. There has been something of a resurgence so hopefully that is not the cse today!
HTH - Don
In a line stage I would just buy another tube. In an output stage I would get a matched pair as matching is more critical for low distortion, and like headlights, if one went out the next is probably on its way.
One caution is that most generic tube testers do not load the tubes anywhere near what the actual circuit does so it may be hard to see how well they are really matched, again and especially power tubes. Back in the day I built a custom tube tester for output tubes (audio and RF).
Finally, I do not know how good tubes are these days, but I found a significant variance in performance (gain, noise, etc.) even in low-level tubes. That is one thing that drove me away from tubes; I was having to buy two or three to find one good one. There has been something of a resurgence so hopefully that is not the cse today!
HTH - Don
Last edited:
Without a particular context the question is too ill-defined to warrant a serious answer.
Power tubes? Push pull or parallel SE? Channel to channel matching? Negative feedback or not?
The answer may vary between
- no matching needed at all for cathode follower stages
- matching practically unimportant in low level stages with deep NFB
- matching will reduce distortion in PP stages
- matching will improve interchannel balance and equalise distortion for low feedback preamps or input stages of power amps
- matching will improve distortion and max power for parallel output stages
- matching may be vital for the health of both amp and speakers in OTLs
Power tubes? Push pull or parallel SE? Channel to channel matching? Negative feedback or not?
The answer may vary between
- no matching needed at all for cathode follower stages
- matching practically unimportant in low level stages with deep NFB
- matching will reduce distortion in PP stages
- matching will improve interchannel balance and equalise distortion for low feedback preamps or input stages of power amps
- matching will improve distortion and max power for parallel output stages
- matching may be vital for the health of both amp and speakers in OTLs
Matching does not mean just having same measurement. The construction of the tubes are sometimes more important. For example, Mullard 12ax7 long plate v short plate, they have very different tone and character even though the long and short plate tubes may measure very similar.
Analogsa is right, depends heavily what the tubes are used for
Thanks for the replies... For example, do 300B tubes need to be matched across the monoblocks for Audion Golden Nights (1 300b tube in each amp)?
Do all 6550s need to be matched in a VTL 750? Within a single monoblock? Across both monoblocks?
Do all 6550s need to be matched in a VTL 750? Within a single monoblock? Across both monoblocks?
Similar threads
 | Steve Williams Site Founder | Site Owner | Administrator |  | Ron Resnick Site Co-Owner | Administrator |  | Julian (The Fixer) Website Build | Marketing Managersing |