Yes, I know this is probably a dumb question - but I couldn't find the answer by searching.
In an SET design is it ever necessary to buy the output tubes matched? If so why?
I can see if you had a push-pull design how it could make a difference.
I'm generally a cheap SOB, and I'd rather not waste money on a "feature" I don't really need.
In an SET design is it ever necessary to buy the output tubes matched? If so why?
I can see if you had a push-pull design how it could make a difference.
I'm generally a cheap SOB, and I'd rather not waste money on a "feature" I don't really need.
Matched tubes for the Single Ended amps is to ensure that your Left and Right amplifiers are similair....
Thats unless you are running in MONO
Push-Pull amps need at least two things matched with the tubes... First the DC plate current should be fairly close to keep the Output transformer flux excursion symmetrical and close the the origin to avoid saturations at either quadrant... which is debatable, but another discussion...
Another thing that needs to be matched is the AC plate current with respect to AC input grid voltage, known as transconductance.... Somehow some people think that DC matched tubes means that the tubes are gm matched...oh well..
Also CATHODE current is not the same as PLATE current....most people seem to measure current in the cathode instead of plate...
The cathode current is a SUM of the plate current and screen current...
The screen currents on some modern tubes can be quite high at idle and even more rediculous when amplifiing signal... Typical EL34 can have anywere from 2mA to 8mA of screen current at idle...This will jump up to about 24mA when the amp is at full clean power output... Upon realy cranking the amp into clipping the screens can pull as much as 32mA each, thus saturating the screen filter inductor....
So you can see that cathode current measurements can be eroneous when matching tubes even at idle conditions...
Matching DC current for Push-Pull is also important for HUM reduction in the output stage....
Some amps have generous amount of negative voltage feedback applied to the phase inverter tail...In these situations any AC signal imbalance due to non gm matched tubes will be corrected by the Phase Inverter since it will "push" in opposition to the offset feedback signal, thus balancing the signal in the output stage...This is not always linear with respect to frequency, since at the frequency extremes the imbalance will return depending of course on the closed loop response of the system....
Chris
Thats unless you are running in MONO
Push-Pull amps need at least two things matched with the tubes... First the DC plate current should be fairly close to keep the Output transformer flux excursion symmetrical and close the the origin to avoid saturations at either quadrant... which is debatable, but another discussion...
Another thing that needs to be matched is the AC plate current with respect to AC input grid voltage, known as transconductance.... Somehow some people think that DC matched tubes means that the tubes are gm matched...oh well..
Also CATHODE current is not the same as PLATE current....most people seem to measure current in the cathode instead of plate...
The cathode current is a SUM of the plate current and screen current...
The screen currents on some modern tubes can be quite high at idle and even more rediculous when amplifiing signal... Typical EL34 can have anywere from 2mA to 8mA of screen current at idle...This will jump up to about 24mA when the amp is at full clean power output... Upon realy cranking the amp into clipping the screens can pull as much as 32mA each, thus saturating the screen filter inductor....
So you can see that cathode current measurements can be eroneous when matching tubes even at idle conditions...
Matching DC current for Push-Pull is also important for HUM reduction in the output stage....
Some amps have generous amount of negative voltage feedback applied to the phase inverter tail...In these situations any AC signal imbalance due to non gm matched tubes will be corrected by the Phase Inverter since it will "push" in opposition to the offset feedback signal, thus balancing the signal in the output stage...This is not always linear with respect to frequency, since at the frequency extremes the imbalance will return depending of course on the closed loop response of the system....
Chris
cerrem said:
If matching reduces hum that means a power supply needs to be greatly improved. Idle current is usually lower than peak current, so at peak current hum level will be proportionally higher. That means, hum will be added to the music, and what is worse, due to non-linearity of plate curves, it will inter-modulate with the signal.
To my mostly-unfinicky ears, it's not very important at all. The two JJ 6L6GCs I have for the Simple SE are completely and utterly unmatched. I bought one to try in a Champ circuit and used it for a bit. About 6 years elapsed before I bought the second so that I could use them in this amp.
I have separate volume controls for this reason, but the difference in "volume" is barely noticeable. It might be noticeable at super-low volume levels...I haven't really paid attention enough to try.
Often the price for a "matched" pair is the same as buying two singles, so I'll just get the pair. In reality what I do for these "standard" octal tubes is that I generally buy a matched quad instead so that I can use them in any of my amps...PP or SE. I just have to put a tick mark on the box of the pair I have used for any significant amount of time in the SE so that the quad has some amount of wear-leveling. That way if I don't care for the sound in the SE, I can always use them up at some point in the Dynaco and/or in a room where I don't really do "critical" listening. I have tube amps all over the house now....
I have separate volume controls for this reason, but the difference in "volume" is barely noticeable. It might be noticeable at super-low volume levels...I haven't really paid attention enough to try.
Often the price for a "matched" pair is the same as buying two singles, so I'll just get the pair. In reality what I do for these "standard" octal tubes is that I generally buy a matched quad instead so that I can use them in any of my amps...PP or SE. I just have to put a tick mark on the box of the pair I have used for any significant amount of time in the SE so that the quad has some amount of wear-leveling. That way if I don't care for the sound in the SE, I can always use them up at some point in the Dynaco and/or in a room where I don't really do "critical" listening. I have tube amps all over the house now....
Wavebourn said:
You are 100% correct.....
Many Push-Pull amplifiers are not well designed...instead they try to save money and use under filtered power supply in the hopes the output stage will cancel most of the hum....
It's really not the plate current matching that cancels the hum.... since the plate in pentodes not that sensitive the plate voltage variation.... It's the screen and input grid balancing that has more to do with hum cancelation.....
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.