What is Tube Matching?
- Thread starter Duke58
- Start date
Register to hide this ad
Tubes are mechanical devices, made by small spotwelded components and mounted in mica-spacers. As such they cannot be made identicall, thus
their characteristics are spread across the "allowed range" for the type.
Thus im many applications any tube of a given type will function, but
for some applications ( and push-pull power amps is such) matching
of some parameters will improve the result.
See the last paragraph of :
http://www.aikenamps.com/index.php/the-last-word-on-biasing
Rob-F
Likes Vintage Gear and DIY
What tube matching accomplishes
For small-signal tubes, this is my answer also. Matching for this purpose can be done with a transconductance type tester,such as a Hickok 600a, 6000, etc.
For power output tubes, we care a little less about transconductance, and much more about the grid bias voltage needed to achieve a given plate current. For example, a given tube might require -32V (volts) on the grid to get 50ma (milliamps) of plate current. If so, then the other tube in a push-pull pair should require the same, or nearly the same, grid bias for the same plate current.
But what if one tube needs -30V, and the other tube in the pair needs -40V for the same plate current? Then, when the amp is set up for its prescribed plate current, the second tube will be biased ten volts closer to cutoff then its twin. That's not good. I will post a link to a website that explains, in more detail, why it's not good.
Most tube testers are not much good for matching power tubes. I'd like to make one that has a meter for the grid voltage, and a meter for the plate current. In the meantime, I've bought matched tube sets from both TubeDepot.com and from amplifiedparts.com. TubeDepot puts a sticker on the base of each tube with the bias voltage marked on the sticker. That way, even if I get some tubes mixed up when changing tubes, I can just refer to the sticker to make sure I'm installing a matched pair or quad. I like that.
On the other hand, amplifiedparts.com marks the tube parameters on the box, not the tube. If you get them mixed up--easy to do when switching quad sets--you no longer know which tube is which.
Here's the article by Roger Modjeski.
http://www.ramlabs-musicreference.com/virtues.html
I wonder if Roger marks the values on the tubes.
Hope that helps.
For small-signal tubes, this is my answer also. Matching for this purpose can be done with a transconductance type tester,such as a Hickok 600a, 6000, etc.
For power output tubes, we care a little less about transconductance, and much more about the grid bias voltage needed to achieve a given plate current. For example, a given tube might require -32V (volts) on the grid to get 50ma (milliamps) of plate current. If so, then the other tube in a push-pull pair should require the same, or nearly the same, grid bias for the same plate current.
But what if one tube needs -30V, and the other tube in the pair needs -40V for the same plate current? Then, when the amp is set up for its prescribed plate current, the second tube will be biased ten volts closer to cutoff then its twin. That's not good. I will post a link to a website that explains, in more detail, why it's not good.
Most tube testers are not much good for matching power tubes. I'd like to make one that has a meter for the grid voltage, and a meter for the plate current. In the meantime, I've bought matched tube sets from both TubeDepot.com and from amplifiedparts.com. TubeDepot puts a sticker on the base of each tube with the bias voltage marked on the sticker. That way, even if I get some tubes mixed up when changing tubes, I can just refer to the sticker to make sure I'm installing a matched pair or quad. I like that.
On the other hand, amplifiedparts.com marks the tube parameters on the box, not the tube. If you get them mixed up--easy to do when switching quad sets--you no longer know which tube is which.
Here's the article by Roger Modjeski.
http://www.ramlabs-musicreference.com/virtues.html
I wonder if Roger marks the values on the tubes.
Hope that helps.
maxhifi
Addicted Member
There is an RCA publication called the Radiotron Designer's Handbook version 4. You can likely find it on the web, but it's a book from the early 1950s which many consider to be somewhat of a bible of tube design. It has an example somewhere in its pages, analyzing the result when two completely different tubes are used in a push-pull configuration, in that case, a 45, and and a 2A3. I would suggest tracking it down and reading it, it's a good illustration of why push-pull tubes must be matched.
In a practical setting, tube matching is critical for two reasons.
1. If the bias currents of output tubes are not matched, there will be unbalanced current through the core of the output transformer. This results in "core saturation", which affects how much power the transformer can pass at low frequencies, and also increases low frequency distortion substantially. Even a 5mA imbalance can cause a measurable difference, which prevents the amplifier from performing properly. This is why when we match output tubes, we measure how many mA are passed for a given grid voltage. This parameter is significantly less critical when an amplifier can be adjusted to equalize cathode current independently for each output tube, either with individual bias pots, or a bias pots and balance pots. Note that this value changes over the life of a tube, and should be periodically re-adjusted. I usually adjust it when tubes are first installed, then a day later, then a month later... and then whenever the amp is serviced for any reason.
2. Transconductance - this caused gain differences between halves of the push-pull amplifier. This is more critical in amplifiers where negative feedback isn't used, so in most modern amplifiers isn't really very critical, because gain is determined more by the negative feedback circuit than by the tubes themselves. All the same, some amplifiers still employ controls to zero out any imbalance, to optimize performance.
3. I wouldn't worry too much about matching tubes between left and right, unless there are gross differences, mainly because of point 2. Same brand and age should be enough.
Note: The best tool for matching power tubes is something you probably already have - your amplifier! Most larger amplifiers have variable grid bias voltage, and the ability to measure cathode current, and you can use these features to match tubes. A tube tester does a good job of measuring transconductance, but it's not really intended for matching power tubes.
In a practical setting, tube matching is critical for two reasons.
1. If the bias currents of output tubes are not matched, there will be unbalanced current through the core of the output transformer. This results in "core saturation", which affects how much power the transformer can pass at low frequencies, and also increases low frequency distortion substantially. Even a 5mA imbalance can cause a measurable difference, which prevents the amplifier from performing properly. This is why when we match output tubes, we measure how many mA are passed for a given grid voltage. This parameter is significantly less critical when an amplifier can be adjusted to equalize cathode current independently for each output tube, either with individual bias pots, or a bias pots and balance pots. Note that this value changes over the life of a tube, and should be periodically re-adjusted. I usually adjust it when tubes are first installed, then a day later, then a month later... and then whenever the amp is serviced for any reason.
2. Transconductance - this caused gain differences between halves of the push-pull amplifier. This is more critical in amplifiers where negative feedback isn't used, so in most modern amplifiers isn't really very critical, because gain is determined more by the negative feedback circuit than by the tubes themselves. All the same, some amplifiers still employ controls to zero out any imbalance, to optimize performance.
3. I wouldn't worry too much about matching tubes between left and right, unless there are gross differences, mainly because of point 2. Same brand and age should be enough.
Note: The best tool for matching power tubes is something you probably already have - your amplifier! Most larger amplifiers have variable grid bias voltage, and the ability to measure cathode current, and you can use these features to match tubes. A tube tester does a good job of measuring transconductance, but it's not really intended for matching power tubes.
Last edited:
Tubes have no moving parts. They are electrical devices, though your description of their construction is generally accurate. Some plate are stapled/riveted rather than spot-welded, though.
Hiya,
No you would not necessarily be better off buying 4 tubes because most matching services don't take into account the differences that exist within your own amplifier.
Its a baseline nothing more or less these days and IMHO a way to bump up the cost on the part of the seller.
Yes its a good thing but too much of a "Good Thing" exists. Too many sellers don't have a darn clue how to properly test a tube much less claim matching.
And as stated right here in this thread. Your best matching device is your own amplifier. Learn how to match up tubes in a amp and you can bypass a lot of annoyance and confusion.
Frannie
No you would not necessarily be better off buying 4 tubes because most matching services don't take into account the differences that exist within your own amplifier.
Its a baseline nothing more or less these days and IMHO a way to bump up the cost on the part of the seller.
Yes its a good thing but too much of a "Good Thing" exists. Too many sellers don't have a darn clue how to properly test a tube much less claim matching.
And as stated right here in this thread. Your best matching device is your own amplifier. Learn how to match up tubes in a amp and you can bypass a lot of annoyance and confusion.
Frannie
richamor
amateur house husband
One of the reasons I went with the Classic Valve power supply in my ST-70 was to avoid the whole balanced quad thing as it gives the ability to bias and balance each channel.
I agree with Frannie and feel like there's a lot of snake oil hoodoo out there in the tube world. We all know a "matched and balanced quad" will fetch more money and most buyers don't have the ability of knowledge to say otherwise.
eta: The above is why I prefer to buy either new or NOS from reputable dealers or used from AK members.
I agree with Frannie and feel like there's a lot of snake oil hoodoo out there in the tube world. We all know a "matched and balanced quad" will fetch more money and most buyers don't have the ability of knowledge to say otherwise.
eta: The above is why I prefer to buy either new or NOS from reputable dealers or used from AK members.
Last edited:
jaymanaa
RIP 1961-2018
Yep, believe 1% of what you read, and match them where you run them.
Then there are the seller's like Jim McShane who tests them on a special rig at operating voltages, and matches them like you would on an amp. Kind of the same way Dave Gillepie tests his tubes on his home made test rigs.
onemug
Super Member
Buying tubes from Jim and the way he tests them made a big improvement in my preamp and his prices are not out of line at all.
Hiya,
Its a mess to be honest as every single vendor has different, mostly dodgy logic as to what matching actually is.
As far as I am concerned matched tubes indicate exact same structure and brand, same Current Draw and matched Mutual Conductance if reading on a higher quality tube tester. (You can get even finer and look for matching date codes if you really want to get detailed)
Throw in Balanced Sections as far as Current Draw and Mutual Conductance if we are talking Preamp tubes.
Also you can throw Rectifiers into this if Full Wave as you usually want those two sections to match up.
But most dork vendors out there call them matched if the boxes look the same, emission test results look the same on some crappy tube tester, if they are the same brand, same looking, or the Cat thinks they roll around the floor of the Hovel the same.
Frannie
Its a mess to be honest as every single vendor has different, mostly dodgy logic as to what matching actually is.
As far as I am concerned matched tubes indicate exact same structure and brand, same Current Draw and matched Mutual Conductance if reading on a higher quality tube tester. (You can get even finer and look for matching date codes if you really want to get detailed)
Throw in Balanced Sections as far as Current Draw and Mutual Conductance if we are talking Preamp tubes.
Also you can throw Rectifiers into this if Full Wave as you usually want those two sections to match up.
But most dork vendors out there call them matched if the boxes look the same, emission test results look the same on some crappy tube tester, if they are the same brand, same looking, or the Cat thinks they roll around the floor of the Hovel the same.
Frannie
:lmao: You might as well add the Dog to that too. Lucy just about went nutz when i dropped a 6AU6 on the floor and she couldn't get hold of it. She'd put her paw on it and the slippery little devil would shoot across the floor. Then she'd look up at me with a "WTF was that?" look on her face, and go chase it down. She never did get the tube.
sKiZo
Hates received: 92644 43.20°N 85.50°W
Factory matching seems like your best bet as they tend to test more values than the average vendor. As mentioned, that's pretty much a crap shoot for quality purposes and actual results. The cat roll test may actually be higher up the scale than some of the sellers apply ...
I recently had some fun going back and forth between a factory direct reseller that applies "premium matching" to their tubes before they leave the door, vice an aftermarket source that provides cryonic treatment to these same tubes. Apparently, the Gm can go up 3-5% after the process. They assure me the tubes are re-matched, but no details on that, and no telling what all else changes. What's interesting is the cryo place sells a "matched quad" for $20 less than the official factory distributor here in the states.
No names mentioned, but as you can tell, I'm a TAD bit confused here ... <G>
I recently had some fun going back and forth between a factory direct reseller that applies "premium matching" to their tubes before they leave the door, vice an aftermarket source that provides cryonic treatment to these same tubes. Apparently, the Gm can go up 3-5% after the process. They assure me the tubes are re-matched, but no details on that, and no telling what all else changes. What's interesting is the cryo place sells a "matched quad" for $20 less than the official factory distributor here in the states.
No names mentioned, but as you can tell, I'm a TAD bit confused here ... <G>
