Quad II output tube matching

[sempai]

I recently received 4 British made GEC Kt66's.
Clear glass, double O getters.
For use in a pair of Quad II's.

Much to my chagrin one began to redplate soon after being put into use.

I know that with the Quad II's the output tubes can be mismatched by as much as 20% and the amp will still be happy.
So I'm guessing that tube matching is not as critical with the Quad II's as it might be with many other amps.
I also know that searching for a matching single GEC KT66 with the same date code could turn into a full time job.

My questions:
How critical is tube matching for the Quad II's?
Will I be ok with just finding 1 British made, clear glass, double 0 getter GEC KT66 with close specs?
Do I now need to source out another complete matched quad?


Art

 

[vinylkid58]

My questions:
How critical is tube matching for the Quad II's?
Will I be ok with just finding 1 British made, clear glass, double 0 getter GEC KT66 with close specs?
Do I now need to source out another complete matched quad?

I would have thought having matched pairs to be reasonably important, as there's no provision to adjust individual cathode voltage for each KT66. Is there anyone that can test the tubes you have, to see if they are at all close? That's the first thing I'd do, before looking for more.

jeff

 

[gadget73]

I'd also swap those tubes around to make sure the problem follows the tube, and isn't in fact an issue with the amplifier.

matching is an interesting subject. By the book, yes it should be matched. In reality, things will function with a fairly bad mismatch, though it will not be ideal at all. Its just how particular you wish to be. The date codes are actually a lot less important here than the actual operational characteristics of a tube. Same or close date codes would make the tubes theoretically built pretty much the same, but they could still be electrically different due to manufacturing tolerances.

 

[gogofast]

20% margin of mismatch can be tolerated by a lot of other tube amps in my experience. the thing is, with a dozen tubes from same manufacturer, differences in plate currents will be way more than 20% in most cases. also, as mentioned already by others, your problem might be in bias circuit instead of the tubes.

 

[sempai]

I'd also swap those tubes around to make sure the problem follows the tube, and isn't in fact an issue with the amplifier.

I did move the faulty tube.
To all positions on both amps.
In all cases the redplating followed the tube.
I'm currently using Sovtec 5881's none of which redplate.

So I'm reasonably confident it's not a bias problem.

The date codes are actually a lot less important here than the actual operational characteristics of a tube. Same or close date codes would make the tubes theoretically built pretty much the same, but they could still be electrically different due to manufacturing tolerances.

This is what I was hoping to hear.

I take it then that finding a single closely matched GEC with the same build (British made, clear glass, double 0 getters) is ok regardless of date.
I wasn't sure about the relative importance of the date when it comes to matching.
Many of the matching quads of vintage KT66's I've seen for sale seem to stress the matching date codes as much as the closeness in output.

Art

Edit: The Sovtecs 6L6GC not 5881

 

[vinylkid58]

I take it then that finding a single closely matched GEC with the same build (British made, clear glass, double 0 getters) is ok regardless of date.

"Matched" implies that you are buying at least a pair of tubes. A single tube can't be matched to anything.

Many of the matching quads of vintage KT66's I've seen for sale seem to stress the matching date codes as much as the closeness in output.

Well, for the price they're going for, I'd want them to be made on the same day.

jeff

 

[sempai]

"Matched" implies that you are buying at least a pair of tubes. A single tube can't be matched to anything.

jeff

I meant matched to the three remaining good tubes.

Well, for the price they're going for, I'd want them to be made on the same day.

jeff

thus the reason for this post :sigh:

Art

 

[vinylkid58]

I meant matched to the three remaining good tubes.

How do you know your three good tubes are a close match? Wouldn't it be easier, and a LOT cheaper to get a matched quad of reissue Genelex KT66's from a reputable dealer?

jeff

 

[sempai]

edit the sovtecs are

How do you know your three good tubes are a close match? Wouldn't it be easier, and a LOT cheaper to get a matched quad of reissue Genelex KT66's from a reputable dealer?

jeff

You can find a month's worth of reading on KT66's on the web.

Reissue kt66's run hotter than the vintage British GEC's.

Running the newer hotter reissues are not good for the aging Quad II transformers

In deffernce to the age of the transformers I would rather not stress them any more than necessary.
And I like these amps and have no plans of ever replacing them.

The Sovtec 6L6GC run very cool and sound ok but not nearly as nice as the vintage KT66's.

Art

 

[gogofast]

it is very possible that you will find a matching old-stock single tube. what you'll have to do is find out the plate current and transconductance of the three good tubes you have, inform the test results to possible seller and find something that's as closely testing single. it's not impossible, but also not easy since there aren't that many of them out there. it can also be pricy if the seller doesn't know what he/she is doing.

when you did the "red plating" test, did you move the tubes as pairs or did you try pairing the red plating tube with two other tubes? if you did just move it around as a pair, try pairing it with other two tubes, that might work if that red-plating tube is still good but just not matching with currently paired tube.

 

[gogofast]

Reissue kt66's run hotter than the vintage British GEC's.

Running the newer hotter reissues are not good for the aging Quad II transformers

Art

i'm pretty sure there are mods out there that can lower the voltage and make the new production tubes run cooler.

imhe, between the old stock and new production tubes, biggest difference is the reliability and not too much on sound - though there is a noticeable difference. i do not have the quad amp in question, but that was the case in all of my tube amps - eico, marantz, mcintosh, and etc.

so, if it turns out that you can't find a single old stock tube to complete your quad, don't feel too bad. new production tubes can be used with very similar results.

 

[sempai]

it is very possible that you will find a matching old-stock single tube. what you'll have to do is find out the plate current and transconductance of the three good tubes you have, inform the test results to possible seller and find something that's as closely testing single. it's not impossible, but also not easy since there aren't that many of them out there. it can also be pricy if the seller doesn't know what he/she is doing.

when you did the "red plating" test, did you move the tubes as pairs or did you try pairing the red plating tube with two other tubes? if you did just move it around as a pair, try pairing it with other two tubes, that might work if that red-plating tube is still good but just not matching with currently paired tube.

Yes i did try pairing with the other tubes.
Unfortunately with no luck.

But I'll give it another try anyway.
It's Christmas.
Miracles do happen

I don't think it will be too terribly hard to find the single tube I need.
Just need a little patience.

Art

 

[bhamham]

I don't think it will be too terribly hard to find the single tube I need.
Just need a little patience.

Art

That's what I'd do. I just buy a single and give it a go. These amps are pretty forgiving for not being an exacting match.

 

[sempai]

That's what I'd do. I just buy a single and give it a go. These amps are pretty forgiving for not being an exacting match.

This was my interpretation of what I had read elsewhere. (vintage-radio.net)
But I wasn't absolutely sure.

I have also read that -ve feedback means effort with valve matching and concern over valve linearity / quality is probably a waste of time.
Within reason of course
This apparently is part of the genius of Peter Walker's design.

However there also seems to be so much emphasis on tube matching.

I have already started searching for a single match.

Art

 

[petehall347]

i know its a bit long winded but tubes can be matched in the amp using resistors to check the current ..i cant remember offhand what to connect to but someone here will sure remind me .

 

[gogofast]

i know its a bit long winded but tubes can be matched in the amp using resistors to check the current ..i cant remember offhand what to connect to but someone here will sure remind me .

i think the quad II's have common cathode resistor. measure voltage across that resistor for each tube. from one end of the resistor to the other (that goes to each tube's cathode - pin #8).

 

[sempai]

i think the quad II's have common cathode resistor. measure voltage across that resistor for each tube. from one end of the resistor to the other (that goes to each tube's cathode - pin #8).

Thanks gogofast.
While this is a very good bit of knowledge that you have passed along that one day may get put to good use, I am still on the lower slopes of the learning curve with valve amps, am very mindful of the voltages wending their way about the Quad's, and have no wish for my heart to stop beating before its time.

Art

 

[VacuuMe]

I also noticed overload of one of KT66 in my Quads. One plate was becoming red. Tubes were well matched.
It was due to aged resistors in the net of R5 - R9 (in this schematic: http://www.drtube.com/schematics/quad/quad-ii.gif). They were not burnt totally but out of value significantly. Original resistors were out of tollerance in both of my Quads. This led to signal asymetry and output tubes overload. So check the values of all resistors.

Also check coupling capacitors C2, C3 and cathode bypass capacitor C5. Original Hunts capacitors in my Quads seemed to be ok when cold and after couple minutes when warmed they became leaky (and killing one of precious GEC KT66).

Hope it helps.

 

[sempai]

I also noticed overload of one of KT66 in my Quads. One plate was becoming red. Tubes were well matched.
It was due to aged resistors in the net of R5 - R9 (in this schematic: http://www.drtube.com/schematics/quad/quad-ii.gif). They were not burnt totally but out of value significantly. Original resistors were out of tollerance in both of my Quads. This led to signal asymetry and output tubes overload. So check the values of all resistors.

Also check coupling capacitors C2, C3 and cathode bypass capacitor C5. Original Hunts capacitors in my Quads seemed to be ok when cold and after couple minutes when warmed they became leaky (and killing one of precious GEC KT66).

Hope it helps.

Sorry to hear about your loss.
Loosing a GEC is painful.

All caps and resistors in my Quads where replaced recently so I don't think my problem lies there...

Thanks for the help though.

Art