Matched tubes at one point, but not anymore?

minestaken

TDPRI Member
Joined
Nov 18, 2017
Posts
32
Location
Michigan
I built a Princeton Reverb Mojo kit several years ago and put in some NOS 6v6GTY RCA. Everything checked out OK from what I remember. Fast forward to now and I have 16Mv and 22Mv on each side after I tested with a EuroTube tester. Huh... Over time a slight drop in volume might have occurred, but I never really thought much of it. I don't really crank the amp - i'm an at home player.

I recently dropped in a set of matched wafer base GE and got about 15Mv on both tubes. All good, but a bit cold. Btw, I'm using a NOS 5AR4 rectifier.

On to my questions:
+ Is one of the 6V6GTY tubes going (gone?).
+ Wafer base GEs - with such a cold bais, what are my options?

I could mod with a bias pot.. But then again, I might just get a set of Tesla/JJ and call it a day.
 

sds1

Friend of Leo's
Joined
May 4, 2017
Posts
2,161
Age
46
Location
Orlando, FL, USA
Is one of the 6V6GTY tubes going (gone?).
I dunno if that's fair to assume...

You might swap the tubes to opposite sockets to confirm the tubes idle the same on either side.

Also, tubes do drift over time... mostly at the beginning of their life AFAIK...

I could mod with a bias pot..
Or 2x bias pots even. This avoids having to match tubes and deals with the drift problem as well.
 

minestaken

TDPRI Member
Joined
Nov 18, 2017
Posts
32
Location
Michigan
Yep, I did swap tubes during my bias check, and the 16Mv travels with the tube. I didn't check plate however - that's my next check.
 

Phrygian77

Poster Extraordinaire
Gold Supporter
Joined
Apr 30, 2016
Posts
6,007
Location
Crawfordville, FL
The center tap on the primary is not perfectly center. You definitely need to swap the tubes around and measure both current and voltage.

You said several years ago, so how many hours do you think you have on those tubes? It may be that they're already going south, probably starting with the one one the hotter side of the primary.
 
Last edited:

minestaken

TDPRI Member
Joined
Nov 18, 2017
Posts
32
Location
Michigan
I probably have 3 years on them -- maybe 500 hours maximum? They are NOS, so who knows. I bought them from a reputable source that I've worked with for 10+ years.
 

robrob

Poster Extraordinaire
Ad Free Member
Joined
Dec 29, 2012
Posts
8,633
Location
United States
I recently dropped in a set of matched wafer base GE and got about 15Mv on both tubes. All good, but a bit cold. Btw, I'm using a NOS 5AR4 rectifier.

Out-of-balance power tubes generate more even-order harmonics so they can actually sound better than balanced power tubes. As long as the imbalance isn't so high to allow heater hum to increase the amp will usually sound better (warmer).
 

minestaken

TDPRI Member
Joined
Nov 18, 2017
Posts
32
Location
Michigan
Just measured plate -- i'm getting 406VDC on both sides.

So with an unmatched set at 16mV and 22mV and a bias pot, how is bias actually dialed in? By ear? Or do you average the 2 readings?

I can't say I notice much of a tonal difference between the 2 sets - both sound good. If anything the wafer base matched pair is a bit louder at the same settings and maybe a bit more focused. Could all be placebo too :)
 

Lowerleftcoast

Poster Extraordinaire
Joined
Dec 30, 2019
Posts
5,762
Location
california
Is one of the 6V6GTY tubes going (gone?).
Hard to say. If they remain reasonably stable at these numbers they will last. They may be just a set that are no longer matched.
Wafer base GEs - with such a cold bais, what are my options?
Well, you can warm the bias. A change in the resistance of the bias tail resistor. Changing the resistor, adding a resistor in parallel to the tail resistor, or adding a pot to adjust the bias voltage, can all get you there.
I could mod with a bias pot.. But then again, I might just get a set of Tesla/JJ and call it a day.
What makes you think a different set (Tesla/JJ) will run warmer?
So with an unmatched set at 16mV and 22mV and a bias pot, how is bias actually dialed in? By ear? Or do you average the 2 readings?
Imo, always by ear. When you have found a pleasant setting check the hottest tube to make sure it is not running too hot. If it is too hot, back off until the tube can handle the current. With the Princeton Reverb, the tremolo will become weak as the bias is set hotter so many will set it by tremolo performance. This usually ends up cooler than where one may set the bias otherwise.
 

schmee

Telefied
Silver Supporter
Joined
Jun 2, 2003
Posts
21,278
Location
northwest
I've had it occur. As Rob Rob says, it may be a non issue. I assume the tubes change a bit with wear. The other thing to watch for when biasing is to leave the amp on for 10 -15 minutes. I've thought things were good and stable before, only to find after a while the bias may move as much as 5 ma as they continue to heat up/break in.

When we had a bad tube in the 60's we went to Safeway and bought one out of the bottom cabinet in the big tube tester...... and only replace that one. We knew nothing about bias.
Looking back, it was probably OK most of the time, but also likely a reason why certain amps just never sounded that good and others were great.
 

Wally

Telefied
Ad Free Member
Joined
Mar 17, 2003
Posts
42,202
Location
Lubbock, TX
I built a Princeton Reverb Mojo kit several years ago and put in some NOS 6v6GTY RCA. Everything checked out OK from what I remember. Fast forward to now and I have 16Mv and 22Mv on each side after I tested with a EuroTube tester. Huh... Over time a slight drop in volume might have occurred, but I never really thought much of it. I don't really crank the amp - i'm an at home player.

I recently dropped in a set of matched wafer base GE and got about 15Mv on both tubes. All good, but a bit cold. Btw, I'm using a NOS 5AR4 rectifier.

On to my questions:
+ Is one of the 6V6GTY tubes going (gone?).
+ Wafer base GEs - with such a cold bais, what are my options?

I could mod with a bias pot.. But then again, I might just get a set of Tesla/JJ and call it a day.
this is why I believe in ‘burn in’ treatment before the matching process is done. Tubes can change their operational parameters during the first 24-36 hours. Whether or the match is unsatisfactory after that drift is dependent on how great the mismatch is and what the user hears. After this initial,burn-in drift…if it occurs…..the tube will be stable.
 

dan40

Friend of Leo's
Joined
Aug 19, 2015
Posts
2,852
Location
Richmond Va
It's very possible that the tubes drifted apart in their first few hours of use years ago and have been like this ever since. A 6ma difference wouldn't bother me one bit and as long as the trem functions good, I wouldn't sweat it.
 

King Fan

Poster Extraordinaire
Ad Free Member
Joined
Jan 1, 2013
Posts
8,531
Location
Salt Lake City
Great thread, good answers throughout.

As a practical detail, measured mA isn't bias; you were right to measure plate voltages. If you plug those *and* current into a bias calculator like Rob's, and look at %MPD, you can then actually see what you're 'hearing.'

Also, a bias adjust pot is almost mandatory for me -- you can't tune by ear any other way. The PR implementation of this simple mod is especially easy and well-tested. I like the time-tested Doug Hoffman version, but Rob has a nice one on his PR page too.

https://el34world.com/charts/Biascircuits.htm

My drawing if it helps:

Hoffman simple bias pot.jpeg
 

Phrygian77

Poster Extraordinaire
Gold Supporter
Joined
Apr 30, 2016
Posts
6,007
Location
Crawfordville, FL
Great thread, good answers throughout.

As a practical detail, measured mA isn't bias; you were right to measure plate voltages. If you plug those *and* current into a bias calculator like Rob's, and look at %MPD, you can then actually see what you're 'hearing.'

Also, a bias adjust pot is almost mandatory for me -- you can't tune by ear any other way. The PR implementation of this simple mod is especially easy and well-tested. I like the time-tested Doug Hoffman version, but Rob has a nice one on his PR page too.

https://el34world.com/charts/Biascircuits.htm

My drawing if it helps:

View attachment 1016030

IMO, your pot is backwards, but it functions the same as long as you remember which way is which.
 

minestaken

TDPRI Member
Joined
Nov 18, 2017
Posts
32
Location
Michigan
Thanks everyone - the most frustrating part about all of this is now I also have my 5f1 on the bench testing bias :). Does anyone know where I can source a variable ~5W resistor? I saw one of Uncle Doug's videos where he was servicing an early champ(?) where it had one of these resistors. No need to solder, just a few turns of the screw driver. Thought it would be super convenient to try different tubes.

btw, @Phrygian77 i was getting plate voltages between 403 an 407 on both sides of the PR. I forget the exact measurements, but you are correct, they were not the same.
 

jimbo735

Tele-Afflicted
Joined
Sep 19, 2011
Posts
1,488
Location
Michigan
You will not regret having a Bias pot, you will want a Bias pot installed in everything! It really is that good.WORD!
 

Phrygian77

Poster Extraordinaire
Gold Supporter
Joined
Apr 30, 2016
Posts
6,007
Location
Crawfordville, FL
@minestaken you don't want to change the bias resistor on cathode biased class A amps. If you're voltage is above the safe class A operating range, you just need a tube that can handle it, or you need to bring the voltage down in some way. If you intentionally bias it cold, you're going to push the output off center towards cutoff clipping.
 

King Fan

Poster Extraordinaire
Ad Free Member
Joined
Jan 1, 2013
Posts
8,531
Location
Salt Lake City
IMO, your pot is backwards, but it functions the same as long as you remember which way is which.
It depends on shaft-up vs. shaft-down. My actual amp, with an external-adjust pot, is wired like Doug's drawing.

1660315733720.png

But mine above is a version I drew for folks retrofitting (like minestaken) who wanted a 10KL Fender pot like Doug's but who didn't want to drill their chassis for external adjust. (The 10KL works extremely well here and gives you room to drop the tail resistor, if necessary, without going below 10K).

Shaft-up also applies to folks want to use a shaft-adjust pot, like Rob --

1660315585543.png
 

Phrygian77

Poster Extraordinaire
Gold Supporter
Joined
Apr 30, 2016
Posts
6,007
Location
Crawfordville, FL
@King Fan another option for someone doesn't want to drill or modify a vintage amp for example, is to replace the original 22k with a 50k Piher in series with a 10k resistor.

biascircuit3.gif


This works really well. It is sensitive to adjustment though.
 




Top