Matched tubes at one point, but not anymore?

King Fan

Poster Extraordinaire
Ad Free Member
Joined
Jan 1, 2013
Posts
8,522
Location
Salt Lake City
Yep, Doug's page is a winner. I like the 10KL for the 'sensitivity' reason you mention (unless you get a multi-turn 50KL mini-pot, which IMHO gives sort of the opposite problem -- too many turns). But I especially like it if you can do external adjustment. And for those with a PR chassis who don't want to drill, it works great to mount it in the ground-switch hole. But if you don't have external tip jacks, or at least a voltage *and* current bias probe, that advantage (no need to drop chassis) goes away.
 

minestaken

TDPRI Member
Joined
Nov 18, 2017
Posts
32
Location
Michigan
Just wanted to follow up -- a big thank you to everyone!

Princeton: I added a bias pot to the princeton this weekend and tried a couple of tubes to dial things in. I first tried to re-use the existing 27k bias resistor, but that didn't provide enough current: about 8mA-19mA. I changed it up to 10k, and now, I get a sweep of 18mA - 30mA. I have it biased at around 20mA, but will soon try 25mA to see how it sounds.

Champ 5f1: Special thanks to @Phrygian77 -- I was using an 820ohm cathode bias resistor and changed it back to stock 470. Sounds much much better to me (big difference!). It originally started life as a weber kit, with the wrong transformer... So, throughout the years, I struggled to bring the voltage down and ended up at the 820. Anyway, back to spec!
 

Wally

Telefied
Ad Free Member
Joined
Mar 17, 2003
Posts
42,181
Location
Lubbock, TX
Just wanted to follow up -- a big thank you to everyone!

Princeton: I added a bias pot to the princeton this weekend and tried a couple of tubes to dial things in. I first tried to re-use the existing 27k bias resistor, but that didn't provide enough current: about 8mA-19mA. I changed it up to 10k, and now, I get a sweep of 18mA - 30mA. I have it biased at around 20mA, but will soon try 25mA to see how it sounds.

Champ 5f1: Special thanks to @Phrygian77 -- I was using an 820ohm cathode bias resistor and changed it back to stock 470. Sounds much much better to me (big difference!). It originally started life as a weber kit, with the wrong transformer... So, throughout the years, I struggled to bring the voltage down and ended up at the 820. Anyway, back to spec!
And what are your plate voltages at that current draw? Are both tubes hitting 20ma?
If your plate voltage is 415vdc at 20 ma of draw, the tubes are dissipating about 59% of max plate dissipation. At 25ma and 415, you are at about 74% of MPD.
Either should sound good. Things get richer at higher plate dissipation levels, ime. If you are shooting for cleaner sonics and/or extended tube life, the cooler setting might be for you.
 

Wally

Telefied
Ad Free Member
Joined
Mar 17, 2003
Posts
42,181
Location
Lubbock, TX
Ime, that 70% number is not written in stone as the hottest plate dissipation that one can run Class AB fixed bias. The bias point can be a very subjective and personal choice.
 
Last edited:

King Fan

Poster Extraordinaire
Ad Free Member
Joined
Jan 1, 2013
Posts
8,522
Location
Salt Lake City
Did I already say it in this thread? If so, I love repeating it. You don't bias amps in terms of mA -- some amp companies can say you do cuz they bias so cold, some can cuz they sell you very specific $$ output tubes, but mA doesn't measure the *effect* of the bias. To quote Yoda: "There is no mA. There is only %MPD and your ears."
 

minestaken

TDPRI Member
Joined
Nov 18, 2017
Posts
32
Location
Michigan
@King Fan -- I guess I don't understand the difference? 70%PD is 24ma; 60% is 20ma for my specific readings. I just can't bias by ear yet - so this is all I got for now. As I progress, one of the first questions to myself is 'can I tell a difference between 60% and 70%' - we'll see, need more testing :).

On a separate note - I'm really enjoying your threads on the Princeton, specifically the mod report card. Super good info!!
 

King Fan

Poster Extraordinaire
Ad Free Member
Joined
Jan 1, 2013
Posts
8,522
Location
Salt Lake City
Aha. If you never change output tubes, and your tubes never age into different behavior, *and* you make a table of each mA value (say 18, 20, 22..) with its resulting plate voltage and %MPD, you could say that 24mA = 70%. Unless things change.

But you shouldn’t be getting the same PV at markedly different mA readings on *any* tubes. Changing the bias current changes the plate voltage. Have you measured PV at different points in your bias current range? What method are you using to calculate %MPD? Rob's calculator makes it easy to get %MPD *but* needs PV updated — in the first box — each time.

As far as mismatched tubes, you could average, or some people just set the hotter tube as the upper or more sensitive 'indicator' of how hot is hot.

For bias by ear, turn your bias all the way down (negative bias current at max, mA at minimum). Measure PV at that mA and calculate %MPD. Turn the amp volume well up. Play a big chord. Turn bias up some, repeat that chord at that bigness. It’s super helpful to have mini-grabber probes attached to watch mA as you turn — and yes, at this point knowing roughly where the ~70% MPD 'speed limit' sits will help. Feel free to range back down from that point, still listening. *Make sure you also listen to the trem at each level.* When the tone and trem sound best to you, do PV and MPD again to be sure you’re inside roughly 50-70%. You can leave it there, but recheck as your tubes age and if you change tubes.

If you’re hearing no change in tone over the range, check that those PV measurements are actually changing. Let us know if problems.
 
Last edited:

Wally

Telefied
Ad Free Member
Joined
Mar 17, 2003
Posts
42,181
Location
Lubbock, TX
@King Fan -- I guess I don't understand the difference? 70%PD is 24ma; 60% is 20ma for my specific readings. I just can't bias by ear yet - so this is all I got for now. As I progress, one of the first questions to myself is 'can I tell a difference between 60% and 70%' - we'll see, need more testing :).

On a separate note - I'm really enjoying your threads on the Princeton, specifically the mod report card. Super good info!!
Re: setting bias by ear. There is no time like the present to learn and to develop what your ear likes, imho. 30 ma of current draw at a hypothetical 400vdc on the plates is 85% of max plate dissipation for a 6V6. Ime, this will not redplate a healthy tube, but one might want to watch the tubes while playing in a darkened room. I have witnessed fixed biased amps running at 85% of MPD with no ill effects, and some prefer to run them at such a dissipation. There are amps that run even higher than that with no redplatino. Yes, higher plate dissipation works the tube harder and shortens life to some extent.
So, with a cautious approach, you could use the amp you have there to develop/educate the ear as to how you want that amp to function for you.
 

robrob

Poster Extraordinaire
Ad Free Member
Joined
Dec 29, 2012
Posts
8,633
Location
United States
Some of the best sounding amps of all time, Fender blackface amps, came from the factory with bias around 55%. Hotter isn't necessarily better. Merlin Blencowe uses 80% as his "max safe bias" for Class AB fixed bias amps so 70% definitely isn't a hard and fast limit.
 

minestaken

TDPRI Member
Joined
Nov 18, 2017
Posts
32
Location
Michigan
Alright, I think I have it dialed in, thanks for the encouragement! I could definitely tell a difference in low, medium, high bias readings. I landed at about 65% for now, but will continue to test.
 

Phrygian77

Poster Extraordinaire
Gold Supporter
Joined
Apr 30, 2016
Posts
5,985
Location
Crawfordville, FL
Princeton: I added a bias pot to the princeton this weekend and tried a couple of tubes to dial things in. I first tried to re-use the existing 27k bias resistor, but that didn't provide enough current: about 8mA-19mA. I changed it up to 10k, and now, I get a sweep of 18mA - 30mA.

So, you used a 10k pot with a 10k resistor? Originally, I would use a 22k "tail" resistor in my builds, but I found that wasn't enough negative voltage for some tubes that want to run on the hotter side. I ended up using a 27k instead. However, that was with a 5AR4 and a higher B+.
 

minestaken

TDPRI Member
Joined
Nov 18, 2017
Posts
32
Location
Michigan
Exactly - 10k pot w/ 10k resistor. Likely splitting the difference would be the best option? I don't have another matched set to experiment with.
 




Top