How DO you bias an amp correctly?

[jezzzz2003]

I thought my amp was properly biased but apparently it's not.
//www.aikenamps.com/whatisbiasing.htm
Aspens Tube Amp Book says otherwise and I have been going from that for years now, never had any problems yet, although Ive had a set of winged C's in my old Marshall for a year now without a hitch and I didnt even bias at last swap, the tubes were previously svets but a different rating though, the amp still sounds even better than it used to for some reason.
Maybe I can bias by ear in the future and watch for plate glow, is this possible??

How DO I properly bias my amp?

[John Phillips]

In purely technical terms, the correct bias point for a Class AB push-pull amp is that at which crossover distortion just disappears for a full-power clean signal, or very slightly above this. The most accurate way to do this is with a scope and a signal generator (and a dummy load, unless you like listening to sinewave test signals at full output volume :)). Usually (depending on the amp design, and whether it uses the full capability of the tube) this will occur at an idle current of around 50% of the tube's maximum rating. You can do it by ear, but it's not very accurate... though it also doesn't need to be.

In musical terms, they don't always sound best like this, and need to be run a bit hotter. But not too hot - you always need to keep the maximum dissipation at any output power below the maximum rated power for the tube, or you will shorten the tube life (possibly a lot, depending how much and for how long you exceed it). In a typical Class AB circuit, this means keeping the idle dissipation below about 70% of the tube's rated value. In some amps, which push the limits of the tube rating, it needs to be less than this. In others it can be hotter, because the voltages in the amp may be low enough to allow that... the 70% figure is not an accurate, or 'magic' target. It just works pretty well for a typical Class AB amp designed to use most or all of the tube's capability.

Basically, the correct bias for a guitar amp is where it sounds best to you, and doesn't exceed the tube's maximum ratings - or if it does, you can accept the shorter tube life.

It's also my opinion that tone changes from biasing hotter (although not non-existent) are often less to do with the actual tone, and more to do with the perception of better tone caused by the slightly increased gain - this is a well-known psychoacoustic effect. I always bias as cool as reasonably possible when repairing amps, because there's absolutely no doubt that this gives better reliability and longer tube life... and I don't get too many complaints about tone.

Plate glow is not usually a good indicator of when you're too hot BTW... it's an indicator of when you're WAY too hot, and the tube is at risk.

[Wakarusa]

In purely technical terms, the correct bias point for a Class AB push-pull amp is that at which crossover distortion just disappears for a full-power clean signal, or very slightly above this. The most accurate way to do this is with a scope and a signal generator (and a dummy load, unless you like listening to sinewave test signals at full output volume :)). Usually (depending on the amp design, and whether it uses the full capability of the tube) this will occur at an idle current of around 50% of the tube's maximum rating. You can do it by ear, but it's not very accurate... though it also doesn't need to be.



I tend to advise strongly against using crossover distortion as a bias method. John's absolutely correct in his definition of correct bias for class AB1 operation, but it also assumes an ideal output transformer.

Slight mismatches (due to design, age, or the phase of the moon) in the output transformer primaries of a push-pull design will introduce some crossover distortion that simply cannot be eliminated with tube bias. More than once I've seen folks bias way too hot because of this.

[John Phillips]

Slight mismatches (due to design, age, or the phase of the moon) in the output transformer primaries of a push-pull design will introduce some crossover distortion that simply cannot be eliminated with tube bias. More than once I've seen folks bias way too hot because of this.
True! Perhaps it's just that I've learned to tell the difference between tubes-biased-too-cold crossover distortion and inherently-imbalanced-power-section crossover distortion :). The first tends to be symetrical about the zero crossing and sounds 'harsh', the second tends not to be and sounds 'fuzzy'...

[jezzzz2003]

Is there a propper method of biasing then or is biasing something alot of techs do not perform right?

My father is an amp technician and has been for 30 odd years, he has always performed the oscilloscope method with my amp which reads crossover distortion and notches it out.

since this method is not very accurate I gues all his customers who came in for a change of tubes over the years really havent had propper service?

Is biasing really that important provided they are operating within a safe range as long as the amp sounds its best?

[John Phillips]

Is there a propper method of biasing then or is biasing something alot of techs do not perform right?Often not performed right... possibly because the current 'accepted method' - biasing to 70% of maximum dissipation, often by using cathode (not plate) current, irrespective of other circuit parameters - is actually wrong... or at least, not always right.

My father is an amp technician and has been for 30 odd years, he has always performed the oscilloscope method with my amp which reads crossover distortion and notches it out.

since this method is not very accurate I gues all his customers who came in for a change of tubes over the years really havent had propper service?Not at all, he's been doing it exactly right, providing he knows (and compensates for) the difference mentioned by Todd above (if it occurs). And IMO, as long as he listened to the result too... too many properly trained electronics engineers go purely by the book without either listening, or in some cases understanding how musicians will actually use an amp - ie thrashed at full power, beyond the point of distortion :).

Is biasing really that important provided they are operating within a safe range as long as the amp sounds its best?IMO that's all that matters. I tend to err on the side of tube life if I have the choice, but that's about it.

[Wakarusa]

Is there a propper method of biasing then or is biasing something alot of techs do not perform right?

My father is an amp technician and has been for 30 odd years, he has always performed the oscilloscope method with my amp which reads crossover distortion and notches it out.

since this method is not very accurate I gues all his customers who came in for a change of tubes over the years really havent had propper service?

Is biasing really that important provided they are operating within a safe range as long as the amp sounds its best?


I think it's best to split the answer into methods and results. The crossover distortion method is a "correct" method, but it can easily be used for poor results. Same is true, of course, with any method.

I'll go further and say that any result that leaves the power tubes biased below their maximum rated thermal dissipation is "correct". "Correct" biases can be further limited by class of operation, but that still leaves a pretty big range of correct settings.

"Optimal" results are anything that falls into "correct" that sounds good -- a compromise between tone, tube life, etc. Really a personal decision. So Dad using a crossover distortion method is fine as long as the result is below rated dissipation and achieves class AB1 operation. :)

[Wakarusa]

IMO that's all that matters. I tend to err on the side of tube life if I have the choice, but that's about it.

Curses! Beaten to the punch again!

[jezzzz2003]

I like, I like.
:BEER

[A positive]

Sorry for the dumb question but how crossover distortion sounds like??

[Wakarusa]

Sorry for the dumb question but how crossover distortion sounds like??


John pretty much nailed it with "harsh". Other adjectives that work are "cold" and "edgy". To my ears it gives the notes an abrupt, hard quality instead of the smooth, rounded quality we'd expect.

The only other way I can think to describe it is "Musicman RD-50".

[A positive]

Ok thanks!! :)

[reaiken]

If you used the crossover distortion method with this amp, it would burn up the tubes in short order:

http://www.thegearpage.net/board/showthread.php?t=193340

Likewise, if you use the crossover distortion method with a true class A amp, it will be biased way too cold, and will sound like crap.

Crossover distortion biasing is used for "properly" designed class AB amps to limit the active device dissipation for best output stage efficiency. It is primarily useful in solid-state output stages or tube stages designed for "proper" AB operation. Guitar amps are a completely different animal. For starters, your carefully adjusted-out crossover notch comes right back when the amp is cranked up to the point where the output stage distorts. In addition, they aren't always designed to be biased to the point where the crossover notch disappears. The 500V Marshall is a good example - it really should be biased class B, because there is no bias setting that will keep the tubes inside the safe operating area, and the crossover distortion method will bias it way too hot. Who knows, maybe it was originally designed to be a class B amp?.

Guitar amps are not designed to be highest-efficiency, lowest distortion linear amplifiers, they are tone generators. Bias is just another tool for the designer/tech/player to set the amp where it sounds best. Yes, this goes against "conventional" engineering, but so does running the output stage into full clipping!

Randall Aiken

[conundrum]

The only other way I can think to describe it is "Musicman RD-50".


:rotflmao

You are my new personal hero.

[John Phillips]

If you used the crossover distortion method with this amp, it would burn up the tubes in short order:

http://www.thegearpage.net/board/showthread.php?t=193340


Crossover distortion biasing is used for "properly" designed class AB amps to limit the active device dissipation for best output stage efficiency. It is primarily useful in solid-state output stages or tube stages designed for "proper" AB operation. Guitar amps are a completely different animal. For starters, your carefully adjusted-out crossover notch comes right back when the amp is cranked up to the point where the output stage distorts. In addition, they aren't always designed to be biased to the point where the crossover notch disappears. The 500V Marshall is a good example - it really should be biased class B, because there is no bias setting that will keep the tubes inside the safe operating area, and the crossover distortion method will bias it way too hot. Who knows, maybe it was originally designed to be a class B amp?
It's funny you should say that, because I learned how to bias amps by this method... by reading Marshall's service instruction manual, circa 1986.

:eek:

:)


I agree about justonwo's amp though, and many other old Marshalls I've worked on... you just have to try to compromise between several factors to get them to run safely and sound good.