Bias settings on ampifiers...

C Dixon · Post by **C Dixon** » 26 Jun 2004 4:37 pm

This subject appears on this forum a lot. And much rhetoric has been expended and hashed and rehashed. Been thinking about it since the last post. And I would like to share with those of you who might be interested.

Note: this might be a bit too technical for some. If so, you may wish to exit out now. I fully understand.

For any who might elect to learn more about bias, and the why's and wherefore's about it, I thought we could discuss it in a little different way.

In other words, instead of "cliche's" like, set it for minimun hum or "adjust it 'til it sounds the best", I would like to describe first what it is and remove some of the mystery, if you will; and then explain some of the anamolies concerning bias.

First the name bias. I am not sure of the source of the word. Not even sure it fits; nevertheless it has been one of the basics since Dr Lee DeForest put a third element in a diode vacuum tube and called it a triode. Or the beginning of the world's first electronic amplifier.

Biasing a tube is a method of controlling the amount of current that a tube draws with NO signal present at its input. It is done with certain specifics in mind.

Think of a venetian blind for a moment. When it is wide open, maximum light (saturation) is allowed to flow thru it. When it is closed, no light (cutoff) flows thru it. Note: I am talking about a perfect venetion blind, not the el cheepo's that one buys at Walmarts.

The way a blind controls the light is set by a wand that adjusts the slats to allow more, or less, or maximum, or no light to flow, depending upon where the slats are set.

The wand is the "bias" of the blind. The object is the amount of light flow. On a tube the "bias" sets the amount of current that flows thru the tube when no signal is present. From nothing (cutoff) to maximum (saturation) with a given power supply and other associated parts connected to that tube.

Ideally, the rate of change in current flow in a tube would be even and smooth (linear) all the way from cutoff to saturation. Sadly, that is NOT how any tube or transistor works.

Rather it is non linear leaving cut off, turns linear for about the next 80% of the way, then turns non linear again as the tube approaches saturation.

Because of this anamoly, electronic engineers have had to deal with a built-in nightmare from day one.

In a class A amplifer, the tube is "biased" so that with NOTHING coming into it (no signal) the tube is operated at precisely half way up the linear portion of the above anamoly, called "characteristic curve".

This way, the tube can increase current or decrease current linearly depending upon whether the incoming signal aids the bias or opposses it. As long as the signal does not drive the tube into the non linear part near cutoff (or saturation).

Now in a Class B amplifier, the tube is biased at cutoff. In a Class C, the tube is biased well beyond cutoff. Now the bad part of this is, in BOTH cases (B or C) this biasing would cause intolerable distortions of all kinds in any amplifier.

This is due to the non linearity characteristics close to cut off before the linear part begins. Or beyond cutoff in Class C where the output would not only be extremely distorted, but it would not even contain all the signal.

Now class A has a problem also. First of all, it wastes mucho power. Because it is always ON! Also, the amount of amplification is ONLY half what if could be IF there was a way to bias it where the linear part starts and uses two tubes to cover both halves of the input signal.

Then we could double the amount of amplification without having to build a power supply twice as large.

So how can we do this? Very simple. Bias at least two tubes so that the tubes begins operating JUST at the linear portion. So this means, a bias pot is necessary; because tubes even of the same brand and type can have slightly different curves.

This method of biasing is between A and B. Thus the Class AB label. So, biasing means; we are setting the push pull output tubes so they are NOT cutoff; rather they are conducting a small amount of current, just enough, so the tubes NEVER are operated in the non linear portion of its characteristic curve (near cutoff).

Having said that, what is class AB1 and AB2? Ok, Engineers found a way to squeeze EVEN more power, everything else being equal, IF they biased the tubes partly IN the non linear portion, without having the extreme distortion appear in the output.

And how did they achieve this? Ok, by a system of feeding back a tiny portion (the non linear part) back into the input in opposite phase so it would cancel out the distortion. Clever huh?

Indeed. And that is precisely what they did. They called it AB1 biasing; or simply class AB1. Tinkering with it further, engineers discovered that most ears can tolerate (and even like, in some cases) a tiny amount of crossover distortion*. So they adjusted the bias even further towards cutoff to take advantage of this quirk to gain even more power. This is called AB2.

And there you have it folks. ONLY one big problem. MOST manufacturers are negligent and very bad about NOT telling us what the proper procedure is to adjust this bias pot. And this is indeed sad.

I have been in the elecrtonic business for most of my adult life. I have taught it for over 30 yrs in my carreer. And I have yet to see (there maybe one) a set of instructions that told the technician the proper way to adjust it!

So it ends up with high tech "good ole boy" technology. "Well I set it for minimun hum." "Well I don't, I set it for 30ma of current." Well I don't, I just set it 'til it sounds the best"

To WHOSE ears?

, And on and on, Ad infinitym.

Oh well,

carl

*Note: "Crossover distortion" is where one tube is shutting down and its push-pull mate is just starting to turn on, once signal is applied. This is a critical point. This is why I have had a 40 yr pet peave with amplifier manufacturers.

TELL US THE PROCEDURE! SO WE ALL CAN GO DOWN THE SAME ROAD; IT'S FOR THE CUSTOMER, DUMMY!!

[This message was edited by C Dixon on 26 June 2004 at 05:58 PM.]

Jim Florence · Post by **Jim Florence** » 26 Jun 2004 4:49 pm

Thanks Carl, that's the best explanation I've heard on the subject.
Jim

John Daugherty · Post by **John Daugherty** » 27 Jun 2004 3:21 am

Excellent work Carl. I am sure there are many grateful readers who thank you for your time.
I have made my living from elex since 1958(except for a few years in the army and the music business) after gradusting from Ranken Tech in St.Louis. While working in a government research lab, I was taking courses from Devry.
When I first went to school, all we studied was tubes. I am wondering if the younger electronics majors are taught enough about tubes to be able to write the "bias adjust procedure". It takes a very wise and knowledgeable person to write operating manuals and service manuals. Most manuals leave a lot to be desired.
While we are on this subject I want to give a tip to all the tube users: A "cathode emission" tube tester is,in my opinion, just about worthless. If it shows the tube is bad,it is probably bad. If it shows the tube is good, it can still be bad. When I serviced tube equipment I checked tubes by substituting a "known good tube" to see if it help the performance. A "transconductance" tube checker injects a signal and measures gain but not necessarily under the same conditions as the equipment your tube is used in.
I say "Don't put a lot of faith in tube checkers".
One more service tip: Some amps do not have a "hum pot". They have 2 resistors. One from each side of the transformer filament winding to chassis. If one of these resistors burns and opens or changes value, the amp will hum. I have seen this in Fender amps.
I hope I haven't bored you by getting off the track. I just want to add to this already informative discussion.

Gino Iorfida · Post by **Gino Iorfida** » 27 Jun 2004 4:26 am

Thank you for taking the time to explain what biasing is by the way (saved a lot of typing).

A couple poitns I'd like to add to this discussion:

1) As Carl mentioned, bias controls the current flowing through the tube-- if you let too much, the plates will glow red hot and will burn up, not enough and the tube will be cutting off... another analogy would be your car... think of it idiling in neutral (no signal present in amp)... if you floor the gas and redline the engine for any length of time, you are putting in a new engine... if the idle is set too low, the engine will run roughly etc...

2) Hum balance pots in some of the silverface fenders:: These amps were set up from the factory to (this also applies to amps with NO bias setting controls -- most amps they are internal btw, to keep the customer out of where they can get fried!) run almost all current production tubes within their specs-- likewise most of these amps run the tubes rather on the cold side... The silverface fenders 'bias' control controlled whihc side of the push pull pair got the most current (or ti was balanced - allowing unmatched tubes to run happily), the hum balance affected the current running through the filaments-- this is NOT a bias control-- it has nothing to do with the plate current etc -- in fact deals more with controlling things in the preamp tubes!

3) how I bias an amp -- first I look up the tube spec sheet for the power tubes used, and find the maximum dissipation for the tube used. This is listed in WATTS, so I measure the plate voltage of the tubes in the amp i"m working on (never go by the schematics-- ther are variances in parts that can cause these numbers to not be exact!), and divide the max. dissipation by the plate voltage, and that will tell me how many amps (actually milliamps

these tubes are running at. Generally, most that I've talked to and all I've read suggests running the tubes at about 70% max dissipation for amps that are played clean, and a bit lower for amps that are being pushed into overdrive (not talking amps with master volume etc, talking blues style caranking the volume until it hurts thing

. I measure the current flowing thru the plates (I use a transformer shunt method -- hooking one lead of my meter to the plate lead of one tube, then the center tap of the transformer-- the internal resistance of the meter is so small you can get a fairly accurate number this way)... I also make sure to measure this on BOTH sides of the transformer (transformers, tubes etc are not perfect, and it's normal to get a small variance betweeen sides)-- I always use the side that is running the hottest as my reference-- always better to be safe than sorry -- also a good time to note, in amps that use 2 tubes per side (i.e. 4 power tubes total), the number you read will be DOUBLE what you want to see -- remember this transformer is feeding 2 tubes this current per side). (b) adjusting:: if the tubes are too hot, I turn the adjustmetn pot to feed more bias current,which should lower my current draw. If they are too cold, I do the opposite. After I think I have them in a safe range, I start my measurements over--note, you will HAVE to take the plate voltage reading again, and recalculate -- as the current draw goes up, the plate voltage can creep down and vice versa...
-- one caveat of the type of biasing i describe, you NEED a good meter to bias with (Fluke etc).. cheaper meters have a higher internal resistance, and that will mess everything up.

Jay Fagerlie · Post by **Jay Fagerlie** » 27 Jun 2004 7:29 am

Bravo to each and every one of you guys.
The whole biasing thing has been a sore subject for me for a long long time.
I think it would be a good idea from now on that when a post comes up about biasing, they should immediately be fowarded to this post.
Together, we can rid the world of bias-ignorance!

Jay

Stephen Gambrell · Post by **Stephen Gambrell** » 27 Jun 2004 6:29 pm

Can anybody give me ONE good reason why Bobby Cox shouldn't be fired, and replaced with Carl Dixon as manager of the Atlanta Braves?
Thanks, brother Carl, for another informative, and excellent post!

Jay Fagerlie · Post by **Jay Fagerlie** » 28 Jun 2004 5:37 am

I can

He wouldn't have any time to post here on the forum.

And we just can't have that!

Jay

Ray Minich · Post by **Ray Minich** » 28 Jun 2004 9:28 am

Venetian blind... Carl you are the master of the metaphor

Excellent!

Jennings Ward · Post by **Jennings Ward** » 28 Jun 2004 5:45 pm

Men , for a good bias job, I prefer to add a [1] one ohm cathode resistor.., measure the voltage across it, which when extrapolated will give the plate current in ma. milla-amps. takes more work and a little longer,but by far more accurate, and saves parts and time trouble shooting. And it will reduce callbacks expoentually...Carl do you agree and you outher techs?
Eight 6L6 tubes or their equiv. are dangerous as can be . even a pair will lite you up and give you a bad burn..Hope this helps someone..Note: be careful if you are not trained in electronics.... Jennings

------------------
EMMONS D10 10-10 profex 2 deltafex ne1000 pv1000, pv 31 bd eq, +

Jay Fagerlie · Post by **Jay Fagerlie** » 29 Jun 2004 5:45 am

Yes Jennings, the 1 ohm resistor on the cathode is the way to go. I add these plus a tip jack to all of my amps for easy bias testing/adjusting. Makes life easier.
The shunt method works well too, but not for the un-initiated.
Last week I was setting up an amp using the shunt method and managed to get bit by the B+....OUCH!!! It sure woke me up.
That's why I ALWAYS try to have one hand in my pocket (in this case, there was no way-I had to hold two probes....I think I need the grabby-type probes)
The lesson here is: Pay attention, this stuff CAN KILL YOU!!!!

Jay

Ray Minich · Post by **Ray Minich** » 29 Jun 2004 6:05 am

Check out the POMONA product line for grabby type probes. I've got a bunch of different ones and they've saved my butt a few times.

Eight 6L6's running at the same time in the same chassis... Good place to warm a TV dinner

Jay Fagerlie · Post by **Jay Fagerlie** » 29 Jun 2004 6:44 am

Thanks Ray,
Those are Ed Zachary what I was talking about
Time to pull out the Newark catalog.....

Jay

C Dixon · Post by **C Dixon** » 29 Jun 2004 8:29 am

Jennings,

I am in agreement on the 1 ohm thingy's

It has been a long time, but it seems to me that as I was leaving the business, that Techtronics Corp was coming out with a DC "amprobe" probe for their famous scopes.

If I was not dreaming (which is highly possible

), this would be a blessing for setting bias. Imagine just opening up the tiny jaws on that probe, clamping them around the cathode ground wire, and whamo, you have a picture of the actual current flowing thru a given tube! Amazing huh?

Believe this or not, but in my last year of "working for RCA", I used ONLY a scope, EVEN for DC voltage readings. With the advent of the DC scope and an instant "picture", it opened up things to me that I "knew" but never realized.

I will give you an example, I always believed as taught (and many others believe) that when one measures the highly negatve DC voltage on the grid of an oscillator tube that that there are two voltages present; AC and DC. This is NOT true. The DC voltage turns out to be nothing more than a limitation of one's meter.

Also there is NO AC voltage there. What there is; is a varying DC voltage that is going more or less negative with respect to the cathode.

It was NOT until I started using the DC scope exclusively that it really hit me. To this day I can't get over what a revelation that was. I would have sworn that the "bias" voltage my meter was showing was fixed and the "cash register" wave I saw in an oscilloscope was superimposed around that DC bias voltage. NOT so!

Finally, a coupla things:

1. I am humbled by the kind words expressed in this thread.

2. It is unlikely that B+ voltages would actually kill a person. But it can often make you wish you were dead. It is interesting to note, that a person COULD be killed instantly touching the 110V AC line voltage coming in; and yet get hit by 500 volts of DC B+ votage a number of times and only come away with "shock"; physically AND mentally

I been hit by it dozens of times, and I NEVER got used to it. I have also been hit by 30,000 volts on the anode of a picture tube enough times to make a preacher cuss. And it will indeed wake any person up and quick.

But the worst hit I have ever taken, is the "RF" hit around the low end of the Hi voltage damping circuit of a TV set. It not only makes you feel like a Fork lift had bolled you over, but it even leaves a burn mark on your skin that is about as painful as one can stand. One time it knocked me across the room. The RF voltage at that point is usually around 4700 volts of Varying DC; called "RF" because it is varying at 15.75 KHZ. It actually burns the skin and leaves a nasty sore.

Note: for those who do not understand why 110V AC line voltage can kill, and 300 times that much will not kill, has to do with the current delivery capibility of B+ Circuits and HI voltage circuits (in TV sets).

The reason is the human body will load those circuits down instantly and they are NOT capable of delivering thru the body even the very critical .5 amp of current necessary to kill. But again, it will make one wish they were dead for a while.

Also, remember voltage has NEVER killed anything. It is the current flowing thru something that is the "killer". Think of it as water pressure and the water flow itself.

Water pressure=voltage

Water (itself) flowing=current.

Again, thanks fellows,

carl[This message was edited by C Dixon on 29 June 2004 at 12:14 PM.]

Jennings Ward · Post by **Jennings Ward** » 29 Jun 2004 12:12 pm

Carl, it is good to know someone understands me..some dont. Yes an O'Scope was and is my favorite test and troubleshooting inst. T'Tronics, especially, and HP as well. Yes I have been knocked down and stomped, because I was careless and not attentive to what I was trying to do...If I am not mistaken, Raytheon made a Color TV that had, Get this, 2 Daamper tubes. 2 horiz outs, and 2 vert . outs...The worst part was, A METAL BELL ON THE CTR OR FUNNEL AS SOME CALLED IT. THE PLAT CURRENT ON THE HOP WAS ABOUT 1.5 AMPS AT 45KV..... I owned one. not for long though. small kids, too dangerous.... My esperience was with your competitor, Motorola , later Zenith....lots of fun. I say that to say this, an audio amp is the simplest of all the consumer and industral products, because you only have one signal to deal with. Digital is a little different , but basically the same...still fun... YU'aLL have fun, best reguards to all ,,,,,,,,Jennings

------------------
EMMONS D10 10-10 profex 2 deltafex ne1000 pv1000, pv 31 bd eq, +

Gino Iorfida · Post by **Gino Iorfida** » 29 Jun 2004 1:08 pm

Cathode 1ohm resistor> I have a few problems with this as well as good things to say:
GOOD: Safer to work with, easier to deal with etc-- can also be brought out to test leads etc.
BAD: 1)YOu are not just reading plate current, rather the sum of the plate + grid currents -- in other words, your readings will be off a bit, luckily, the readings will err high-- in other words, the amp may be biased a bit cold-- cold == safer, but the tone may or may not be the best
2) I suggest using a 10 ohm resistor, since a 1 ohm resistor is so small, the tolerance may not allow for accurate readings-- not to mention, most meters can't evne accurately measure the resistance of the resistor. If you use a 10 ohm resistor, even if it reads at 9 or 11 ohms (10% tolerance-- I recommend high quality, 1% or better resistors!), you have a better chance of taking more accurate readings... and yes, I always advocate reading the resistance of the resistor BEFORE taking the voltage measurements (amp off, tubes out etc), and do the math, in other words, IF AND ONLY IF, the resistance is 1 ohm, the voltage and current will be the same... if the resistor is exactly 10 ohms, the current will be 1/10th (move one decimal point) of what the voltage will be... if the resistor measures anythig else, the current will be the voltage divided by the resistance (ohms law

Ray Minich · Post by **Ray Minich** » 29 Jun 2004 1:58 pm

Carl, did you ever take an audio output transformer from an old tube radio, and hook the secondary up to 110 volts. I did, as a kid, and I still have the scar on my thumb where the arc from the tranformer primary passed thru it. Fun days...

Funny thing, the transformer only arced for second then burned up

[This message was edited by Ray Minich on 29 June 2004 at 02:59 PM.]

C Dixon · Post by **C Dixon** » 29 Jun 2004 5:48 pm

No Ray I haven't. But it makes sense

In traveling the country teaching technicians from every walk of life, I found it sad just how few EVER used a scope. Course it was not all entirely their fault. The manufacturers usually did us no favors by not giving us accurate waveforms and gain specs, etc. If at all.

Even sadder was the high number of techs that had NO electronic training that were working on sophisticated products; with absolutely no knowledge at all of the product they were working on, let alone how to adjust a bias pot or any other adjusment.

We called them in the business, "parts changers". When those of us who were given the task of "making real techs out of them" and it was apparent some of them were just NO cut-out for the job, their managers would more often than not become incensed when we pointed this out. It was NOT uncommon to hear the following,

"They may not be have a PHD (as if any PHD would work on a lowly TV; lol) in electrical engineering, but they can sure sell the heck out of service contracts. So I will trade them for any "tach-nishun any day".

Interestlingly, we had no intention of trying to make them PHD's or even engineers. We would settle if they could EVEN read a meter. Which sadly many couldn't. Even sadder is they still can't. But do they ever make the money selling "them service contracts".

But the saddest part of all was the low paid technicians spent most of their time undoing the butchering the high paid "selling techs" had caused.

Oh well,

carl

Ray Minich · Post by **Ray Minich** » 30 Jun 2004 6:19 am

A METAL BELL ON THE CTR OR FUNNEL AS SOME CALLED IT. THE PLAT[E] CURRENT ON THE HOP WAS ABOUT 1.5 AMPS AT 45KV...

One of the reasons for the metal bell was that above 28.5 KV acceleration voltage a picture tube can become a real nice x-ray generator. I always thought TV techs should carry a single exposure of Polaroid Land Pack Film, in a light-tight carrier, in their shirt pocket, and then develop it at the end of the day to see what the streaks look like.

Jay Fagerlie · Post by **Jay Fagerlie** » 30 Jun 2004 6:33 am

Gino,
The 10 ohm resistor is an option, but my main reason for using the 1 ohm is that it reads out directly without computation, ie, 39 millivolts on the meter equals 39 milliamps of current.
I use 1% resistors so I don't think there is a tolerance problem.
In a typical circuit, how much grid current is there? Is there actually enough to skew the readings?
Interesting....

Jay

Ken Fox · Post by **Ken Fox** » 30 Jun 2004 6:48 am

I use a Bias King Pro, two sockets for tubes and a digital read out for both via a switch. Weber has an even better setup that now had the plate voltage reading as well. You can get all the readings and never open the amp up!
http://www.webervst.com/bias.html

Bill Llewellyn · Post by **Bill Llewellyn** » 30 Jun 2004 7:00 am

My brother used to work at a TV repair shop in his teens. He (also a ham radio operator) would tell me stuff about the insides of our Magnavox color TV set in our living room...like the x-rays given off by this big old rectifier tube that was partially housed in a metal casing. I remember the rule of thumb was not to spend too much time directly behind an operating color TV set because that's the direction most of the x-rays went.

I never got to know very much about tubes. When I started my BSEE schooling ('73) that part of the cirriculum had just been phased out. I have worked in audio amplifier design (7 out of the past 8 years) but that involved class-D amplifiers.

------------------
Bill, steelin' since '99 | Steel page | My music | Steelers' birthdays | Over 50?

Ray Minich · Post by **Ray Minich** » 30 Jun 2004 8:35 am

Bill, that probably would've been the 1B3 in the high voltage section.

My dad was "operating" on an older RCA color TV a few years ago, and the HV probe, when stuck under the anode connector, was reading 39KV. Turns out, one of the "orangedrop" caps that helped set the Horizontal oscillator frequency had gone bad (open) and the oscillator was operating at twice the frequency. We shut that thing down right away before somebody ended up "glowing in the dark".

C Dixon · Post by **C Dixon** » 30 Jun 2004 11:33 am

""glowing in the dark".

Oh Ray, does that ever bring back memories. I used to do a little trick for mostly younger students (but occassionaly some old ones too) back in the lab after teaching them how the High Voltage circuit worked in a TV set.

You mentioning the 1B3* sparked my memory. I would gather the class around an old black and white TV set; and with my left hand holding the center of a 4' neon bulb up in the air, touch the metal plate cap of that HV rectifier using my right hand with a pair of uninsulated long nose pliers. Yes, UN insulated!

When they saw that neon bulb light up, you can imagine the looks. But the ones that were aghast were the techs who had been working on them for years.

As some of them left shaking their heads, I could often hear comments like,

"carl actually touched the plate cap of that High voltage tube with the set ON! NO way would I ever do that!"

Course what I never told them was, it took everything in me to ever try it, the first time I saw someone do it.

For those of you who are familiar with electricity, my left hand was holding that neon bulb in free air space, my body of course was floating with the voltage on that plate cap. At that point it was about 40,000 volts AC with respect to ground.

Therefore I was in no danger of getting shocked. EVEN though, as the long nose pliers approached the cap, a sparkling fat 1" blue electical flame ensued between the tip of the long nose and the cap. But knowing that did not lesson the horror and fear that was exuded by those older techs.

Incidently for those of you that would like to understand it better. The exact same thing happens when birds land and rest on bare 130,000 volt hi tension wires strung across super highways. (no flame at 60HZ)

They are simply only touching ONE side. IF any part of their flesh would even hint at touching ground, they of course would be electrocuted instantly.

But you brought about some

nostalgia to me Ray.

Luv ya for that. I needed a good laugh dear friend,

carl

*You would be surpised how many many customers would come in and ask for "one of them I-Bee-three tubes". I would often tell them that there had been a change and they ar being replace by the newer,

"2 times 2 tubes" (2X2's)

[This message was edited by C Dixon on 30 June 2004 at 12:41 PM.]

Ray Minich · Post by **Ray Minich** » 30 Jun 2004 6:12 pm

Carl, the little standoff capacitors that the 1B3 tube socket stood on (it was a small round brown hard cylinder about 1 inch diameter, about 3/4 inch high, and had an 8-32 stud sticking out of each end) made neat little wheels for a bench cart

Jennings Ward · Post by **Jennings Ward** » 30 Jun 2004 6:21 pm

Carl that bulb trick reminds me of how smart we were and sure could do some dumb things....Ever get bit by a degausing capicator? Have a 6BK4GT go bad while you were working on the chasis? With a CRT Jig? Try to do a convergence job after a Christmas Party? Memories..Must be a reason for my heart trouble..... The good Lord takes care of dumb animals and fools.. In God we trust, every one else pays cash....BBest to all,,,, Jennings

------------------
EMMONS D10 10-10 profex 2 deltafex ne1000 pv1000, pv 31 bd eq, +