Setting Bias: It Isn't Easy Sometimes But It Makes All The Difference

PreCd

I have been playing with the bias on my Adcom GFA-555 for a few weeks now and have learned a few things about it. It makes all the difference in the world in your sound if it is not set correctly. The audio can sound fuzzy or distorted depending how much the current is off. It may even affect sound stage which was the case for me. I have a procedure for setting my amplifier bias but found that if I set it at the level specified in the manual it doesn't cut the mustard.

What I ended up doing was setting it at the level specified which was 10mv for my amp after warmup. If I left it there it did not sound good. Slightly Fuzzy. The instruments and vocals did not sound the same coming through the speakers as they did the headphones. So I dropped it down a bit. Listened for awhile then adjusted a bit more. I finally settled it in at 6mv. Now some may say well that is only 4mv....come on what difference can that make. Well if you take a critical listen it can make a lot of difference I found. The fuzziness is all but gone. I mean I have listened to the aja album so many times that I can tell the parts that show up fuzzy on my system and on a scale of 1-10 it may be a .05! I give the rest to having a solid state system! Now the fuzziness I am speaking about is not something that just jumps out at the normal listener. You really have to pay attention but when you hear and then you remove it it is joyful. And the sound stage opened up quite a bit also. That was possibly the most satisfying part of the whole experiment.

I even had my current set at 5mv for a few weeks and noticed the amp sounded worse as it warmed up. On my amp the bias current is higher when cold then settles down when warm. So I upped it a mv and it made a difference. I may listen to it at 6mv for awhile and try 7mv just to see if it gets better. I don't think it will but it is worth a try.

So if you have a few hours on your hands and know what you are doing give it a whirl. Take music that you are very familiar with and listen to it as you tweak your bias current. I bet many of us out there have bias current that is out of whack and just tweaking it could give you a brand new nice sound.

dragon1952

PreCd wrote: »

What I ended up doing was setting it at the level specified which was 10mv for my amp after warmup. If I left it there it did not sound good. Slightly Fuzzy. The instruments and vocals did not sound the same coming through the speakers as they did the headphones. So I dropped it down a bit. Listened for awhile then adjusted a bit more. I finally settled it in at 6mv. Now some may say well that is only 4mv....come on what difference can that make.

While 4mv may not be a lot of current, that is a 40% change from the recommended setting so actually is it a lot.

FTGV

Interesting as one would expect an increase in bias to sound better than a decrease since higher bias increases the output level at which it operates in Class A before it transitions into Class B.Ofcourse increasing bias beyond the recommended manufacturer level is risky as thermal limits of output devices can be exceeded ,potentially leading to premature failure.

PreCd

I understand your statement and thought about that when experimenting. Let me ask this. Wouldn't the scenario below depend on the design of the amplifier?

FTGV wrote: »

Interesting as one would expect an increase in bias to sound better than a decrease since higher bias increases the output level at which it operates in Class A before it transitions into Class B.Ofcourse increasing bias beyond the recommended manufacturer level is risky as thermal limits of output devices can be exceeded ,potentially leading to premature failure.

FTGV

I can't see any scenario where reducing the level of Class A operation would be sonically or from a objective meaurement stand point benificial.With lowerthan optimum bias levels crossover and other forms of distortion rear their ugly head which usually leads to a cold,sterile hard sounding amplifier.

PreCd

Cannot explain it. Only can say that the procedure that I followed in my original post obtained better sound than I had when I started with the higher bias current. I will also say this. I am not pushing for anyone to lower their bias current. What I am saying is that it would be beneficial for those who have not looked at the bias on their amp to do so and experiment a bit with their ears. Especially vintage gear.

FTGV

While it's counter intuative if you find the lower bias level preferable then run with it.Reliabilty certainly will not affected in a negative way.

PreCd

Have we taken into consideration that a vintage piece of gear more than likely would contain components that would not be as close to spec as when new? That could be a difference maker and could explain why a different bias current, be it lower of higher, would be ideal.

FTGV wrote: »

While it's counter intuative if you find the lower bias level preferable then run with it.

FTGV

Heres an interesting read from the designer of your 555. http://www.passdiy.com/pdf/leaving_class_a.pdf

PreCd

Thanks for the link. I will read over that when I get some rest! Tomorrow.

Boys and Girls.........check your bias current.

PreCd

Read the link FGTV and it was informative. Thanks.

I do understand the correlation between current and the operations of A, AB and B type amplification. There has to be a cutoff in current to have ideal linearity at some point. For instance we cannot keep adjusting the current upward and expect the circuit to perform better just because we are feeding it more. If that was the case the manual would not have a spec for bias voltage. I have spoken with Chris many times in the past and this is what he has said about setting bias current for the GFA-555. The procedures in the manuals are flawed! The procedure he outlined for me a few months ago took more than an hour to complete and even that did not net me the results that listening and adjusting has. It got me close though.

Again I would like to state that with age components degrade over time and the bias current spec also changes. Tube guys can attest to this fact for sure, more so that transistor users.

My two cents worth.

heiney9

This is the first time I have heard anyone recommend lowering bias in a push/pull circuit. If it's because of worn components, replace the worn components. If after reading the article by Nelson Pass you still don't think it's a poor idea then I don't know what to say, except if you like it that way, great.

But, I wouldn't go around telling people to "check their bias" and recommend lowering it unless it's way too high out of spec or too low out of spec.

H9

sda2mike

precd...didn't you recap that 555?

PreCd

Read post #6 where I stated that I am not recommending anyone lower their bias. However I am recommending that for anyone that has not checked the bias of their amp needs to do so. In agreement there?

Do I trust Nelson Pass more than I do my knowledge......you betcha. But when is the bias current too much? You must agree that there is a limit no? I am sure he would too. Being a tube guy surely you adjust the bias on your amps when the tubes get some age on them don't you?

heiney9 wrote: »

T

But, I wouldn't go around telling people to "check their bias" and recommend lowering it unless it's way too high out of spec or too low out of spec.

H9

PreCd

Yes I did. Even the filter caps. Why?

sda2mike wrote: »

precd...didn't you recap that 555?

heiney9

PreCd wrote: »

Read post #6 where I stated that I am not recommending anyone lower their bias. However I am recommending that for anyone that has not checked the bias of their amp needs to do so. In agreement there?

Do I trust Nelson Pass more than I do my knowledge......you betcha. But when is the bias current too much? You must agree that there is a limit no? I am sure he would too. Being a tube guy surely you adjust the bias on your amps when the tubes get some age on them don't you?

Too much is usually when you don't have the hardware to dissipate the heat and/or the output devices can't take the heat, again speaking in very general terms as the complexity of the design has something to do with it as well.

I have one tube amp that has manual bias, in two years of using used power tubes in it, I haven't had to adjust it. To be fair I don't play it everyday. Many amps have bias compensation up to a point, but the more complex and the more gain stages utilized it becomes complicated and yes, bias should be checked on an older, well used amp. But I DON'T agree with adjusting the bias to compensate for worn components, if they are worn replace them.

Have fun, tweaking and messing around can be fun. I just don't recommend someone who has little knowledge going in and twisting dials/pots, etc. And it is rare that reducing bias improves the sound, both objectively and subjectively. In fact, IMO, I'd go as far as saying it shouldn't be done. But in the end they are your ears, in your rig so turn away.

H9

PreCd

Here is a snipet from John Broskie's site:


Regardless whether the amplifier is single ended or push pull, solid-state or vacuum tube, class-A or class-AB—to achieve the best performance—all power amplifiers require a precisely set idle current for their output stages. In a single-ended amplifier, too much or too little idle current will limit the potential power output. In a class-A push-pull amplifier, too much idle current will needlessly overheat the output devices and too little will force the amplifier out of class-A into class-AB operation. In a class-AB or class-B push-pull amplifier, too much idle current will add distortion and an inconsistent output impedance due to gm doubling and too little will produce crossover distortion.

We have to admit that there is a place where bias current has a sweet spot and it may require lowering or increasing bias current. I am an intelligent person and know what I hear and tread carefully before jumping into things most times There is more at play here than just set it as high as you can, dissipate the heat, and go with it. We are talking about a AB amp here which does run at lower power than a fully Class A amp (see Nelson's quote below):

Again I have not set it low enough to pull the amp out of Class A. It sounds too good for that to have happened. And as for replacing components that are aged I would rather tweak a variable resistor any day than to spend countless hours and $ replacing components that are aged but not broken. Make sense?

Brock read post #1 where I clearly state that if you know what you are doing check your bias. And what problem would we have if you don't know to learn. Isn't that we have been doing from day one?

heiney9

I don't agree with the John Broskie quote. It's worded very vague and general vs. the specifics of the NP article where he explains what's going on both graphically and verbally.

Have fun, that's all that matters, we can agree to disagree about the fundamentals of what's going on.

It's your rig, do what you want to make it sound good to you.

I was simply reiterating that people who don't know their way around the inside of an amp shouldn't try this--we agree on that :biggrin:

H9

heiney9

Another good read is this one and I think it hits more succinctly the points John Broskie is trying to make. I should say I dislike the John Broskie quote because it's too vague and general, not because it's wholly untruthful. It's not near specific enough since Bi-polar, MosFet and JFet transistors all have different characteristics when bias is applied, you can't generalize like he did in the quote.

https://passlabs.com/articles/the-sweet-spot

H9

FTGV

PreCd wrote: »

Read the link FGTV and it was informative. Thanks.

.

Your welcome it is a good tutorial on the subject.Also I'm for the most part in agreement with the Broskie quote having essentially covered the same points in earlier posts.The main take away is that there is an optimum bias level for each amplifier.

11tsteve

heiney9 wrote: »

Too much is usually when you don't have the hardware to dissipate the heat and/or the output devices can't take the heat, again speaking in very general terms as the complexity of the design has something to do with it as well.

I have one tube amp that has manual bias, in two years of using used power tubes in it, I haven't had to adjust it. To be fair I don't play it everyday. Many amps have bias compensation up to a point, but the more complex and the more gain stages utilized it becomes complicated and yes, bias should be checked on an older, well used amp. But I DON'T agree with adjusting the bias to compensate for worn components, if they are worn replace them.

Have fun, tweaking and messing around can be fun. I just don't recommend someone who has little knowledge going in and twisting dials/pots, etc. And it is rare that reducing bias improves the sound, both objectively and subjectively. In fact, IMO, I'd go as far as saying it shouldn't be done. But in the end they are your ears, in your rig so turn away.

H9

i am completely new to the interior of an amp, but would like to consider myself to be educable. i recently acquired an Acurus A250 and a matching RD-11 preamp. these have a reputation for shrill highs... and while my new RD-194s and the crossovers are not fully burned in yet, there is just a bit of an over-accentuated high frequency. overall, i like the amp. so far.
after doing some reading around forums and such, it is a general consensus that the A250 came out of the factory with a bias setting a bit too low. i have some threads, schematics, and info built up, and, of course will start my own thread when the time comes, but i would like to start learning my way around inside of my amp, possibly even change some caps if need be. here in the middle of Iowa, i have little recourse to having audio equipment serviced without shipping.
i guess the question is, are you dead-set against somebody learning via the forum, or will there be guidance available? i am not one to jump into things until my ducks are in their proverbial row.

thanks

steve

11tsteve

so... no then?

FTGV

11tsteve wrote: »

i guess the question is, are you dead-set against somebody learning via the forum, or will there be guidance available?

Your ? was'nt directed at me but IMO you could get useful input and guidance from individuals that actually have practical experience doing such procedures (and not just parroting something they have read in a wiki or on a manufacturers web site).

PreCd

Steve:

I have no experience with that amp. If you don't get an answer here it is because no one has experience either. Post your question over on DIYAudio or AudioKarma. You may get someone there who has that amp or experience with the amp.

11tsteve wrote: »

so... no then?

lanchile

heiney9 wrote: »

This is the first time I have heard anyone recommend lowering bias in a push/pull circuit. If it's because of worn components, replace the worn components. If after reading the article by Nelson Pass you still don't think it's a poor idea then I don't know what to say, except if you like it that way, great.

But, I wouldn't go around telling people to "check their bias" and recommend lowering it unless it's way too high out of spec or too low out of spec.

H9

Best post! I agreed 100%.

PreCd

Whew. How quickly things get blown out of proportion......

lanchile wrote: »

Best post! I agreed 100%.

11tsteve

thanks for the thoughts. i just joined DIYAudio a couple of days ago and is actually where i am getting some history on this amp, so i'll look into that further... all i want is to learn some and be more self sufficient. it was a ton of fun recapping my Xovers and modding my 11T's... it would be nice to be able to take care of my amp as well.

[Deleted User]

Hello 11steve,
Here is a copy of a recent blog article from Paul McGowan (the "P" of PS Audio) that might be of interest:

"A bias right in the middle
How boring if you’re biased right in the middle eh? You don’t lean to the left or the right, you’re just middle of the road for everything – as if you don’t actually have an opinion. Well, in audio amplifier designs that’s exactly what you want – a totally neutral middle-of-the-road stance when it comes to music.

Several of you have asked me to expand this little mini series on tubes and transistors by one day to help understand bias, what it means and how it works. Understand that this is actually a bigger subject than I can comfortably cover in a small daily post but I’ll do my best – and to do this let’s design a single ended amplifier together – right here in this post.

Remember in our understanding of the different types of devices we have two: voltage devices (tubes and FET’s) that are always on and we have to turn them off – and current devices (bipolar transistors) that are always off and we have to do the opposite. With this understanding in mind we can now easily appreciate that each of these devices can be used in one of two ways: as either a switch (on/off) or an amplifier (moving between the two extremes of on/off). To make a tube or FET a switch we need to simply turn it all the way off or let it go back to all the way on – and then the opposite for the bipolar transistor. Just as a side note – in the late 1940′s early computers used vacuum tubes as electronic switches requiring huge rooms and building full of them to perform even simple calculations. The term “bugs”, associated with computer program glitches, is rumored to have come from this era as the glowing tubes attracted moths to the light that wreaked havoc in the systems.

To make an amplifier we need to stay away from using the device as a switch – we never want to reach the two extremes of always on or always off if music is going to come out of our device without distortion (the always on state is called clipping). This means we need to use the device in its in between state, and we want to begin our journey right in the middle.

So, let’s design our little amplifier using a single device – in this case we’ll choose a JFET and design a single ended amplifier. To do this we need a few elements: a battery and 4 resistors. We want a gain of 10 for our amplifier – meaning if we put 1 volt on the input we will get 10 volts on the output. This is easy. Remembering our device has 3 nodes: the signal input (gate) and the 2 battery inputs (drain and source) we place a resistor on the drain and connect it to the + of our battery and another resistor on the source and the – of the battery. We make sure the + resistor is 10 times bigger than the – resistor (this sets our gain). We’re almost ready to amplify our signal.

On the input side (gate in a FET, grid in a tube and base in a transistor) we want to make sure the device sits right in the middle of the battery voltage – this is important because we want our musical signal to have room to move – and move somewhere away from the two extremes (always on or always off). This process is called biasing the amplifier.

Let’s imagine we have a 30 volt battery – we want our device to therefore rest at half of that or 15 volts. In a FET, tube or transistor this is really easy, we simply use two more resistors on the input that have a ratio such that we get 1.5 volts at their meeting point. One goes to the – of the battery and the other to the + and when our little amplifier sees this, it produces 15 (1.5 x 10) volts on the output. Bingo, we have an amplifier. The output of this amplifier is taken at the junction of the + resistor and the drain.

Now, when you put our phono cartridge on the input, the tiny voltage is amplified and we get a 10 times larger signal on the output which makes music!

Sorry this was so technical and it’s the last one that will be – promise. Tomorrow we wrap up the series with some comments about how all this ties together.

Let me leave you with one interesting notion, that of field effect. Remember that in a tube or FET there is no physical connection between the input and the output – everything is controlled through a mysterious “field” that goes through the air and magically turns the device off or somewhere in between on and off? How the heck is that possible and is there an example of this in real everyday life?

Sure – and this is one I delight in leaving people with because it gets your brain spinning. Picture yourself driving in your car down a busy city street. Suddenly, the light ahead turns yellow and you slow down. The light turns red and you stop and wait. The light turns green and you go. Sound familiar? This is the same exact way a field works – only this time the field is working on you. In the traffic example a change in small particles called photons has this effect and in our transistor the particles are called electrons. In both cases there’s no apparent physical connection or touching yet control happens perfectly.

Now you know you’re no different than an amplifier. Just make sure you sit in the middle and don’t go running any red lights."
I hope this is helpful information.
Regards, Ken

11tsteve

good read. thank you Kenneth.

PreCd

Schematic and users manual..sorry but no service manual. You will need to register to download.

http://www.hifiengine.com/manuals/acurus/a250.shtml

11tsteve

nice. i had the schematic from DIY, but i see now it was incomplete. thanks PreCD.