Author Topic: Tube bias theory  (Read 2246 times)

0 Members and 1 Guest are viewing this topic.

Offline jeffjmrTopic starter

  • Regular Contributor
  • *
  • Posts: 144
  • Country: us
Tube bias theory
« on: February 13, 2021, 01:34:18 pm »
I understand there is an ideal bias voltage for tubes. But besides the difficulty in accurately measuring the current to match the particular tube’s ideal spec, I am put off by some who recommend “set it to what sounds best”.

I’m curious to learn what exactly happens to the tube’s output when slightly underbiased and slightly overbiased, ignoring for now tube life differences. I’d like to err on the side of caution to extend tube life by underbiasing, but I can’t hear the difference in my amp, and would like to know what is actually happening when I depart from the ideal.

Thanks,
Jeff
 

Offline Cubdriver

  • Supporter
  • ****
  • Posts: 4201
  • Country: us
  • Nixie addict
    • Photos of electronic gear
Re: Tube bias theory
« Reply #1 on: February 13, 2021, 06:26:25 pm »
It depends on how the tube is being used.  If class A, you want to set the bias point to put it in the center of its most linear region to maximize the output swing on either side of the bias point.  If you bias it too low, you'll start to get distortion on negative input swings as it starts to cut off.  If too high, it will begin clipping on positive input swings as it approaches saturation.

If in class B (or AB) - push-pull - biasing the tubes too low will cause crossover distortion because there will be a 'dead' zone on either side of zero where the first will turn off before the other side turns on.

What is the application?

-Pat
If it jams, force it.  If it breaks, you needed a new one anyway...
 

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 10172
  • Country: gb
Re: Tube bias theory
« Reply #2 on: February 14, 2021, 11:12:34 am »
To do it properly, you need the tube operating curves from the datasheet and then (as Cubdriver says) chose the most linear operating point. Under or over biassing from that operating point is going to increase distortion - the people who recommend 'setting for what sounds best' are choosing increased 2nd harmonic distortion (or just don't know how to work out operating points and read datasheets).
Best Regards, Chris
 

Offline Circlotron

  • Super Contributor
  • ***
  • Posts: 3362
  • Country: au
Re: Tube bias theory
« Reply #3 on: February 14, 2021, 11:54:23 am »
the people who recommend 'setting for what sounds best' are choosing increased 2nd harmonic distortion
Which may actually be a valid approach for a musical instrument amplifier, but not for hifi.
 

Offline bob91343

  • Super Contributor
  • ***
  • Posts: 2675
  • Country: us
Re: Tube bias theory
« Reply #4 on: February 14, 2021, 06:29:52 pm »
There is no substitute for knowing what you are doing and using decent test equipment.  Everything else is a compromise with widely varying results.
 

Offline jeffjmrTopic starter

  • Regular Contributor
  • *
  • Posts: 144
  • Country: us
Re: Tube bias theory
« Reply #5 on: February 15, 2021, 04:52:19 am »
The amp in question is a TriodeUSA version of the Dynaco ST70, which I run in ultralinear mode (Class A?).

It is fitted with JJ EL34s. I assume you are familiar with this popular amp, but I will mention that the bias controls each are shared by a pair of the EL34s. Originally 1.56v was specified but later recommendations fell to 1v or even a little less. Thus I am pretty sure I am not at the ideal level.

I have toyed with the idea of modding it to provide individual bias. Still learning about tubes, and have suitable skills and equipment to make the mod. Was just curious what is actually going on when not at the ideal bias. Now I get it and thanks for the clarification. I know an actual check of the current is required to get it right. 50ma rings a bell but I need to check.

Jeff

 

Offline bob91343

  • Super Contributor
  • ***
  • Posts: 2675
  • Country: us
Re: Tube bias theory
« Reply #6 on: February 15, 2021, 06:17:14 am »
Don't dwell too much on this.  Bias can be set over a wide range with almost no audible change.  It's mostly effective on weak signals although there is a slight influence on maximum power.

Setting it too low makes the current increase, a good thing for weak signal operation but can reduce tube life if carried too far.  When the bias is too high, weak signals won't be able to be reproduced since the plate currents are too low.

That Dynaco was a well respected amplifier in its day, probably late 1950s.  Its secret is the output transformer.  I used a pair of them back in the day and they worked well.  Perhaps my ears aren't so good but after I tried a few amplifiers I decided they all sounded pretty much the same.  I switched to a Marantz 2275 for a number of years and then installed an Onkyo which I still use.  No audible difference to me at least.  I am no longer fussy, as my old ears don't hear that well any more.  I still use the Onkyo for the background to my music practice.

Interestingly, I retired the Marantz not because of any fault, but because of the radio tuning dial.  I found myself spinning the dial a lot and I wanted a tuner with presets, which the Onkyo provided.
 

Online TimFox

  • Super Contributor
  • ***
  • Posts: 8996
  • Country: us
  • Retired, now restoring antique test equipment
Re: Tube bias theory
« Reply #7 on: February 15, 2021, 05:56:21 pm »
By the way, the Dynaco amplifiers chose a current shunt resistor value in the cathode circuits that would produce 1.56 V at the factory-recommended bias current, so that one could use a cheap voltmeter to compare the voltage with the voltage of a fresh Zn-C cell.
"Ultra-linear" does not necessarily mean Class-A.  The term was used for a circuit where the screen grids of the output tetrodes were connected to special taps on the output transformer (typically at 43% of the winding) to provide local feedback.  The operation of that circuit is often considered to be intermediate between tetrode (screens at constant voltage) and triode (screens connected to plates) operation of the output tubes.
 

Offline Johnny10

  • Frequent Contributor
  • **
  • Posts: 901
  • Country: us
Re: Tube bias theory
« Reply #8 on: February 15, 2021, 07:01:25 pm »
 :popcorn:
Tektronix TDS7104, DMM4050, HP 3561A, HP 35665, Tek 2465A, HP8903B, DSA602A, Tek 7854, 7834, HP3457A, Tek 575, 576, 577 Curve Tracers, Datron 4000, Datron 4000A, DOS4EVER uTracer, HP5335A, EIP534B 20GHz Frequency Counter, TrueTime Rubidium, Sencore LC102, Tek TG506, TG501, SG503, HP 8568B
 

Offline jeffjmrTopic starter

  • Regular Contributor
  • *
  • Posts: 144
  • Country: us
Re: Tube bias theory
« Reply #9 on: February 16, 2021, 04:02:07 am »
I certainly do NOT know how to read data sheets beyond the voltage and current limits.

The amp maker suggests 1v for each pair of EL34s. Being conservative leaning I am at between .940 and .950. They tend to drift lower as the amp heats up. I believe the bias resistor is different from the original Dynaco design hence the 1v recommendation versus the original 1.56v.

Is there a Cliff Notes version of what to look for here?

Data sheet attached.

Thanks,
Jeff
* el34_e34l.pdf (0 kB - downloaded 51 times.)
« Last Edit: February 16, 2021, 04:03:50 am by jeffjmr »
 

Offline Cubdriver

  • Supporter
  • ****
  • Posts: 4201
  • Country: us
  • Nixie addict
    • Photos of electronic gear
Re: Tube bias theory
« Reply #10 on: February 16, 2021, 06:51:01 am »
To the best of my knowledge, Dynaco style amps are not known to be particularly hard on output tubes, so why not just bias them to what the amp manufacturer suggests?  I have a VTA ST-70 clone, and while the driver is different (it biases the tubes individually rather than in pairs), I've run it for many hours using the bias setting from the manual and the outputs are fine.

-Pat
If it jams, force it.  If it breaks, you needed a new one anyway...
 

Offline Johnny10

  • Frequent Contributor
  • **
  • Posts: 901
  • Country: us
Re: Tube bias theory
« Reply #11 on: February 16, 2021, 06:46:07 pm »
To know more about tube Biasing...

There are three YouTube posters that have great descriptions of this.

El Paso TubeAmps
Uncle Doug
Blue Glow Electronics

I don't think you can go wrong with any of the three.
Tons of experience with the Dynamo Amps and Biasing Tube Amplifiers.

Here's a sample:


« Last Edit: February 17, 2021, 10:25:11 pm by Johnny10 »
Tektronix TDS7104, DMM4050, HP 3561A, HP 35665, Tek 2465A, HP8903B, DSA602A, Tek 7854, 7834, HP3457A, Tek 575, 576, 577 Curve Tracers, Datron 4000, Datron 4000A, DOS4EVER uTracer, HP5335A, EIP534B 20GHz Frequency Counter, TrueTime Rubidium, Sencore LC102, Tek TG506, TG501, SG503, HP 8568B
 

Offline jeffjmrTopic starter

  • Regular Contributor
  • *
  • Posts: 144
  • Country: us
Re: Tube bias theory
« Reply #12 on: February 17, 2021, 01:55:03 pm »
Cubdriver, why not follow the manufacturer’s recommendation was exactly the meaning behind my question.  What happens if I don’t?

And It seems there is good reason to go with what is recommended. Given that the numbers they suggest are probably an average, I’d feel better if I had a better understanding of the real parameter we are trying to manage, which I believe is the current flow, which is what the bias is intended to control.

There are still some gaps in my knowledge and these posts have helped fill some of them. Going to read up some more to try to learn what I am looking at on tube data sheets.

Thanks,
Jeff
 

Offline basinstreetdesign

  • Frequent Contributor
  • **
  • Posts: 466
  • Country: ca
Re: Tube bias theory
« Reply #13 on: February 18, 2021, 04:59:16 am »
STAND BACK!  I'm going to try SCIENCE!
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf