ttradio.net   /   Terry's DIY tube amp pages


 

 


Information about tube bias arrangement

  Self bias comes from voltage drop of a cathode resistor, in case of minor gas inside glass envelope, the slightly plate current increasing will lead to negative bias increasing so that this is a self protection to prevent over current. See picture below, an example of cathode resistor bias. (348 ohm, -34V )
  Fixed bias comes from a negative voltage power supply so that cathode resistor is not necessary. If gas is inside it will lead to plate current drift (increase) but the bias is constant, the over heat effect will further increase plate current and so on. Eventually red hot plate comes (thermal run away). See picture below, an example of fixed bias. ( Cathode grounded)
  Conclusion : - some marginal gassy tubes might pass in self bias amplifiers but could fail in fixed bias amplifiers, that means fixed bias test can filter out those marginal tubes. Though slightly gassy tubes still work on self bias but sound quality is questionable.


The tube amplifier story

The first vacuum radio tube was made around 1900 in laboratory and became commercial mass production in 1920.

In mid 1920's tube radio became popular and lot of manufacturers such as RCA victor, ATWATER KENT, KOLSTER, MARCONI etc. made famous radio models. At that moment, stand alone tube amplifier was not popular.

********* 1929 tube radio sale advertize in north America. **********

In early 1950's, 33-1/3 rpm long play records replaced the old type 78 rpm steel pin play record which did not want an amplifier. The good sound quality of 33-1/3 records led to high quality speakers and tube amplifiers.

The triode single ended output stage of old radios was designed to drive a few watts speaker and obviously it is not enough to provide strong base to modern speakers. 30-40 watt output power amplifier demanded in the market.

The pentode push pull output amplifier can provide 30-70 watt depends on different out put tubes. (EL34. 6L6GC around 30W, 6550, KT88 around 60W)

I built my first 6V6 6SN7 low cost tube amp in late 60's, and then 6BQ5, played with crystal cartridge and self built speaker cabinets. Sound quality was just ok but I enjoy a lot. In 70's most folks changed to build transistor amps with 2SB337 2SB426 Japanses products and later use 2N3055 2N2955 US products. Eventually power MOSFET available from market. Actually the sound quality was not good yet, most audiences become tired after few hour listenning.

From mid 80's I went back to tube amplifiers and tried to bulid a 30W 6L6 push-pull mono block pair with 6SN7 drive tube.

 

I selected a popular circuit diagram see picture

 

Stage 1 voltage amplifier calculation, plate output. Grid input load of next stage neglectable.

 

Stage 2 phase spliter calculation, plate and cathode output.
Grid input resistor load of next stage parallel to plate and cathode resistor.

 

Stage 3 driver stage calculation, plate output.
Grid input resistor load of next stage parallel to plate resistor.
Cathode resistor neglected due to differential amplifier operation.

 

6L6 output stage, pentode connection operating point Vp=395V, Ip=52ma, RL=2.5K
Po=(Vp-Vpmin)Ipmax/2
   =(395-75)0.18/2
   =28.8W
Grid input 42V p-p for 28.8 Watt push pull output power.

The open loop voltage gain to the power tube is: (NFB not yet considered)
(-12.24)x0.902x(-14.8)=163.4
i.e.   42/163.4=0.257 Vpp

That means 0.257 Vpp input can deive out 28.8W output.
6 db NFB arranged, 0.514 Vpp can drive same output.
See 2.5K load line below.


Ok, each stage was working good in linear portion, no overload drive,
every thing seemed perfect.
The sound quality is quite good but I expected it should be better.
After about a month listening I rearrange the output stage as ultra-linear connection
and cathode NFB arrangement. ( another -6 db NFB, 1.028V pp input for full out put)


Final test result -

Sensitivity: 1.028 Vpp for 28W output

Frequency response :

10 HZ -1.6 db
15 HZ -0.6 db
20 Hz to 45 KHZ +/- 0 db (flat)
60 KHZ -0.6 db
80 KHZ -3 db

Where 0 db=28W power output.


THD (total harmonic distortion)

20 HZ 28W 1.2%
2 KHZ 28W 1.2%
20 KHZ 28W 1.2%

The final sound quality is now much better and can easily drive
different type of speakers.



Thank you for looking.

                                                                                          Written by    Terry  Tam   I. Eng  MIEEIE  MIET  (U.K.)

  ttradio.net   /   Terry's DIY Audio pages