|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
68.108.208.249
In Reply to: 6DJ8 or 12AT7 posted by jimdandy on February 21, 2007 at 07:38:55:
the long tailed pair. 12AT7 has amplification factor of around 60
as I recall, and 6DJ8 is only around 30. Go with the 12AT7 for
the invertor. If you are attempting to drive output tubes which
use grid resitors much lower than around, say, 330K, then I'd
consider using a buffer stage between the driver and the power
tube. A 5687 is a good choice for buffer, and yes, 6SN7 works too.
If you are using EL84 tubes for output, or cathode-bias EL34, then
you could use either the 12AT7 or even possibly better a 12AX7
long-tailed invertor with no buffer needed --AS LONG AS: you will
be operating the ouyput tubes mostly in class-A, AND you are NOT
using grid resistors for your power tubes of less than 470K.
I'd use 470K grid-R's for EL34s, and 1Meg for the EL84. Again,
the values I give presume cathode biased output stages operating
mostly in class-A (triode or ultralinear mode both okay).What kind of output tubes and config are you contemplating?
Follow Ups:
EL34s, I'm not using cathode bias. Running class A/B.
best results, and you cannot use a 12AX7 long-tail directly
because of the lower value grid resistor required for fixed bias
operation of EL-34. I'd consider using a simple split-load/concertina
invertor with either a 6SN7 or a 12AU7 --both generally sonically
(linearity) superior to 12AT7. Be sure to feed the concertina
and it's preceding input stage with WELL REGULATED/Filtered supply,
or you will absolutely suffer sonically detrimental LF feedback
through the power supply chain. This is particularly true with
class AB operation.Frankly, unless you are building a guitar amp,
I'd go cathode bias and not try to bias too far up-into class
AB. The classic Mullard (who designed the EL34!) " Mullard 5-20"
design with cathode bias and 12AX7 long-tailed inverter is a
VERY nice sounding design, and good for an honest 20 watts out.
IMHO, trying to squeeze a few more watts/dB out of the EL34 by
using fixed bias is not really worth it, and zaps some of the
sonic refinement. Google "Mullard 5-20" and you should find a
schematic. Also: the long-tailed pair has high PSRR, and so an
extremely well-filtered power supply feeding it is not needed.my two cents.
With its high gm and low Rp, the 12AT7 can directly drive EL34s. The critical issue is sufficiency of gain. Do you have a line stage with gain? If you do, a 'T7 differential splitter has enough gain to allow a 2 stage amp.
Eli D.
This post is made possible by the generous support of people like you and our sponsors: