|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
198.209.41.132
In Reply to: RE: Bias source from B+ or seperate trans, which is better? posted by DAK on August 08, 2014 at 09:29:28
A separate bias transformer, and a FULL WAVE bridge, is always superior to a tap off a HV winding and a typical half wave, quasi rectified bias voltage !!
Your VA rating will be fine, backwards, forwards or side ways. Separate bias supplies are the best way to do fixed bias. Maybe you can fit a ( small ) choke input filter after the Full Wave Bridge. !!!
Always remember, the bias supply is on the MUSIC GRID of the tube, and it needs to be squeeky clean in execution !!!
Jeff Medwin
Follow Ups:
I will tap off the 8A 6.3v lines to a full wave rectifier consisting of 4 3A 100v Schottky diodes, then to the 6.3v filament transformer, or vice versa ( does it make a difference?), then to a Pi filter, then to the 2 bias pots. I will need to put in a resistor to ground on the third leg of the pots. I am shooting for at least 40v from the bias pots. The amp is a parallel single ended kt88/6550 stereo.
You need to place the rectifiers after the last transformer winding.
Transformers don't work with DC.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
That is exactly what I did, anyway. I guess intuitively I knew that is how it should be from looking at hundreds of schematics. thank you for pointing that out. cheers, Dak
that the bias supply should "feel" the dynamics of the main power supply?
Thanks!
What ??? I would like to know what you are talking about.
The bias supply, in fixed bias, is mixed directly with the music signal, on the control grid of the tube, supply needs to be squeeky clean !!
Jeff Medwin
Jeff, The argument is that the tap off the PT for the bias supply is part of the PT. Therefore, it "sees" EMF changes of the PT due to demands on the power supply.
Guess it's not an issue. Just wondered what your thoughts were, on this topic.
Agree that a clean bias supply is important, but disagree that a separate transformer is necessary.
I feel that a full wave supply is wise, but to my surprise I've never heard the common half-wave supply cause hum.
Bias supplies need furnish no current unless the amp is overdriven, and that lets us get away with the half-wave cheapness (which I still don't like, empirical evidence admittedly to the contrary). But a whole separate transformer for a supply that furnishes no current seems a waste.
Tap it off the B+ winding, full-wave filter, resistor or choke, filter, and you're good. Use a film cap for the final filter and you're extra good. If you're going to use a choke you should probably have a 1K or so resistor to ground so the choke "has something to do." But I wouldn't use the choke.
And don't regulate the C- unless you regulate the B+. You want the two supplies to track, so that as B+ goes up with line voltage (causing more current to flow in the tubes) the C- also goes up, countering the rise in current.
Lee,
What you write seems reasonable to me, but I was specifically told to do fixed bias with a separate PT and a FWB by an audiophile tenured EE Professor, a genius. No EE got through his class until they built a successful sounding tube amp !! This E.E. PhD. was in charge of USA's early NASA project of putting a monkey into space.They were using an EARLY version of Jeremy Epstein's DC 2A3 amp circuit. To calm the mammal, in the capsule, they installed WE 755s, played Bach and Beethoven, using an amplifier that was intially called "Monkey on a Rocket ".
Seriously, there are all kinds of high frequency nasties in the modulating HV secondary, and I could underdstand intuitively - a separate PT for fixed bias would be better. Rather than..."that will do" designing. The EE Professor was specific about this !!
Jeff Medwin
Edits: 08/08/14 08/08/14
Jeff, why is it that all these people to whom you have stories like this are long dead and buried with absolutely no published documentation for the claims you make.
Do you have some reference as to this professor and the requirement for students to build an audio amp to pass the course? As this is a rather unique credit requirement I'm sure there would be something documented of this practice.
What was the professors name and the school where this was practiced so some of us may research this unique course requirement further?
monkeys in space. that made me laugh.
I *heard* through a brilliant engineer that cathode bias was superior. Said engineer had actually developed apparatus to help pigs fly.
I believe the spikes from rectification not only affects the forward B+ supply but is also reflected back into the power transformer and cross contaminates all th secondaries.
To that end I placed three snubber caps, .1 uF, across the legs of the filament leads and ground, hoping that the effect would have some change in the B+. It certainly did. Everything became significantly warmer and fuller with an increase in midrange dynamics and detail. Cheap and very effective.
I only had 120 VAC rated caps so I couldn't place them on the B+ leads.
Also one of the major issues with bias circuits is to adequately buffer it from the ground. In most cases, the adjustable pot ( thinking Dyna circuits here) simply has a a cap to ground. This allows the bias voltage to swing a bit more than I prefer. I place a simple cap across the resistor to ground with great results, better resolutions and dynamics. The cap ( a small one, 5 UF is OK and electrolytics certainly work very well) buffers the bias circuit extremely well. You can see this in metering out the bias. Turn the knob and the voltage moves significantly slower, thus maintaining better control over the signal, IMHE.
Using this cap actually eliminates the need for a separate full wave bias bridge, although it should work there too.
YMMV and certainly FWIW
But is it not good practice to have the bias supply immediately available to the power tubes? With a separate trans running off the 6.3v filament tap the bias voltage will appear immediately. But on the other hand since my B+ is via tube rectifiers, there will be a lag if I tap off the B+ voltage.
You don't tap off the B+ for bias. You tap off the AC and rectify that.
Bias will appear immediately.
The biggest advantage to tapping off the HV AC is that bias will follow the tubes current draw quite well, protecting the output tubes. The biggest disadvantage is the potential for noise. In PP that noise will cancel in the output transformer, however in single ended amps, bias needs to be fairly quiet.
Dan Santoni
Or you could go way over the top - see schematic below. Seperate tap, TUBE rectified and choke input!
Critical inductance choke input, no less! (a little higher than 3 times critical inductance)
Of course, most don't consider a non-critical inductance choke in the input position a "choke input filter".
It will behave much more like a cap input filter than a choke input filter with short, high current pulses instead of constant, low current (just a little greater than the current drawn by the load) delivery.
Tre'
Have Fun and Enjoy the Music
"Still Working the Problem"
Post a Followup:
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: