|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
In Reply to: Re: Hum pot in DHT preamp ??? posted by Thomas Mayer on November 01, 2001 at 07:06:29:
did you ever consider connecting the bias resistor to the positive side... this will buy you some extra bias voltage, and allow for a slightly smaller resistor for the same bias point... you will lose some headroom on your AC swing before A2, but in early stages this generally isn't a problem.the above is more of a question than a statement...
i assume since you are using it, you like the sonic results of the bias resistor :-)
dave
Follow Ups:
Hi!> did you ever consider connecting the bias resistor to the positive > side... this will buy you some extra bias
> voltage, and allow for a slightly smaller resistor for the sameI'm not sure if I understand what you mean by this,
but with the filament bias approach this wouldn't work.
With filament bias, the filament itself is actually a
substantial part of the bias resistor.
> i assume since you are using it, you like the sonic results of > the bias resistor :-)Im my 801 linestage the bias resistor is only 33 Ohms.
I still use the ultrapath connection. The B+ side of
the lineout transformer is returned to the - side of
the filament. So the AC signal actually bypasses the
bias resistor (at least from the midrange on upwards)Thomas
lets assume more traditional R/C coupling....the filamnet has a 7.5V dc potential across it.... now tie one side to ground (through the RC) and build two different supplies a +7.5V one and a -7.5V one the two different filament supplies will considerably change where the tube biases up, and for the same operating point two different bias resistors will be needed...i think the negative supply will require a smaller bias resistor for the same current... the down side is grid current will start to flow when the grid hits -7.5V rather than zero...this is a problem in output tubes, but in drivers and such were we tend to keep well clear of the 0 bias point... maybe this is a place to make a small gain???
in other words with the -7.5V supply when one 1/2 of the filament is at -7.5V the other half is at zero and some grid current will start to flow. but with the +7.5V supply when one 1/2 of the filament hits 0 the other 1/2 is still at -7.5V so the grid current will start flowing with a difference of bias of 7.5V
is this making sense???
dave
Hi!I see now what you mean. Yes with conventinal biasing through
a cathode resistor, this might make a small difference.
However I thibk if one is concerned with grid current issues,
maybe he should use a different approach anyways.Thomas
the more i think of it the difference is pretty trivial... but i could see some instances where it could buy you something...when do you use filament bias, do you use a current source for your filaments??? or just a traditional linear LCLC supply at an elevated voltage?
dave
Hi Dave!> when do you use filament bias,
I do it wherever the power dissipation in the bias
resistor is managable.So far I use it in my all DHT phonostage,
in the 801 linestage and 801/801 power amps.With tubes like 26 or 71A the power in the bias
resistor would be quite small.Josef Aigner described a filament biased 71A stage
in the latest VALVE magazine.
> do you use a current source for your filaments??? or just a
> traditional linear
> LCLC supply at an elevated voltage?No current sources. I used them in the past, but found
passively filtered supplies to be better.
I let the filament/bias supply end with a choke.3-phase briged with Schottky diodes -> L-> C-> L -> filament.
In the linestage the filament supply is about 45-50V @1.25A
Thomas
i'm with you on the passive supplies... i guess i just have to get past the idea of putting a 50V supply on a tube that needs a 7.5V filament... i know on paper its stable and safe.. but still it scares me a bit...i guess i have to add this to my list of bias schemes to try...
dave
Hey Dave,Give it try, it's worth it.
You're right about the 50V with a 7.5V tube though.
So far I blew two 801s. The first due to a wiring error
and the second, when I measured the voltage across a
filament biases tube and accidentally shorted one
side to the chassis, which applied 45V to the filament.
A short flash and the tube was gone.Thomas
sure at some point i'll give it a try... i'm thinking i need a way to bias some parallel 6C33's @ -50V or so... they will be running 700ma per pair so the damper diode may not cut it... but with 7.5A of filament and plate current combined... oops.. make that 15A... i forgot i share cathodes :-) thats just a 3 ohm resistor :-) lessee... where on earth can i find a 3 ohm 1KW resistor.... maybe a halogen lightbulb :-) and hmm... a 60V 15A DC supply... hmmmm.... sounds like a concept amp to me :-)dave
Thomas, good to see you posting. Remember having dinner with me at ny noise a couple years ago at the steak house? I just wanted to chime in and say that I too find it best to CLC or LCLC rather than current sources for DHT filament supplies.I've been thinking about this filament bias thing too. Assuming you pick a tube that you wish to bias that has a low enough filament current, say a 71A, for example. i.e.) 5V@.25A. Then one may be able to select a diode dampener, one that will pass 250mA in this example, say two 6DM4As (or perhaps two 6ax4s like you use, if I remember correctly) in series dropping 20 volts each and experience a reduction in the dynamic impedance seen in the cathode?? This gets rid of all that power dissipation in that cathode resistor, say the 10 watts in the 160 ohm resistor as shown in the VALVE article.
Thoughts?
Hey Larry,of course I remember you!
On that evening you told me that my article in SP16
inspired you to build a 10/10 amp. Well, your 10/10 amp
inspired me to build a 801A/801A amp. So the circle is
closed.Thomas
Thomas -Your article in SP 16 inspired me, too - just ask Dave, I was at his loft place wearing some black clothing and smoking a cigar just last night.
-j
This post is made possible by the generous support of people like you and our sponsors: