P2P Redd 47 - a few questions

[PeteC]

I'm still trying to understand whether its OK to run both channels off the one regulator.......  i think it is ......

I guess I'm trying to understand how to calculate the current that the 2 OA2's will draw.  ( or the zener string, if I go that way instead of teh Reg tubes. - 6x51v string )

Cheers
Pete

 

[PeteC]

Ok,
been playing around some values.

my mains TX is 550v CT ( at 150ma - loads to play with )

So, i reckon I will have rectified Cap input HT of circa 388v.  550/2x 1.41.
I'm shooting for 300v regulated, and as I'm running 2 channels, about 60ma total current

So I need to size the PS dropping resistors in the RC networks to get me to say 320v just before the OA2's.

In the REDD47 circuit there are x3 1K resistors in the RC filters , and another optional one jumpered out ( which can be put in circuit of the raw HT > 410v according to the manuf info sheet. )

So......I need to drop about 60volts before the Reg tubes. 
So.....60volts dropped at a current draw of 60ma requires a total resistance between raw HT (388v ) and the pre-Reg B+  ( 320v ) = 1000 Ohms.  1k.

So... with 3 R-C resistors they each need to be circa 330 Ohms ( not the 1K's in the original design )
Power dissipation rating for each needs to be c. 60ma x 20v = 1.2Watts , so 2Watts for comfort...

What I dont understand is how to calc what current the OA2's actually draw with this circuit.

Cheers
Pete




 

 

[CJ]

looking up the tube from Franks place (see signature at bottom of this post)

it says 75 ma start-up, 5 ma min, 30 ma max,

you might want to increase the plate resistor value a bit, 18 ma is a pretty big load for this type of tube, it will wear out pretty quick at this level,  they run the tube  hard because the distortion comes down with increased Ip, at the sacrifice of tube life,

24 K might be a better choice for two parallel sections,
Audio Research SP-12 is an example of high end amp that eats tubes pretty quickly,

here is a pic of a 6DJ8 i took apart, the frame grid is a moly metal frame that supports the grid wires in a more precise fashion than the usual round posts, this allows the grid to be placed closer to the cathode, which means greater Gm, and it also lowers distortion due to the more uniform placement of the grid wires, this info courtesy of Eric Barbour, Vacuum Tube Valley issue #7.

 

[ilfungo]

sorry for the maybe stupid question
I can use a UTC A10 as an input?
thanks

 

[My3gger]

Of coure you can use it. Just take away zobel filter for the old transformer and choose proper loading resistor for A-10.
I would use A-10 for something original, there are many nice tube RCA preamps that would be candidats for A-10.

 

[ilfungo]

Thanks  My3gger
sorry i'm not an exper
proper loading resistor for A-10?
What can I use valid and not too expensive as output?
thanks!

 

[Kamel]

A10 is 150:50k (1:18,25) if used as mic input, be sure you will not overload the ef86 input

if you use it as 600:50k (1:9,18) it will work but impedance will not be strickly matched so you have to use the proper load on secondary

try 166k to have a 2k reflected at the primary

 

[My3gger]

What can I use valid and not too expensive as output?

You could use Edcor WSM 15k/600:
http://www.don-audio.com/Edcor-WSM15K-600
It is cheap, sounds good and many people used it in R47 with success. Original output ratio was 7:1 (Edcor 15k-600 is 5:1), so their 15k-200 model might work closer to original. Ask about it, i'm not really sure.
I wouldn't waste my money and time with reg. tubes, use zeners instead. There are remarks in original datasheet about 1st and 2nd tube distortion, this is why i run E88CC at 18mA.

 

[ilfungo]

thanks My3gger!!!
You suggest to use solid-state power?


last question
Hammond 370BX  or 369JX?
THANKS!!!

 

[My3gger]

thanks My3gger!!!
You suggest to use solid-state power?


last question
Hammond 370BX  or 369JX?
THANKS!!!

Yes, i would do it with solid state power supply, Hammond 369JX looks ok. You could use simple zener regulator or capacitance multiplier for HT, AC heaters and phantom from 50V tap like here:
http://groupdiy.com/index.php?topic=42745.40

 

[ilfungo]

I've found two Beyerdynamic transformer (TR / BV 352 007 004 - 1:7 - 200:10 k)
I would like to try one on input
How do I calculate zobel network?
Thanks

 

[My3gger]

I don't know Beyers well myself so can't suggest zobel filter. You can try it without, just load it with something like 100k loading resistor across secondary. This should get you close and it would be nice to compare it with A-10.
Beyer transformers have very small cores and distort easily, so you will need a pad in many situations. This preamp was made to sound pretty clean, so "better" input transformers like Lundahl, Cinemag or Jensen are better match for them.

 

[abbey road d enfer]

I've found two Beyerdynamic transformer (TR / BV 352 007 004 - 1:7 - 200:10 k)
I would like to try one on input
How do I calculate zobel network?
Thanks

One does not "calculate" a Zobel. Zobel calculation is based on the actual leakage inductance and stray capacitance of the transformers. These two factors are generally not published. They are in fact impossible to measure directly. Their valuation is based on indirect measurements, that are subject to many set-up errors and misinterpretations.
Most often a Zobel is optimised by trial and error, starting with well defined source and load impedance and analysis of the frequency response up to several octaves above the audio spectrum, or square-wave response.
Please note that a Zobel can be optimized for one and only one source impedance. All practical implementations make assumptions on the most likely range of source impedance.

 

[ilfungo]

Thanks  My3gger
Thanks Abbey road d enfer
I will try some configuration
Thanks!!!

 

[jdbakker]

Most often a Zobel is optimised by trial and error, starting with well defined source and load impedance and analysis of the frequency response up to several octaves above the audio spectrum, or square-wave response.

Indeed. One procedure to tune a Zobel can be found here: http://groupdiy.com/index.php?topic=1221.msg15714#msg15714

JDB.
[back when thread-IDs were still in the low thousands]

 

[jonathanmorbin]

Hi,

I see this thread has been dormant for a while but I'm hoping someone is still on it. Its full of a lot of good info!

I have built one of these Pre Amps P2P to the same design as the original schematic. However i noticed that the 0A2's, when given a good amount of signal, blink and i can hear a small audible pop when they do so. Im presuming they are drawing to much current. Is there a solution to the PSU design to stop this from happening? Or am i just feeding it too much signal and this is what they do?

 

[ruffrecords]

Hi,

I see this thread has been dormant for a while but I'm hoping someone is still on it. Its full of a lot of good info!

I have built one of these Pre Amps P2P to the same design as the original schematic. However i noticed that the 0A2's, when given a good amount of signal, blink and i can hear a small audible pop when they do so. Im presuming they are drawing to much current. Is there a solution to the PSU design to stop this from happening? Or am i just feeding it too much signal and this is what they do?

When you say 'blink' I presume you mean they briefly turn off i.e. the glow stops. If this is the case then either the downstream circuit is drawing too much current OR the the combination of HT supply and 0A2 feed resistor is wrong.

Downstream current depends on the quiescent current and the signal level. The quiescent is supposed to be about 20mA so the first thing to do is check that. The total current draw is supposed to be about 30mA so if you put a meter across the 1K resistor feeding the 0A2 you should get a reading of about 30 volts.

Cheers

Ian

 

[jonathanmorbin]

When you say 'blink' I presume you mean they briefly turn off i.e. the glow stops. If this is the case then either the downstream circuit is drawing too much current OR the the combination of HT supply and 0A2 feed resistor is wrong.

Downstream current depends on the quiescent current and the signal level. The quiescent is supposed to be about 20mA so the first thing to do is check that. The total current draw is supposed to be about 30mA so if you put a meter across the 1K resistor feeding the 0A2 you should get a reading of about 30 volts.

Cheers

Ian

Yes that is what i mean by blink. It occurs when a fairly hot signal goes in. more so on bassier sounds, i may just be overloading it but its not audibly distorting when this occurs.

Across that 1k resistor i get 23.28v

 

[abbey road d enfer]

When you say 'blink' I presume you mean they briefly turn off i.e. the glow stops. If this is the case then either the downstream circuit is drawing too much current OR the the combination of HT supply and 0A2 feed resistor is wrong.

Downstream current depends on the quiescent current and the signal level. The quiescent is supposed to be about 20mA so the first thing to do is check that. The total current draw is supposed to be about 30mA so if you put a meter across the 1K resistor feeding the 0A2 you should get a reading of about 30 volts.

Cheers

Ian

Yes that is what i mean by blink. It occurs when a fairly hot signal goes in. more so on bassier sounds, i may just be overloading it but its not audibly distorting when this occurs.

Across that 1k resistor i get 23.28v

Yes, from the description I gather the OA2 doesn't draw enough current - or doesn't get enough voltage, whatever words can convey. Clearly the unreg voltage is not high enough. The idle current is 23mA, should be closer to 30. It may not be really consequential because the transition from regulated (OA2 conductive) to non-regulated (OA2 cut-off) is smoothed out by the capacitors.

 

[jonathanmorbin]

So are you saying that I need to increase the voltage going in to the 0a2 by changing the resistors before (lowering the value?) so that my current draw is closer to 30mA? Would this hopefully stop the regulators from going off?