Paralleling Tubes and SRPP

[Higo]

Hi,

I´m ending my first tube pre-amp design. I decided a SRPP configuration for Phone and a modification of Original PAS(http://www.normankoren.com/Audio/Spice_preamp.html) for Line.
When I started the design of the PCB board, I read something in an old magazine that I don´t know if It´s true.
I´m using twin triodes (12AX7/12AU7/6DJ8).
1- In SRPP: Must I use one tube for the upper section and another for the lower section or can I use the same tube for upper-lower section?
If I have to use one for the upper and another for the lower, Why???
2- In SRPP or another configuration like Original PAS, It´s possible "to paralleling" the tubes??? I mean If I´m using one triode in each stage I´ll be wasting the unused tube section. How can I do that??? I have to change something???

Any advice/answer will be appreciated.

Thanks!!!

Marcelo from Argentina

 

[Asen]

Yes you can use one tube per channel in SRPP. It's better to take some precautions though. Connect the ground of the heater to the high voltage using a 2/3 devider so that the voltage between the anode of the upper tube and the heater could not provoke a short circuit in the tube. If the high loltage is 180V you will have in the middle point of the devider, lets say 60V. Connecting the ground of the heater to this point you'll measure 113,7V between upper anode and heater voltages and circa 23,7V between lower anode and heater voltages.

Use the search engine for more info. Or take a look at this thread:
http://www.diyaudio.com/forums/showthread.php?threadid=250&highlight=srpp

Asen

 

[phil]

Hi Higo

Most SRPP circuits show the layout using one tube for the upper and lower part of the circuit.
You can do it either way but it is best if you use two heater supplies. Use one tube for the upper cicuit and one tube for the lower circuit
The upper tubes heater must be lifted to or near to the cathodes potential. This way your tubes will last longer.
Also consider a simple turn on delay for the HT allowing the heaters to warm up.
Phil.

 

[Higo]

Thanks for your answers Ansen and Phil!!!
However It was impossible to link to http://www.diyaudio.8m.com/srpp.html or http://www.diyaudio.8m.com/lpsh.html.
I have been reading the post, and, presently I´m a little confused about another thing more!!!
First:
Asen: about paralleling tubes, in the Original PAS, do you think that it´s suitable for do that?? I mean I´ll achieve a better noise figure and/or less distorsion???
Second:
Phil: I never saw a turn on delay in HT section like you mention. Is it common or It´s something that you do for experience??
Both:
What do you think that is better for a PSU, a tube (like 5T4) or a LM317 type regulator???
Thanks.
Marcelo from Argentina.

 

[phil]

Hi Higo

The turn on delay can be a simple 555, do a search on Manfred Huber, he has a great site full of good ideas and circuits that work.
You will not need a turn on delay if you use a valve rectifier in the psu.
Allen Wright of VSE suggests using two heater supplies for SRPP and even lifting the lower potential heaters by about 30v.
Phil.

 

[Asen]

This circuit is for slow warming up of the tubes. I'm sorry I cannot reply your questions about paralleling tubes and PAS. I've never done projects like these.

-------------

I'm sorry - the picture didn't appear. Probably I've done something wrong. Send me your e-mail and I'll post it to you.

 

[Higo]

Thanks a lot guys.
I´m planning to build a merge between a LM317 regulator and 555 for the heaters.
I´ll simulate a SRPP stage paralleling the upper and lower sections. I hope that It works and I´ll not be wasting the unused tube section !!!
Yesterday I simulate The Dynaco Original Pas (the modificated circuit that Norman Koren did it) and I obtain extrange results. Today I´ll probe again.

Bye, Bye
Marcelo


Asen My email: Lorenzo_Marcelo@hotmail.com

 

[dice45]

Higo,

please avoid paralleling tube sections. IMO and TME this eats up low level resolution and sonic subtleties. Allen Wright would second that, i know it

No matter how you do it, you are better on with one tube doing the job. With the right tube doing the job.

 

[Brett]

about paralleling tubes, in the Original PAS, do you think that it´s suitable for do that?? I mean I´ll achieve a better noise figure and/or less distorsion???

Paralleling will give lower noise and distortion, but in every single example I've ever heard, sounds worse. If noise is an issue, try quieter tubes or a better topology. Allen Wright has one in his TPCB that is a radical mod of the PAS, using a 2SK170 of similar as the bottom device in a cascode. I know this topology works well and sounds good, at least with tubes other than a 12AX7.

Second:
Phil: I never saw a turn on delay in HT section like you mention. Is it common or It´s something that you do for experience??

Easiest delay is to use TV damper diodes as rectifiers in the supply. The come on over about 30 secs and have very low switching noise, much lower than any SS devices. You get the delay, and better sound in one.

What do you think that is better for a PSU, a tube (like 5T4) or a LM317 type regulator???

Neither. The LM317 stinks sonically. Use a CLC filtered supply and a shunt regulator using a glow diode string, like OC3s, or much better still a proper active shunt regulator.

Cheers

 

[Higo]

Hi,
Dice45:
I´m planning to build a SRPP. I´m using Protel SE99 for sim. I have 12AX7, 12AU7, 12AT7, 6DJ8, 6AV6 and E80CC. Presently I only simulate 12AX7 and 12AU7. I don´t know if it´s convenient to use 12AX7 for lower section and 12AU7 for upper. What do you think??
Later I´ll post the circuit.

Brett: Do you think that CLC filter It´s better PSU that a solid state PSU? I was thinking in use LM317 regulator ´cause It´s cheaper than TL783 (u$13 vs u$2). Don´t fe afraid for the prices...I´m living in Argentina... . Can you give me another option??
I read that SPRR is very sensitive to bad regulated PSU. I´m right ???
I couldn´t find the PAS modification that you mention (Allen Wright ). Do you have any link?

Thanks a lot guys,
Grettings,
Marcelo.

 

[Higo]

Circuit

Dice45:
I attach the circuit that I simulate.

 

[Brett]

Brett: Do you think that CLC filter It´s better PSU that a solid state PSU? I was thinking in use LM317 regulator ´cause It´s cheaper than TL783 (u$13 vs u$2). Don´t fe afraid for the prices...I´m living in Argentina... . Can you give me another option??
I read that SPRR is very sensitive to bad regulated PSU. I´m right ???
I couldn´t find the PAS modification that you mention (Allen Wright ). Do you have any link?

Hi Marcello,

Yes my preference would be for a CLC or CLCLC filter over either the LM317 or TL783, neither of which I'd use. The only SS regs I will use are shunts, which none of the chip manufacturers make TMK. The older, but still good version of the circuit I use, can be found <a href="http://www.vacuumstate.com/images/PS_HT_CS_Circuit.gif">here</a> and <a href="http://www.vacuumstate.com/images/HT_Reg_Circuit.gif">here</a>. The later version simply replaces the current source using an LM317, with a suitable resistor. Tweak the value of the 56k resistor to get the output voltage you want, and use a better opamp, like an AD797 if your budget allows. A good PSU is imperative for performance in a single ended design.

The TV damper diodes I mentioned for the PSU are 6CJ3 or 6D22S, although a 5AR4/5AS4/GZ37 isn't too bad. Much, much, much better than SS devices.

Personally, I'm not a fan of SRPPs either as they don't sound good to my ears, and they're performance is very load dependent.
A really excellent article on SRPPs that explains them much better than I can, can be found <a href="http://www.tubecad.com/articles_2002/SRPP_Deconstructed/SRPP_Deconstructed.pdf">here</a>, as can the first part of a two part series on <a href="http://www.tubecad.com/articles_2002/RIAA_Preamps_Part_1.pdf">RIAA stages</a> by the same author, John Broskie of <a href="http://www.tubecad.com/">TubeCad</a>. Tons and tons of great design info at his site, and he also writes and sells some software for simulating tube circuits that's excellent.

The Wright PAS mod I mentioned is in his Tube Preamp Cookbook. I dont have a scanner so I'll try and draw it out for you. If it were me, I'd ditch the PAS circuitry altogether and build something like <a href="http://www.vacuumstate.com/images/fvp5.gif">this</a>, using the regulators I linked you to above. This is a few generations after the PAS was ditched.

HTH
Cheers
Brett

 

[Asen]

For SRPP I would highly recomend the use of 6DJ8 (ECC88, E88CC, etc.). This is low voltage, high curent tube.

 

[Higo]

Hi Brett
I´ve been reading the TubeCAD article that you mention and the next step will be split R13 in two resistors.
The reason why I want to use SRPP It´s that It´s look simple to build. I´m planning design the Amp, then the load dependance in his behavior It´s not a problem (50K Zin Amp It´s that I want).
I´ll try the preamp circuit that you link.

Assen: Ok. I´ll try 6DJ8, but I have to create a new element in Protel, ´cause It doesn´t appear. Damm It...

Thanks guys.
Marcelo.

 

[Asen]

I have a model for CircuitMaker 2000.

 

[Higo]

I´ll appreciate if you can send me an e-mail with the library...

Thanks.
Marcelo.