Bottlehead Forum Bottlehead Forum
Welcome, Guest. Please login or register.

Login with username, password and session length
News: Come see us February 11, 8-5 at the Head-Fi meet at the Burlingame Double Tree Hotel, Burlingame, CA. We'll be bringing lots of headphones and amps, and our prototype tube DAC used with our music server and the latest version of Amarra.

www.head-fi.org/t/584924/official-2012-bay-area-meet-thread-california-february-11th-saturday
 
   Home   Help Search Calendar Login Register  
Pages: [1]   Go Down
  Print  
Author Topic: Printed circuit board question  (Read 769 times)
0 Members and 1 Guest are viewing this topic.
ducati guy
Newbie
*
Offline Offline

Posts: 18


View Profile
« on: February 14, 2010, 08:13:49 AM »

While building my SR-45 I found that the biggest PITA was attaching the wires from the tube socket/terminal strip to the shunt regulator circuit board I had made. Is there any reason not to put the tube socket on the board as well (noise pickup, etc) other than the obvious one of needing different boards for tubes with different pin configurations?
Logged
Paul Joppa
Hero Member
*****
Offline Offline

Posts: 1642



View Profile
« Reply #1 on: February 14, 2010, 08:41:05 AM »

I have, as much as possible, avoided having signal voltages on PC boards because the board dielectric is really inferior. Never done any experiments to prove whether that's audible, it's just my best guess about why so many audiophiles prefer point to point wiring.

Most of my boards are power supplies, having fixed voltages. The one exception is current sources as plate loads; in those cases the output transistor lead is very close to the board edge and leads directly to a solder pad, total length 0.1 or 0.2 inch. This is to minimize the exposure of signal voltage to board capacitance.

The other problem with PC boards is they make it difficult to swap components like capacitors, which many of us do periodically. The virtues of PC boards are pretty obvious, mechanical stability and wiring convenience - not to mention prevention of wiring errors!
Logged

Paul Joppa
Caucasian Blackplate
Sr. Member
****
Offline Offline

Posts: 498



View Profile
« Reply #2 on: February 14, 2010, 10:52:13 AM »

I will also chime in on this, as I made up a large PC board to accommodate the entire shunt regulated power supply, the driver load, and the regulator for the driver cathode.  All of those components dissipate quite a bit of heat, but when you look at putting the 6CM7 onto the PCB, you have to remember that little tube is dissipating 7-10 watts.  With something like a wire wound resistor, this would be the situation of having a 25 watt resistor with a large body to throw off the heat, and you could space it off the board.  With a tube socket, there are many connections to the board and very close proximity.  This is a lot of heat for the board to contend with, and I generally feel that this isn't a reliable way to build an amp. 

Also, I find something somewhat horrid about designing a PCB with heater traces and the signal circuit all in one.  I like being able to shield the heaters, and having some short jumpers from the PCB to the tube sockets to allow the heat to go elsewhere...
Logged

Paul "PB" Birkeland

Bottlehead Grunt & The Repro Man
Pages: [1]   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.10 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!