PCB design techniques for Tube Amps - Page 2 - diyAudio
Go Back   Home > Forums > Amplifiers > Tubes / Valves

Tubes / Valves All about our sweet vacuum tubes :) Threads about Musical Instrument Amps of all kinds should be in the Instruments & Amps forum

Please consider donating to help us continue to serve you.

Ads on/off / Custom Title / More PMs / More album space / Advanced printing & mass image saving
Reply
 
Thread Tools Search this Thread
Old 11th October 2007, 02:39 PM   #11
diyAudio Member
 
richwalters's Avatar
 
Join Date: Sep 2003
Location: Alps:Tube amp designs over 150W, SMPS guru.
Quote:
Originally posted by d1camero


So far, the signal is on the bottom side, and the B+ is on the top. I am not sure which side to put the ground.

thanks
I did just such a design and after a year's use so much dust had accumulated around the B+ tracks (500V) which was making noises with ajacent tracks so I did a redesign and put both ground and signal on the underneath and B+ above and varnished the tracks. It does imply some complication routing but is worth it.
Where power components are mounted on fiberglass, I drilled air circulation holes directly underneath.
Where there is heat and some airflow, static dust accumulates around B+ tracks. A hint is to see how much dust settles around the EHT sections of a TV !
Ultimately the tag strips method is by far the best.

Major telecoms and computer designs often use 5V at 60A + with multilayer. Even with 2 sided, provided tracks are thick enough perhaps with extra strip Cu soldered....can't see any problem with heater currents..

richj
  Reply With Quote
Old 11th October 2007, 03:16 PM   #12
diyAudio Member
 
Join Date: Nov 2006
Rich thanks for the detailed reply. I will follow your advice!

As for heaters, it is not the current it is trying to keep AC hum out of the signal path. One recommendation was to wire the heaters in separate with twisted pair. Ths is how one builder did it for AX84 Firefly guitar amp.

I did have heaters on one board, but a lot of hum got through. Not sure how much was the board as I was dealing with grounding issues at the time as well - poor grounding scheme can cause lots of hum to creep in.

One further question: do you suggest that the ground be a ground plane or should I follow my "tried-and-true" star ground method that I have on my hand wired amps?

Don
  Reply With Quote
Old 11th October 2007, 05:48 PM   #13
Jeb-D. is offline Jeb-D.  United States
diyAudio Member
 
Jeb-D.'s Avatar
 
Join Date: Nov 2005
Location: SoCal
These are just personal observations as I'm sure some people have had good results doing otherwise.

I've tried the ground plane method a few times, but the performance was never better than the star and buss grounding for Class A circuits. Sometimes it seems better for the current to return to one point before another as it does in star and buss grounding. Ground plane is defiantly a safer solution for someone inexperienced with other grounding methods. But if you are experienced with other methods stick to the tried and true.

Putting signal and power on the same side of the board has never been a problem as long as clearances are generous.

As Cerrem mentioned, keep vias out of the signal path. In fact, it would be best to keep vias out completely if possible. The power supply is part of the signal path too.

AC heaters would be best kept off the board, but DC heaters are no biggie if put on the board.

Place grid stoppers as close to the socket pin as possible.
  Reply With Quote
Old 11th October 2007, 07:03 PM   #14
diyAudio Member
 
richwalters's Avatar
 
Join Date: Sep 2003
Location: Alps:Tube amp designs over 150W, SMPS guru.
Quote:
Originally posted by Jeb-D.

Place grid stoppers as close to the socket pin as possible.
Worth a mention. I knew someone who built a tube amp with alot of SMD stuff in it, and soon after I was the one who repaired it. Tempting it is to use Melf resistors and caps, (if one can get 'em cheap enough) make sure the PCB is mechanically evenly fixed, i.e can screw down with distorting it and components.
The damage is obvious and can easily happen when a tube is withdrawn from socket and stresses the PCB.

I go along with Jeb-D regarding grounding impedance. With point to point wiring doing a stereo amp, keep symmetrical earth layout and circuit amp earth bus to chassis near input tubes. Avoid using the input socket for earth. It is poor. Better is a thick LoZ Cu strip to chassis close by.
Same applies to PCD except as copper plane.
With PCB layout have a massive earth track plane and keep layout symmetrical where poss, esp relative to mains and psu chokes..(this will reduce eddy currents set up around pcb) This won't be quite as easy. I always try and engineer for the worst case that is one might have to find a " compromise sweet spot" common to each stereo half near input circuitry.Shown as arrows with enclosed photo. The PSU ground is done at the other end.With an AC heater supply, the old remedy of using a 100R 3W pot wired across the heater with wiper to chassis can often solve problems.

richj
Attached Images
File Type: jpg bus majestic.jpg (51.1 KB, 490 views)
  Reply With Quote
Old 27th July 2013, 03:57 PM   #15
diyAudio Member
 
Join Date: Jul 2013
Can you design a board for a tube unity gain using both sides of the 12AX7 ?
  Reply With Quote
Old 27th July 2013, 08:32 PM   #16
roline is offline roline  United States
diyAudio Member
 
roline's Avatar
 
Join Date: Sep 2010
Location: East Tennessee
I do single side toner transfer, etched in ziploc bag, eyelets for wire connections and feedthroughts to mount caps to the copper side.
I jump to top side for all; B+, signal and ground. The back side is mostly ground plane. If going 2 layer the top would also be ground plane.
I always run the heaters as a twisted pair 3/4" from the back side. and on some, dual heater or triple heaters wire sets are used for 17V, 5V and 6.3V.

It will take me around 4 to 6hrs to get a PCB ready to drill and assemble/install the eyelets. I would like to have a real PCB double sided with heavy plated feedthroughs instead, but that takes longer and co$ts much more.
Attached Images
File Type: jpg PC270356.jpg (813.3 KB, 186 views)
File Type: jpg amp1c.jpg (157.1 KB, 177 views)
File Type: jpg keg-ear21.jpg (560.4 KB, 172 views)
File Type: jpg all-pcb1.jpg (253.4 KB, 168 views)
File Type: jpg P7070223_b.jpg (155.8 KB, 73 views)
File Type: jpg P7070225_b.jpg (162.4 KB, 67 views)
__________________
SO many tubes, SO little time!!!
  Reply With Quote
Old 27th July 2013, 08:39 PM   #17
roline is offline roline  United States
diyAudio Member
 
roline's Avatar
 
Join Date: Sep 2010
Location: East Tennessee
some more examples... the PCB on the left was stuffed and is the last couple of pics above.
The design was altered to add a power Zener-mosfet in place of a resistor and some of the traces were increased in width. When I use PCB express software I can design for toner transfer or real PCB design which are not the same artwork. For toner transfer I print a combo of top and back for the single back side. I use yellow silkscreen to show where the jumpers go on the top side.
A side shot of the Keg-ear amp, the Keg-ear PCB and finished amp. ( 6L6-Kt88 family) The last pic is one of the earlier 6GV5 sweep tubes from the first couple of pics above.

and what they can turn into at the end......
Attached Images
File Type: jpg P6280207.jpg (836.1 KB, 70 views)
File Type: jpg 6jt6_h.jpg (331.2 KB, 59 views)
File Type: jpg 6.3 chassis side_c.jpg (111.2 KB, 57 views)
File Type: jpg 1.0_pcb1c.jpg (113.2 KB, 54 views)
File Type: jpg 8.0 Assembled_b.jpg (156.1 KB, 52 views)
File Type: jpg P3270500_a.jpg (838.1 KB, 43 views)
__________________
SO many tubes, SO little time!!!

Last edited by roline; 27th July 2013 at 08:48 PM.
  Reply With Quote
Old 27th July 2013, 10:05 PM   #18
diyAudio Member
 
denny9167's Avatar
 
Join Date: Nov 2007
Location: Graham, TX
Really good recommendations so far. I have an example of a current project consisting of a phono stage and line stage preamp. You will notice the narrow layout, to accommodate a low profile enclosure. I'm using DC on the heaters,and for that I'm able to run the traces on board to minimize noise,and wiring. I've been using ExpressPCB for 2 or 3 years, they are a little pricy, but they are consistent on quality and turn around.
Attached Images
File Type: jpg image.jpg (600.5 KB, 71 views)

Last edited by denny9167; 27th July 2013 at 10:15 PM.
  Reply With Quote
Old 28th July 2013, 09:26 AM   #19
12E1 is offline 12E1  United Kingdom
diyAudio Member
 
12E1's Avatar
 
Join Date: May 2012
Location: Near Huntingdon, Cambridgeshire, UK
Quote:
Originally Posted by roline View Post
It will take me around 4 to 6hrs to get a PCB ready to drill and assemble/install the eyelets. I would like to have a real PCB double sided with heavy plated feedthroughs instead, but that takes longer and co$ts much more.
I noticed the use of eyelets. (Great photos, and some really good boards too.)

Have you seen any problems caused by differential thermal expansion between the eyelet and the rest of the PCB? I remember seeing boards using this technique back in the 70's and after a few years they would start yo get a few hairline fractures at some of the though PCB connections. They eyelets used then were probably different and the solder would have been regular tin-lead of course.
  Reply With Quote
Old 28th July 2013, 01:42 PM   #20
roline is offline roline  United States
diyAudio Member
 
roline's Avatar
 
Join Date: Sep 2010
Location: East Tennessee
When I was involved with aerospace electronics at TRW ( TDRSS, IUS, Shuttle), the main problem that we experienced was the thermal fatigue caused by not having the proper lead form on components to allow for thermal expansion. We also had a few failures on 8 layer PCBs with the plated through holes developing cracks in the barrels causing disconnects between the layers. This was improved by allowing the through holes to be a little larger in size and thus allowing the solder to wick from bottom to top during wave soldering. The only issue with the eyelets has been not soldering them to the traces, mechanical flaring only and developing intermittent connections... I solder all eyelets to the traces and no problems.
The 6LU8 amp was done with the PCB designed upside down, traces to the top and components on the bottom. After the tube sockets were installed the leads were bent over for mechanical security and then the leads were soldered to the traces. The leads must be firmly bent over to prevent from pulling the traces off the PCB. Double sided PCB's solve this issue.
Attached Images
File Type: jpg P2090129.jpg (546.8 KB, 63 views)
File Type: jpg P2090134.jpg (498.2 KB, 53 views)
__________________
SO many tubes, SO little time!!!
  Reply With Quote

Reply


Hide this!Advertise here!
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off


Similar Threads
Thread Thread Starter Forum Replies Last Post
For Sale: Tube Amps, Class-A Amps, BG Ribbons, Chassis and Other Goodies! opc Swap Meet 23 2nd November 2009 05:04 PM
Numerical Solving Techniques for Optimum Crossover Design thadman Multi-Way 5 21st May 2009 05:29 AM
First Watt F3 / Tube Audio Design TAD-125 Hibachi JFET amps vizion Tubes / Valves 1 31st August 2008 03:08 AM
Tube pre-amps Glowin Plates Instruments and Amps 5 20th May 2004 06:30 AM
Op Amps for Everyone Design Guide (Rev. B) lohk Solid State 2 14th July 2003 11:15 AM


New To Site? Need Help?

All times are GMT. The time now is 07:14 PM.


vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2014 DragonByte Technologies Ltd.
Copyright 1999-2014 diyAudio

Content Relevant URLs by vBSEO 3.3.2