PCB for stand alone 5W SE - 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 26th April 2011, 08:11 PM   #1
diyAudio Member
 
Join Date: Feb 2011
Default PCB for stand alone 5W SE

Hello I want to build a stand alone 5W SE power amp. Nothing out of the ordinary. Same thing as they use on the ax84.
The only thing is that i want to make that on a PCB. Pretty much because it is easy for me. The power supply will in a separate enclosure.
I have been reading about coupling between the traces, but i assume this has more to do with traces on both sides of the board. Mine will have only one side. Is coupling still strong with traces that run parallel but side by side and not on top of each other?
Keep in mind i have all space in the world for this board so i can make the traces as further apart as i want, but i am curious...
  Reply With Quote
Old 26th April 2011, 08:20 PM   #2
diyAudio Member
 
Tom Bavis's Avatar
 
Join Date: Jun 2002
Location: Macedon NY
The Tubelab Simple SE board will do the job - it's two channels but you only have to install parts for one if that's all you need. 3-5 W with a 6V6 up to 12 W or so with a 6550 (at higher supply voltage...). Just flip on over to the Tubelab forum...
  Reply With Quote
Old 26th April 2011, 08:43 PM   #3
diyAudio Member
 
Join Date: Feb 2011
thanks, i will, but does anyone have an answer for my questions?
  Reply With Quote
Old 26th April 2011, 11:02 PM   #4
diyAudio Member
 
tubelab.com's Avatar
 
Join Date: Jul 2005
Location: West Virginia panhandle
Laying out a PCB for a tube amp is a bit different than laying out a typical low voltage analog or digital board. The impedances involved in tube amps are much higher (hundreds of K ohms) as are the voltage and signal levels.

Trace to trace coupling is a real concern. So is trace to ground plane coupling. I make all my boards double sided but it is possible to make a single sided board for simple circuits. Avoid running high impedance traces, especially grid circuits next to or on the other side of ground or other signal traces.

If using a single sided board it might be wise to leave the heater wiring off the board and run them using twisted pair of wire. On a two sided board run the heater circuits as a differential pair, either side by side, or one on top of the other. The preamp tubes are much more critical than the output stages. In your guitar amps the venerable 12AX7 is especially prone to hum, and noise.

Ground plane fills are commonly used in many PCB's, but don't work here. You will get hum. I use a single point star ground with the star point being the negative pin of the last electrolytic in the power supply. The only place the power supply ground meets the amplifier ground is at this point. This is also where the chassis and power line input grounds get connected.

I must state that I have been laying out PC boards in my full time job and my side projects for over 30 years. It still took me several (like 10) attempts to make my first clean, hum free tube amp design. I have been doing tube amp PCB's for about 10 years now and I still take 2 or 3 passes to make one that I am completely happy with. Once you get a good design though, you can make zillions of them and they will all be the same.
__________________
Too much power is almost enough! Turn it up till it explodes - then back up just a little.
  Reply With Quote
Old 27th April 2011, 12:21 AM   #5
roline is offline roline  United States
diyAudio Member
 
roline's Avatar
 
Join Date: Sep 2010
Location: East Tennessee
Default Tube Amp PCB's

George is correct!, trace length will give you inductance, resistance and the length and closeness will give you a parasitic capacitor for coupling. Traces need to be wide enough to have low resistance and lower inductance with a minimum of 0.050" spacing to the next trace to prevent high voltage arcing (500V rule)and excessive capacitance that can effect cross talk and high frequency roll off. You must verify the signal phase in the line and the ones next to it to prevent positive feedback. Frequency capability will be determined by trace design, u-wave circuits are done on duroidal PCB's with a whole thick book of design rules! almost a black art.... properly designed PCB's are not a limitation for the power bandwidth requirements for audio amps.
Watch where you have high current flow to ensure proper grounding to prevent potential for ground loops and hum.
Always put heater supplies as twisted pair off of the board, I like to follow the 1/2" rule and keep them away from hi impedance points in the input stages.
If you do single sided you can use the toner transfer method and use eyelets for the feed throughs for wire connections and putting caps on the back side. I like to use single sided maximum ground plane with ample stitching to prevent ground loop potentials, it keeps the differential voltage potential in the u volts at most. If the design does not allow for that, star point ground is a must.

Designing PCBs is easier that designing IC's. With IC's I had to also worry about thermal feedback and parasitic MOS devices where the traces passed over the transistor tubs. With ceramic hybrid circuits, that was the most fun. We used gold conductors, screen printed laser trimmed 1% resistors, any mix of IC's, transistors, etc up to 6 conductor layers. The most complex one was a 6 conductor layer, 19 die 256 wire bond digital circuit for HP. I also did a swiss navy project of a ceramic hybrid power opamp for home audio. The biggest issue was getting the heat out through the alumina oxide substrate to the heat sink. It worked but not at the power level that was desired. The design was active crossover filters with minimum band pass phase shift distortion driving a quad amped speakers. A fun project but not practical, not as good as a tube amp....
Remember the PCB's will work within design limit constraints, but to the purists point to point wiring is optimal ($ilver wire, variable gage, teflon insulation, minimum cross capacatance, etc...) and will work in all design cases including UHF circuits. I agree with them, but find a good PCB to my liking.
Attached Images
File Type: jpg compactron 6gm6_6lb6_1a.jpg (170.2 KB, 211 views)
File Type: jpg P11903iyv72.jpg (517.9 KB, 204 views)
File Type: jpg kt88-2a.jpg (242.1 KB, 189 views)
__________________
SO many tubes, SO little time!!!

Last edited by roline; 27th April 2011 at 12:47 AM.
  Reply With Quote
Old 27th April 2011, 02:10 AM   #6
diyAudio Member
 
tubelab.com's Avatar
 
Join Date: Jul 2005
Location: West Virginia panhandle
Quote:
u-wave circuits are done on duroidal PCB's with a whole thick book of design rules! almost a black art....
That's what I do in my day job. I just did a 60 by 100 mm PCB that contained 650 parts and ran at frequencies from 100 MHz to 1.8 GHz, and cranked out 10 watts at those frequencies, and had a receiver too. THe board contained 10 layers using HDI technology. It's a different world from tube amps with BIG leaded parts. I set my minimum design rules for 50 mil traces on 50 mil spacing for tube amps, and many traces are wider than that.

PCB material has improved to where we don't use Duroid in any cost sensitive applications below 2 GHz. Your cell phone uses FR4 at 2 GHz. I recently laid out two identical RF power amps that make 20 watts of LTE at 2.4 GHz. One was done on FR4 and the other on Rogers 4350, a Duroid like composite for low loss at multi GHz frequencies. The microstrip lines were optimized for the individual materials. There are negligible differences between the two designs.

There are a dozen IC designers sitting around me doing IC's for 2 GHz applications. Those guys must learn to worship the simulator and "trust but verify" its results. I lay out the boards for the "verify" part once the chips get made.
__________________
Too much power is almost enough! Turn it up till it explodes - then back up just a little.
  Reply With Quote
Old 27th April 2011, 02:42 AM   #7
roline is offline roline  United States
diyAudio Member
 
roline's Avatar
 
Join Date: Sep 2010
Location: East Tennessee
That is one thing I love about this form, we have a broad range of experience to share. My highest frequency experience was in space flight electronics, TDRSS-IUS satellite project and shuttle communications at TRW. We even x-rayed the duroidal boards when mounted to the chassis to make sure there was perfect bonding, no air pockets. 1.5Ghz ALB1A bipolar IC process at Honeywell. I sat at a dumb terminal with a 300-baud acoustic coupled modem doing H-Spice simulations, eating reams of paper for the printouts for months on end prior to silicon layout design. Our design group had a layout check rule. If you found a layout error, it was worth one cold beer. On one large linear IC, I worked over a holiday weekend checking a layout to meet a deadline and earned a cold keg!
It was most challenging sitting at a probe station on an air table in a screen room doing device characterization and circuit analysis on the wafers. That was in the 80's, I sure that is old and obsolete by todays standards. What I did with ceramic hybrid circuits in the 70's is almost duplicated by surface mount PCB's today. Makes you wonder about tomorrow!
__________________
SO many tubes, SO little time!!!
  Reply With Quote
Old 27th April 2011, 03:04 AM   #8
diyAudio Member
 
Join Date: Feb 2011
ok, thanks a lot you guys, i really really apreciate the inputs, but this is getting way out of hand for this guy here.... Plus my project is really not that complex..
I had already thought about not running heater wires on the board, i guess my instincts were in the right place... double
But anyways, i read a lot about high impedance traces, but i am not too sure what they mean. Are those traces that have a high impedance on their own? Or traces which either end have a high impedance to ground (like a high resistor). Also, doesn't this mean a low current trace?
About the grounding, i read somewhere that would be a good idea to keep signal ground (negative) and earth ground (chassis, power input ground) separate. Even because my power supply is not center taped... Basically i wasnt planning on having ground planes...
keep in mind i am not professional, i am doing the pcb just because i have the equipament at work, so it would be easier...

Thanks a lot
  Reply With Quote
Old 27th April 2011, 12:41 PM   #9
diyAudio Member
 
tubelab.com's Avatar
 
Join Date: Jul 2005
Location: West Virginia panhandle
Quote:
It was most challenging sitting at a probe station on an air table in a screen room doing device characterization and circuit analysis on the wafers. That was in the 80's, I sure that is old and obsolete by todays standards.
It might be old and obsolete, but we still have an air table probe station. The probes have grown smaller and much more expensive over the years since our chips now have over a hundred bonding pads.

Quote:
But anyways, i read a lot about high impedance traces, but i am not too sure what they mean. Are those traces that have a high impedance on their own? Or traces which either end have a high impedance to ground
The impedance that a trace sees is defined by the circuitry. If there is a 100K plate resistor on your 12AX7 and it feeds a tone stack that adds up to 300K or so, then the circuit impedance to ground is 75K at some frequencies. It can be lower at other frequencies because of the capacitors in the tone stack. Because of this impedance a few pF of stray capacitance can transfer significant energy. The impedance of any circuit containing a variable resistance will vary with the adjustment of the pot. This is why some marginal layouts oscillate or hum at some pot settings and not others. The input impedance of the amp changes with what is plugged into it.

A guitar amp puts a lot of gain in a small box and places the whole thing on a vibration table (the speaker cabinet). Plan on multiple tries to make a perfect design. Patience and lots of time are required!!!! That's why my ultimate screamer is still not finished.
__________________
Too much power is almost enough! Turn it up till it explodes - then back up just a little.
  Reply With Quote
Old 27th April 2011, 02:18 PM   #10
diyAudio Member
 
Join Date: Feb 2011
Thank you....
  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
5W EL84 drive EL84 SE Audio_idiot Tubes / Valves 31 31st May 2012 08:38 PM
5w SE guitar Amp N9VOC Instruments and Amps 16 18th August 2010 05:34 PM
5W SE 2A3 monoplate for sell! widowmaker Swap Meet 0 29th February 2008 07:49 AM
SE 1.5W OTL audio3chung Tubes / Valves 81 19th July 2006 02:03 PM
MTM stand mount or floor stand garfieldcat Multi-Way 5 7th July 2004 12:45 AM


New To Site? Need Help?

All times are GMT. The time now is 09:48 AM.


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