active crossover power supply pcb check before making - diyAudio
Go Back   Home > Forums > Amplifiers > Solid State

Solid State Talk all about solid state amplification.

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 29th April 2006, 07:55 PM   #1
mazurek is offline mazurek  United States
diyAudio Member
 
Join Date: Aug 2005
Default active crossover power supply pcb check before making

This is my stab at making a +/- 12 volt power supply. There is a ground plane on top which is not shown, also standoff holes not shown. The motivation is to get a third 12 volt output for relays. The right side of the board will be one array of screw connector terminals, the aux out must be shorted in order to power and unmute the relays. The aux out is fully regulated by design to be flexible. For simplicity, I will be adding a solid state ac controlled relay to control the aux out, but a proper trigger switch or other control circuit could easily be implemented.

A ground isolating network was used. Fusing is design to be flexible.

I placed all the regulators on one side of the board for easy cooling. Regulator capacitance is 100 uF electrolytic. All my opamps are bypassed with ceramic later.

Sources:
-sound.westhost.com
-www.adireaudio.com
-www.diyaudio.com
-www.national.com
-My terrible electrical engineering book which is the most useless thing ever, and I want to send the author hate mail
Attached Images
File Type: png powerpcb.png (32.9 KB, 462 views)
  Reply With Quote
Old 29th April 2006, 11:10 PM   #2
diyAudio Moderator
 
pinkmouse's Avatar
 
Join Date: Apr 2002
Location: Chatham, England
A few things that I have learnt along the way from cleverer people than me.

The traces to the big caps should run directly to their pins with the thickest track you can get away with, and then from them the tracks should then get slightly smaller to run to the regulators. In fact, all the traces could be thicker than you have.

Don't worry about a ground plane, just get a nice star ground set up instead. Think where the currents are flowing, and try and keep the big ones away from the small.

Try and arrange your PCB so all the + terminals of the elcaps point in the same direction. It makes troubleshooting a lot easier.

The regs will work better if you have 0.1uf film caps bypasing the input rails.


Oh, and I hate non isometric tracks in PCB layouts!
__________________
Rick: Oh Cliff / Sometimes it must be difficult not to feel as if / You really are a cliff / when fascists keep trying to push you over it! / Are they the lemmings / Or are you, Cliff? / Or are you Cliff?
  Reply With Quote
Old 30th April 2006, 12:56 AM   #3
infinia is offline infinia  United States
diyAudio Member
 
infinia's Avatar
 
Join Date: May 2005
Location: Californication
Hi
1) Traces from C3 should run directly to C2's terminals (2pl)
also ground plane connection conn close to C2 terminals.
reason- isolate high current noisy traces i.e. rectifiers to big caps to minimize noise coupling to pass transitor inputs.
2) Don't use ground plane over high current noise traces same reason as 1).
3) Up date reference designators ( some repeats) reason- all components on same pcb should be unique.
__________________
like four million tons of hydrogen exploding on the sun
like the whisper of the termites building castles in the dust
  Reply With Quote
Old 30th April 2006, 01:45 AM   #4
infinia is offline infinia  United States
diyAudio Member
 
infinia's Avatar
 
Join Date: May 2005
Location: Californication
Hi
replace my last post with this (edit timed out)


1) Traces from C3 should run directly to C2's terminals (2pl).
In addition ground connection conn close to C2 terminals.
Reason- isolate high current noisy traces i.e. rectifiers to big caps to minimize noise coupling to 3-T regs inputs and ground out.

2) Don't use ground plane over high current noise traces same reasons as 1). and capacitive coupling thru bottom to top into ground plane.

3) Run high current noisey traces from one big cap then to last one C2 not in between them. Also run these traces parallel to keep the loop area small. Mimics twisted wire. Don't really need fat traces if you know where to pick off traces.

4) Update reference designators. ( some repeats) Reason- all components on same pcb should be unique.

Agree with Pinkmouse on this. You don't really need a ground plane on this board, can actually cause more harm than good. Point to point is better for filters and regs.


I want one when you're finished! For active filter PCB Group Buy?
__________________
like four million tons of hydrogen exploding on the sun
like the whisper of the termites building castles in the dust
  Reply With Quote
Old 30th April 2006, 01:55 AM   #5
mazurek is offline mazurek  United States
diyAudio Member
 
Join Date: Aug 2005
Thanks for the tips, I will be working on this more after I get more "real work" done.

"I want one when you're finished! For active filter PCB Group Buy?"

That's the idea in general, active filters are popular here these days. The other group buy power supply is probably a better design, but an aux out would be handy, as well as some decisions that make wiring it up easier for me at least.
  Reply With Quote
Old 30th April 2006, 05:10 AM   #6
infinia is offline infinia  United States
diyAudio Member
 
infinia's Avatar
 
Join Date: May 2005
Location: Californication
Quote:
Originally posted by mazurek
The other group buy power supply is probably a better design, but an aux out would be handy, as well as some decisions that make wiring it up easier for me at least.
Maybe a better design or just different objectives and tradoffs like any thing "beauty in is the eye of the beholder". Have you thought of using a more flexible 3 term adjustable regulator. The noise performance is much better than fixed.
P.S. Like to hear more on your muting concepts. I have a simple idea of using a 24V relay coil sensing the +/- 12 V supplies with its NC contacts shorting the outputs of the active filters outputs.
__________________
like four million tons of hydrogen exploding on the sun
like the whisper of the termites building castles in the dust
  Reply With Quote
Old 30th April 2006, 09:18 AM   #7
AndrewT is online now AndrewT  Scotland
diyAudio Member
 
Join Date: Jul 2004
Location: Scottish Borders
Hi,
Quote:
The regs will work better if you have 0.1uf film caps bypasing the input rails
are these the C3 that appear on your layout?

The two C3a, C3b need to be very close to the input pins of the respective regulators. I really mean very close, even mounted on the back of the pcb direct to the reg pins. Similarly the output pins can benefit from close coupling the cap bypasses.
If you change to 317/337 types the close coupling seems to be slightly less important but still worth doing.
__________________
regards Andrew T.
  Reply With Quote
Old 30th April 2006, 02:13 PM   #8
mazurek is offline mazurek  United States
diyAudio Member
 
Join Date: Aug 2005
Quote:
Originally posted by infinia


Have you thought of using a more flexible 3 term adjustable regulator. The noise performance is much better than fixed.
P.S. Like to hear more on your muting concepts. I have a simple idea of using a 24V relay coil sensing the +/- 12 V supplies with its NC contacts shorting the outputs of the active filters outputs.
I thought about the adjustable regulators, if its a good idea, I'll use them. I originally was using them in my pcb design, but got annoyed at the amount of space it takes to put in a potentiometer. Would it be ok for your use to put them in and use a fixed resistor network? I suppose that AndrewT's comment about capacitor proximity on the lm317 might mean there is more room to play with around the regulators.

I was basing a lot of the relay stuff on Rod Elliot's comments in his P05A project. Supposedly the relays can add noise into a regulated power supply when operating, if this is true, could maybe fix it with some diodes and capacitors. The other factor would be ensuring that they cut off in time. The relays I'm using have a 10% min dropout voltage, and 5ms must release time. I suppose you could play with your supply capacitance until the relays open before you hear voltage drop. I don't have gear detailed enough to measure what part of the voltage drop yields the most audible pop. I was thinking a regulator for the relays is a good idea, because then you can have separate voltages for relays and opamps without adjusting some pot to get the voltage you want for the amount of relays you have.


Quote:
Originally posted by AndrewT
Hi,
are these the C3 that appear on your layout?

The two C3a, C3b need to be very close to the input pins of the respective regulators. I really mean very close, even mounted on the back of the pcb direct to the reg pins. Similarly the output pins can benefit from close coupling the cap bypasses.
If you change to 317/337 types the close coupling seems to be slightly less important but still worth doing.
Thanks for the suggestions, I'll work on fixing that. Right now I have a ceramic decoupling capacitor before the input as well as the electrolytic ones. Can I do with a C0G ceramic across the inputs instead of a film. Film capacitors are very large, and from the specs it seems that ceramic are better with HF anyhow? Or I could find room for both, I suppose the polystyrene capacitors are a little smaller than the polypropylen I was looking at.

By the way AndrewT, I know you told me earlier about how ground planes weren't always good, it was really a matter of my laziness that I used them on this (didn't want to spend the time drawing ground traces). I will fix it though.
  Reply With Quote
Old 30th April 2006, 03:03 PM   #9
infinia is offline infinia  United States
diyAudio Member
 
infinia's Avatar
 
Join Date: May 2005
Location: Californication
Quote:
Originally posted by mazurek


1) I thought about the adjustable regulators, Would it be ok for your use to put them in and use a fixed resistor network?


2) The relays I'm using have a 10% min dropout voltage, and 5ms must release time. I suppose you could play with your supply capacitance until the relays open before you hear voltage drop. I don't have gear detailed enough to measure what part of the voltage drop yields the most audible pop.


3) Right now I have a ceramic decoupling capacitor before the input as well as the electrolytic ones. Can I do with a C0G ceramic across the inputs instead of a film. Film capacitors are very large, and from the specs it seems that ceramic are better with HF anyhow? Or I could find room for both.

1) Pot's are very nice, but don't really need them here.

2) I expect opamps start to sound ugly around +/- 5V, but it depends on the particular ones used. I'm not sure about relay's causing noise or other issue's in steady state operation. For relay's with a tighter dropout, timing should not be a issue as long as they cutout/turn on above the opamps dropout voltage, they should be fine. If you want you can sense at Vreg input instead and use a suitable RC filter to get within relay coil specs.
YMMV

3) My sense around here "audiophile land" that caps that are really sticky issue for some people. Rightly or wrongly the "right"selection depends on their application. Some people unnecessarily avoid ceramics like the plague. C0G are really the best ceramics in lower values, but are too large and costly when values approach .01uF or greater. As you pointed out ceramics when used as supply bypasses (low impedance & low gain) are a better choice, even heaven forbid the X7R type. On the other hand I would not use them (X7R) in opamps filters or in the signal path (high gain/impedance). Again whatever floats your boat.
__________________
like four million tons of hydrogen exploding on the sun
like the whisper of the termites building castles in the dust
  Reply With Quote
Old 30th April 2006, 03:41 PM   #10
mazurek is offline mazurek  United States
diyAudio Member
 
Join Date: Aug 2005
Quote:
Originally posted by infinia


2) I expect opamps start to sound ugly around +/- 5V, but it depends on the particular ones used. I'm not sure about relay's causing noise or other issue's in steady state operation. For relay's with a tighter dropout, timing should not be a issue as long as they cutout/turn on above the opamps dropout voltage, they should be fine. If you want you can sense at Vreg input instead and use a suitable RC filter to get within relay coil specs.
YMMV

It seems to me to be a tradeoff, I agree I don't know how a relay could cause steady state (unless it induces oscillation when switched). With only two supply rails you save money on regulators, but you need better relays and a more application specific relay electrical network. My approach requires a more expensive power supply, but I'm able to use any number of $1.50 relays without tuning the electrical network (most 24v signal level relays I saw on digikey were 5 dollars a piece, and I'm using one relay per output). On the other hand, I also require a 17 dollar bipolar solid state relay for ac sensing, but I figure I could use this power supply to add a 555 timer or other cool trigger in/out features eventually (and make this stuff pro grade).

Also on the subject of relay muting etc, I believe I will try shorting the input also to prevent oscillation during power on transients (I realize I need to fix the feedback networks to make the circuit oscillation proof, but the only time it has ever oscillated was during startup transients). That combined with eventually putting in a 555 timer should make the circuit bullet proof I think.

I should have a revision out by tonight. Got to do my homework so I can keep my grad stipen though.
  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
single-ended power supply (12V) based active crossover r_s_dhar Car Audio 8 15th April 2011 08:44 PM
help powering 41hz amp3 and an active crossover on the same supply Nemophyle Class D 12 28th February 2007 08:59 PM
active crossover power supply pcb check before making (was in wrong forum) mazurek Power Supplies 0 29th April 2006 10:25 PM
active crossover check before finalization mazurek Multi-Way 0 30th December 2005 11:04 PM
Check out this power supply? planet10 Everything Else 3 18th September 2003 08:49 PM


New To Site? Need Help?

All times are GMT. The time now is 09:01 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