XO3: The 1/2/3-way Active Crossover/filter PCB. Design suggestions welcome! - Page 2 - diyAudio
Go Back   Home > Forums > Source & Line > Analog Line Level

Analog Line Level Preamplifiers , Passive Pre-amps, Crossovers, etc.

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 3rd November 2010, 03:42 AM   #11
diyAudio Member
 
Join Date: Mar 2007
Location: California
There is no fundamental, that might be power supply noise or something similar.

This measurement was done using swept sine or white noise input, can't recall which. This is the distortion spectrum, not the noise floor plus some harmonics of a single tone.

-Charlie

Last edited by CharlieLaub; 3rd November 2010 at 03:48 AM.
  Reply With Quote
Old 3rd November 2010, 03:45 AM   #12
diyAudio Member
 
abraxalito's Avatar
 
Join Date: Sep 2007
Location: Hangzhou - Marco Polo's 'most beautiful city'. 700yrs is a long time though...
Blog Entries: 98
Send a message via MSN to abraxalito Send a message via Yahoo to abraxalito Send a message via Skype™ to abraxalito
I'm curious - how can you get a THD reading without a stimulus tone? I agree its odd there's no notch apparent on that plot.
__________________
It doesn't have to take the form of a conspiracy, rather a consensus... James H Kunstler
  Reply With Quote
Old 3rd November 2010, 03:48 AM   #13
diyAudio Member
 
Join Date: Mar 2007
Location: California
I used STEPS, a software for measuring distortion products. It is part of the ARTA suite of measurement software.

If you want to learn how it works, read the technical literature in the manual, found at this link:
http://www.fesb.hr/~mateljan/arta/do...ser-manual.pdf

-Charlie
  Reply With Quote
Old 3rd November 2010, 03:58 AM   #14
diyAudio Member
 
abraxalito's Avatar
 
Join Date: Sep 2007
Location: Hangzhou - Marco Polo's 'most beautiful city'. 700yrs is a long time though...
Blog Entries: 98
Send a message via MSN to abraxalito Send a message via Yahoo to abraxalito Send a message via Skype™ to abraxalito
Thanks for the info - I've only read the explanation briefly, its still doing something very similar to a traditional distortion analyser, rather removing the fundamental by multiplying the output obtained by the original tone. So no notch filter is needed. But there's still a fundamental frequency being applied.

<edit> ah, I see what you mean now - its a plot of distortion harmonics against frequency - the stimulus is not a fixed but swept frequency. Got it - in which case yeah, the power supply noise is terrible!
__________________
It doesn't have to take the form of a conspiracy, rather a consensus... James H Kunstler

Last edited by abraxalito; 3rd November 2010 at 04:14 AM.
  Reply With Quote
Old 18th January 2011, 12:29 PM   #15
diyAudio Member
 
s3tup's Avatar
 
Join Date: Dec 2004
Location: Israel
Some updates...

- Parametric EQs instead of notches
- Ballanced/regular input
- Ability to reroute the sections as you wish
- Jumper for 3-way/2-way operation
- Built-in traces for default 2-way/3-way configurations
- Onboard PSU with LM337/317 regs, propertly implemented (i hope so), with huge caps on output (can accomodate 10,000uf caps).
- Redesigned the whole layout of everything
- Had to go back to 2 SMT resistors instead of 4 in configurable places - still, you can stack-up the SMT resistors to get the right resistance.
- The "global" EQ with reroutinc could be used as fourth EQ at output...
That gives us ability to make... well, 5-way crossover
- I like the Altium's feature of "rounded" traces - they look way better on PCB.

Drawbacks... Well, i can't find a place to put the mounting holes...
Attached Images
File Type: jpg BoardV2_A.jpg (312.6 KB, 399 views)
File Type: png PCB_Naked_Top_V2A.png (150.4 KB, 305 views)
File Type: png PCB_Naked_Bottom_V2A.png (174.2 KB, 279 views)
File Type: png PCB_Top_V2A.png (168.1 KB, 268 views)
File Type: png PCB_Bottom_V2A.png (159.3 KB, 260 views)
File Type: jpg JumperV2_A.jpg (57.1 KB, 55 views)
Attached Files
File Type: pdf xo3_V2_SCH.pdf (99.0 KB, 99 views)
  Reply With Quote
Old 18th January 2011, 04:08 PM   #16
diyAudio Member
 
Join Date: Mar 2007
Location: California
Quote:
Originally Posted by s3tup View Post
Some updates...

- Parametric EQs instead of notches
- Ballanced/regular input
- Ability to reroute the sections as you wish
- Jumper for 3-way/2-way operation
- Built-in traces for default 2-way/3-way configurations
- Onboard PSU with LM337/317 regs, propertly implemented (i hope so), with huge caps on output (can accomodate 10,000uf caps).
- Redesigned the whole layout of everything
- Had to go back to 2 SMT resistors instead of 4 in configurable places - still, you can stack-up the SMT resistors to get the right resistance.
- The "global" EQ with reroutinc could be used as fourth EQ at output...
That gives us ability to make... well, 5-way crossover
- I like the Altium's feature of "rounded" traces - they look way better on PCB.

Drawbacks... Well, i can't find a place to put the mounting holes...
Your approach shows a lot of potential. Here are some comments:

1. Power supply - I am not sure why you have such large caps on the output of the regulators. When the board is powered down, the supply side voltage will collapse before the caps on the output drain, so you will need to use the protection diodes as shown in the LM317 datasheet. If not, you risk damage to the ICs. The DC is already REGULATED. It's not like the caps on a power supply. You only need enough capacitance on the output to stabilize the regulator IC, as little as 10uF. If you use a value under 25uF, you don't need the protection diodes.
Also, you might want to move the voltage regulators to the edge of the board and allow room for the addition of a heat sink to each.

2. It's nice that you included a delay stage. Unfortunately, a single first order section for delay is extremely limited in what it can do. What is really needed is flat delay throughout the crossover region, especially above the crossover frequency. Typically you need multiple second order stages to accomplish this except for the lowest crossover frequencies, because the amount of delay that you can achieve by a single stage for all-pass corner frequencies of a couple of kHz (this is minimum) is too small. Can you make it possible to use multiple all-pass stages?

3. Will the board come assembled? If not, I think that the SMD resistors (and there seems to be a lot of them) are difficult for a DIYer to solder on without some advanced tools/skills. If the board is pre-assembled, how does one change the crossover frequencies and other components/parameters?

4. Without mounting holes, and with components right up to the edge all around, how would one actually mount/use the board? It can't just hover in place!

-Charlie
  Reply With Quote
Old 18th January 2011, 04:37 PM   #17
diyAudio Member
 
s3tup's Avatar
 
Join Date: Dec 2004
Location: Israel
Thanks for feedback!

1. the "huge" caps are there for a reason. I've used a resistor in series with regulator to detach it's output from the board's supply.
Here is why:
- Regulators don't like big caps on their outputs
- Output impedance of refulator is curvy, frequency dependant...
- The regulator has slow reaction to voltage surges at input, therefore it's PSRR/rejection worsens as the frequency of noise at it's input rises. The C-R-C filter on reg's output cuts all the noise that have passed the regulator.

The regs are there to "charge" the large caps with constant voltage.
The "bypass" resistor should limit reverse current flowing thru regulator backwards on shutdown... but i'll add the diodes too, there is a space for them.

2. I agree on this point. The problem is board space, there isn't much space to put 8 stages (4 opamps) (which are sufficient for regular filter instances for correction of woofer-midrange offset).
The way i think of it is
- the All-pass section can be tuned low, and therefore give a larger delay, but gonna add some phase shift in driver's passband. The all-pass filters are there to fine-tune the phase alignment of drivers in both physical and electonic phase response. They couldn't be coherent, but their phase should match around crossover frequency to give a nice "null" on frequency resonse when one of the drivers is being connected in reverse polarity.
I need to investigate this theory in advance...

3. No, the board will be "bare". If you look closely, all the SMT components have "advanced" footprints which have regular thru-hole pads too. It's up to you to decide which components you'll use.

4. Yeah... i know... it's bad... but these holes are almost 8mm in size! As much as single opamp I'll search the way to squeeze them. at least 3 in count...


I'm limited in board size, as i want them cheap, and there is limit 100x100mm for each board in company i plan to order from.

There are electrolytic power supply caps under each opamp - you could use them as "standoffs"
  Reply With Quote
Old 18th January 2011, 05:25 PM   #18
diyAudio Member
 
s3tup's Avatar
 
Join Date: Dec 2004
Location: Israel
Ah, about the heatsink space and placement of PSU at the edge.
When i put the PSU in center, i get short and equal lengh traces to all of the components, and as a plus - the "star" topology power supply, where is no interference between power supply traces of each opamp. I think it is a good decision
Even the star ground is possible, but i doubt it will be better than double-sided pours. I can slice the layers to create it.

Last edited by s3tup; 18th January 2011 at 05:27 PM.
  Reply With Quote
Old 19th January 2011, 06:16 AM   #19
mt490 is offline mt490  Australia
diyAudio Member
 
Join Date: Jun 2010
I'm interested to know what selections of SMD resistors you're designing around. In my research there only seems to be one type of thin film commonly around that is actually economical to use but it's not stackable.

I suppose they're not that expensive if I can put together a bulk order with other items.

Last edited by mt490; 19th January 2011 at 06:23 AM.
  Reply With Quote
Old 19th January 2011, 10:20 AM   #20
diyAudio Member
 
s3tup's Avatar
 
Join Date: Dec 2004
Location: Israel
Thick film are cheap and common. Could be bought over ebay for 100$ for whole E24 values set, 100pcs for each value.
20pcs/value goes for 16$ (don't know whether it's thick or thin film, but the price tells us it's thick)
170 Values 0805 SMD Chip Resistor COMPLETE Kit (0R~10M) - eBay (item 170470904131 end time Feb-03-11 15:01:03 PST)

Thin film are pricey.
Could you share yours thin film resistors source?
Thanks.
  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
Active Digital Loudspeaker Crossover - Part Suggestions solicitation speakerguy79 Digital Line Level 0 2nd December 2009 04:26 PM
active crossover state variable filter Altecboy Multi-Way 0 23rd June 2009 01:31 PM
<AUS> B3S active crossover design - Value woofer suggestions? TheSeekerr Multi-Way 0 25th August 2008 12:46 PM
Generic active filter / crossover... board... thingie... JohnR Parts 5 14th October 2002 10:13 PM
Opamps for active crossover/filter harvardian Solid State 10 3rd January 2002 02:01 AM


New To Site? Need Help?

All times are GMT. The time now is 05:29 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