ilfungo

Vactrol VTL5C4 / 2 matching
« on: July 24, 2010, 04:52:48 AM »
I just finished my first D-AOC
Sound wonderful but can not get it to work correctly in stereo ..
I'm sure the problem is Vactrol VTL5C4 / 2 matching
anyone knows how to do?
how can I calibrate stereo mode?
Thanks


ilfungo

Re: Vactrol VTL5C4 / 2 matching
« Reply #1 on: July 26, 2010, 06:14:52 AM »
some Ideas? ??? :'(

Script

Re: Vactrol VTL5C4 / 2 matching
« Reply #2 on: July 27, 2010, 04:54:43 AM »
Best thing would be to do a search. I'm not an expert on this at all, but I'll have two opto units coming up on the bench. Info on matching whatever type of opto for stereolinking is all scattered and seems to differ between units. Anyway, here are some thoughts:

(1) matching often seems to involve trims: in parallel to or before the LED (e.g., LA2A).

(2) 'stereolinking' the LEDs (before and after) reduces differences between L and R (i.e., D-AOC)
--> search: summing side-chains together

(3) (Forrsell only and unconfirmed, as it seems:) someone suggested raising the 4.99k and 1k resistors (series and shunt) to 49k and 10k before the opto's resistor (here in series, see schamatic) to reduce difference between individual optos. However, a different response curve is to be expected.

(4) To match the optos themselves, I think we'd want to measure their resistances at several different current draws, plus using test signals of different durations to compare attack/release behaviour, and then compare the convex and concave parts of the plotted graphs--a process that sounds like sure headache to me... The desription of a strange thingy for Passive LDR Attenuation mentions the need for multiple optocouplers (25 or more) to find a pair that matches--more or less. I wouldn't take this info at face-value, but it gives an idea.

(5) Silvas reports (D-AOC help thread) that his input transformers created a slight difference btw. L and R. So matching starts with the periphery.

(6) I guess that for LR signals at mild levelling/compression settings we can easily get away with 'stereolinked' optos that are less than ideally matched, so a 'stereolink' switch does make sense after all (maybe rename it 'SC link'). At high settings though I expect differences to become annoying: audible shifts in stereo image/image wandering and bouncing off centre.

(7) To avoid off-centre bouncing, we could feed the unit with MS signals to 'successfully' shift differences between optos from left/right to mono/stereo...

Exhausted for now :) Anyone has more ideas/better advise?

« Last Edit: July 27, 2010, 08:42:50 PM by Script »

Silvas

Re: Vactrol VTL5C4 / 2 matching
« Reply #3 on: July 27, 2010, 12:12:18 PM »

(5) Silvas reports (D-AOC help thread) that his input transformers created a slight difference btw. L and R. So matching starts with the periphery.


Not on opto matching, but on L-R level matching !

My optos seems to be tightly matched, maybe i got lucky or maybe the kubi mod does improve matching, i don't know.
Camilo Silva F.
Camilo Silva F. Mastering
Chia, Colombia

ilfungo

Re: Vactrol VTL5C4 / 2 matching
« Reply #4 on: July 27, 2010, 01:17:43 PM »
Script thanks for the info... ;)
might better explain step 3?

"3) (Forrsell only and unconfirmed, as it seems:) someone suggested raising the 4.99k and 1k resistors (series and shunt) to 49k and 10k before the opto's resistor (here in series, see schamatic) to reduce difference between individual optos. However, a different response curve is to be expected."

Regarding D-AOC which resistors should replace ?
Thanks

Script

Re: Vactrol VTL5C4 / 2 matching
« Reply #5 on: July 27, 2010, 11:10:40 PM »
Hi,

about 3: no I can't explain... But it does in this post:
http://www.groupdiy.com/index.php?topic=16016.msg321620#msg321620
(Sorry for not having included it above.)
The Forssell schematic is here:
http://www.diyfactory.com/projects/forsselllimiter/ffosc.htm
What those two resistors do is explained in the walkthrough by the designer (i.e., change balance between 'feedforward'/'feedback' behaviour). 

On the D-AOC: Looking at the schematic at a proper sized screen now (not netbook), the values of R102 and R103 look suspiciously close to what it suggested in that above post... As I said I'm not an expert in electronics, probably will run the D-AOC in MS, and maybe better shut up now ;)


 

Related Topics

  Subject / Started by Replies Last post
13 Replies
5050 Views
Last post December 10, 2004, 05:55:22 PM
by CJ
11 Replies
7234 Views
Last post May 21, 2014, 02:26:46 PM
by tzman
8 Replies
1617 Views
Last post May 04, 2005, 08:19:25 AM
by analag
19 Replies
4940 Views
Last post February 06, 2011, 07:20:09 PM
by zayance