Go Back   Home > Forums > >

Digital Line Level DACs, Digital Crossovers, Equalizers, etc.

Another (ultimate?) Behringer mod - what can  do to DEQ2496  (PART 1)
Another (ultimate?) Behringer mod - what can  do to DEQ2496  (PART 1)
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
Old 31st August 2021, 08:55 AM   #41
DBMandrake is offline DBMandrake  Scotland
diyAudio Member
 
DBMandrake's Avatar
 
Join Date: Nov 2010
Location: Glasgow, UK
You might have a point about the bias if there is insufficient bias, although I believe you can get non-polar tantalums as well.

Has anyone measured the bias voltage across these 47uF coupling caps to see if there is any ? If the originals are polarised aluminium electrolytics with no bias that's not great either.. and fitting the same would not be a good idea as they would also fail eventually. Whatever goes in should be non-polar if there is little or no bias.

I agree with bypassing it with a 0.1uF film cap - if I can physically squeeze it in. Electrolytics with no bypass directly in a signal path that can supposedly pass audio up to 96Khz is a big no-no, there's no way it will be flat up to those high frequencies without additional bypassing and I'm surprised nobody in this thread has addressed this when trying to perform mods to improve sound quality.

At the moment I don't have a working scope so I'm going to have to sort that out before I can properly look at this. Maybe I will go a bit further than just trying to fix the channel imbalance. At the moment my stereo setup can work fairly well without any EQ so I can pull the DEQ out of the system and pick away at it carefully.

Is there a consensus here that replacing the various mentioned signal path capacitors alone is enough to improve the quality significantly without trying to replace the (difficult to remove) opamps ?
__________________
- Simon

Last edited by DBMandrake; 31st August 2021 at 09:07 AM.
  Reply With Quote
Old 31st August 2021, 12:08 PM   #42
bentoronto is offline bentoronto  Canada
diyAudio Member
 
Join Date: Jan 2008
Location: Toronto and Delray Beach, FL
Another (ultimate?) Behringer mod - what can  do to DEQ2496  (PART 1)
For 10 years, I've been curious about the hunt for perfection in the DCX by mods to the analog sections. Lots of despair about super-sonic jitters but not much evidence of audible shortcomings or audible improvement. REW testing I've ever done of FR, noise, and THD seem extravagantly excellent to my perspective.

How many people are still using the analog input to the DCX or DEQ instead of AES-SPDIF/coax digital input? Or can't find $50 for an upstream ADC instead of this risky mod?

With 1% distortion in speakers on a good day, does .05% distortion - and only north of 10kHz - in the Behringers impair sound quality?

Having said that, I greatly admire the thought shown in this thread. But also the generosity of posters in sharing their knowledge and being helpful to others. Thanks.

B.
__________________
HiFi aspirations since 1957. "When the toilet paper of Experience is depleted, the backside of Reason goes unwiped"
  Reply With Quote
Old 31st August 2021, 02:09 PM   #43
DBMandrake is offline DBMandrake  Scotland
diyAudio Member
 
DBMandrake's Avatar
 
Join Date: Nov 2010
Location: Glasgow, UK
Quote:
Originally Posted by bentoronto View Post
For 10 years, I've been curious about the hunt for perfection in the DCX by mods to the analog sections. Lots of despair about super-sonic jitters but not much evidence of audible shortcomings or audible improvement. REW testing I've ever done of FR, noise, and THD seem extravagantly excellent to my perspective.
I've been happy with the sound quality of mine for years and have found the measurements to be pretty good overall, for me the issue is channel imbalance between the two channels that has developed as it has aged, as well as a shift in frequency response of one channel relative to the other. A small shift but it is quite audible.
Quote:
How many people are still using the analog input to the DCX or DEQ instead of AES-SPDIF/coax digital input? Or can't find $50 for an upstream ADC instead of this risky mod?
I use only analogue inputs and outputs - the amplifier it is connected to is purely analogue.
Quote:
With 1% distortion in speakers on a good day, does .05% distortion - and only north of 10kHz - in the Behringers impair sound quality?
Depends what form that distortion is in. If it was .05% harmonic distortion then no you wouldn't hear it at all and yes speakers are far worse than that for harmonic distortion.

However intermod distortion between high treble frequencies (which on their own might be inaudible to the listener) generating audible products in the midrange could certainly be heard with the right source material, and would sound a bit like aliasing - I think that's what the modifications earlier in this thread are trying to address, and I don't have a problem with that.

Not sure if I'm brave enough to attempt it though, I may just stick to fixing the channel imbalance and getting it back to factory original performance.
__________________
- Simon

Last edited by DBMandrake; 31st August 2021 at 02:12 PM.
  Reply With Quote
Old 8th September 2021, 08:26 AM   #44
DBMandrake is offline DBMandrake  Scotland
diyAudio Member
 
DBMandrake's Avatar
 
Join Date: Nov 2010
Location: Glasgow, UK
Now that I have a scope again I spent some time troubleshooting the I/O board and found the channel imbalance problem does indeed seem to be electrolytic coupling caps - specifically two of them in the right channel.

First I did some accurate gain comparisons between left and right channels and the relay bypass. As expected the relay bypass is a perfect gain match, so the figures below are relative to the relay bypass:

+12dBu mode

40hz
Left: -0.46dB
Right: -3.07dB

1Khz
Left: -0.36dB
Right: -2.06dB

10Khz
Left: -0.43dB
Right: -1.70dB

So yeah, 2.6dB down on the right channel at 40hz vs left, not good...

While there was a fraction of a dB difference in overall level in +22dBu mode the trends were all the same.

Next I wanted to find out how much of that loss was input stage vs output stage, so I used the width processor set to minimum width which effectively turns the signal path into mono mode.

I was then able to feed the test signal into only the left input and compare left and right outputs. (I also tried it feeding only into the right input - same result)

In this mode the channels matched each other within 0.05dB so that confirmed all the channel imbalance is on the input side and the output is perfectly OK. This tallies with the VU meters to be honest because the VU meters show an imbalance on the input signal as well.

I started tracing the circuit from the input connectors referring to the DCX2496 schematic posted earlier in the thread unfortunately there are large differences in the input stage between my DEQ2496 and the circuit given earlier, for example it turns out both signal + and - from the XLR connector go directly to two 47uF coupling capacitors while none are shown in the circuit above...

And funnily enough the coupling capacitor on the positive input of the right channel is the main fault, with a loss of about 2dB across it at 40hz - when I bypass this with a 10uF polypropylene cap I have lying around the level goes up by about 1.8dB, so given this is 1/4 of the correct value of the original cap and it still makes this much difference the electrolytic must either be down to less than 1/10th of its correct value or have a really high ESR... either way it's stuffed.

There is a second 47uF next to it that is also faulty although the loss here is less - around 0.5dB, and bridging it with the 10uF cap still makes a significant difference. So also faulty.

On the picture below C19 is the one that is really bad and C18 is the not so bad but still faulty cap:

Another (ultimate?) Behringer mod - what can  do to DEQ2496  (PART 1)-img_4898-jpeg


Bridging the 10uF across the same caps in the left channel (or indeed any other caps) makes zero difference, so it seems I only have two faulty caps. I will of course replace all of these 47uF caps, even the ones on the phone jack outputs I don't use...

The next question is what to replace them with - I've been swayed away from tantalum but I'm thinking that if I can find some that will physically fit I really ought to replace them with non-polar electrolytics ?

Of the 15 or so 47uF caps more than half of them, including both faulty ones, are operating with zero DC bias. This is obviously not ideal for a polarised electrolytic, especially one going to the input socket.

A few have 2.8v across them and a couple 14.88 across them. Are there any drawbacks to simply replacing all of the 47uF 25v (and the one 10uF) with non-polar electrolytics ?

These ones are 12mm tall by 5mm diameter but there is plenty more height available and they could probably fit 6mm diameter without too much trouble. I haven't yet looked to see what's available.

I'm also considering bypassing the ones being used for coupling with 0.1uF, but what would be good to use but is still small enough to fit under the board and be good for audio ? Polypropylene ?

I've also noticed a bodge wire on my board - while I did replace the relays on mine and had some difficulty with the PCB I have no recollection of having to bypass a faulty trace - does anyone else's board have this blue wire ??

Another (ultimate?) Behringer mod - what can  do to DEQ2496  (PART 1)-img_4897-jpeg

The foam pads on the bottom have completely rotted and gone hard and crumbly, so I've got a bit of board clean up to do and new foam spacers to fit before it goes back together...
Attached Images
File Type: jpeg IMG_4897.JPEG (940.9 KB, 171 views)
File Type: jpeg IMG_4898.JPEG (786.1 KB, 167 views)
__________________
- Simon

Last edited by DBMandrake; 8th September 2021 at 08:52 AM.
  Reply With Quote
Old 8th September 2021, 10:13 AM   #45
DBMandrake is offline DBMandrake  Scotland
diyAudio Member
 
DBMandrake's Avatar
 
Join Date: Nov 2010
Location: Glasgow, UK
Ok never mind, I can't find any non-polar electrolytics in that value/size range from any of the suppliers I use so I guess I'll just have to get decent quality polar types!

I am planning to bypass the ones directly in the signal path with 0.1uF metalized polyester film types as well though if I can fit them under the board, and I'll do some sweeps to see if this does affect the high frequency response in any noticeable way.
__________________
- Simon

Last edited by DBMandrake; 8th September 2021 at 10:19 AM.
  Reply With Quote
Old 8th September 2021, 01:53 PM   #46
bentoronto is offline bentoronto  Canada
diyAudio Member
 
Join Date: Jan 2008
Location: Toronto and Delray Beach, FL
Another (ultimate?) Behringer mod - what can  do to DEQ2496  (PART 1)
Your left channel seems to be about 2 dB louder (readily addressed by gain management somewhere) and there is an inaudible variation with frequency about plus and minus .7 dB between channels.

Not something I'd lose sleep over, or want to warm up my soldering iron and get my magnifying glass, even if I could reliably measure it.

B.
__________________
HiFi aspirations since 1957. "When the toilet paper of Experience is depleted, the backside of Reason goes unwiped"

Last edited by bentoronto; 8th September 2021 at 01:55 PM.
  Reply With Quote
Old 8th September 2021, 02:31 PM   #47
DBMandrake is offline DBMandrake  Scotland
diyAudio Member
 
DBMandrake's Avatar
 
Join Date: Nov 2010
Location: Glasgow, UK
Quote:
Originally Posted by bentoronto View Post
Your left channel seems to be about 2 dB louder (readily addressed by gain management somewhere) and there is an inaudible variation with frequency about plus and minus .7 dB between channels.

Not something I'd lose sleep over, or want to warm up my soldering iron and get my magnifying glass, even if I could reliably measure it.

B.
Each to their own of course, but for me a stereo gain imbalance of 2dB is considerable, pulling the stereo image quite a long way off centre and subjectively also giving the impression of "compressing" the perceived stereo width more towards mono.

I noticed the subjective effects of the problem long before I ever took any measurements - I only took the channel balance measurements because I noticed a fairly obvious problem in normal music playback, not the other way around.

And a 1.7dB up tilt from bass to treble in the right channel is also very obvious and noticeable and certainly explains why the speakers are sounding a little thinner than they should lately.

While I have a workaround in place to balance the levels and frequency response of the two channels to some degree, it's quite inconvenient and not entirely accurate as I've had to put the DEQ2496 in stereo EQ mode where you can adjust the EQ of left and right channels independently, (normally one set of EQ adjustments is shared and mirrored between both channels in "Dual Mono" mode) but this is a PITA when trying to make any adjustments especially if the EQ on both channels has to be set differently to get the same result.

In short, I want it fixed, and there's no way I would leave nearly dead coupling capacitors in place now I know what the problem is and how easily it can be fixed.

The capacitors are on order and will be fitted as soon as they arrive...
__________________
- Simon
  Reply With Quote
Old 10th September 2021, 08:44 AM   #48
DBMandrake is offline DBMandrake  Scotland
diyAudio Member
 
DBMandrake's Avatar
 
Join Date: Nov 2010
Location: Glasgow, UK
All fixed, some 16 capacitors later.

While I identified two caps that were specifically relating to the right channel fault during my initial testing it turns out after measuring them all during removal that of the 16 47uF caps on that board 4 were completely open circuit so I'm not sure how it was even working, although some were in other functions like the RTA input which I haven't used in over a decade...

C19 which made the most difference to the right channel level was down from 47uF to just 0.2uF C18 was down to 10uF.

The values were all over the place at random - some just a few uF or completely dead, some nearly at their full value. I suspect the ones that were still relatively good were ones that have bias voltage although I didn't actually check that to confirm it. There were two different batches of capacitors fitted from the factory - one lot blue, one lot black.

I replaced them with good quality Panasonic types so hopefully they will last longer, and will probably see out the rest of the useful life of the device considering it's now 17 years old.

Here are the before and after measurements. Note that the vertical scale is blown up a lot to measure the differences accurately making the high frequency roll off appear worse than it really is.

Before:
Another (ultimate?) Behringer mod - what can  do to DEQ2496  (PART 1)-deq2496-repair-png

The upwards slope on the right channel is really obvious, at 30Hz the difference is 3.3dB, at 10Khz it is about 1.8dB.

After:
Another (ultimate?) Behringer mod - what can  do to DEQ2496  (PART 1)-deq2496-repair-png

Channel matching is near on perfect now with the two channels within 0.06dB up to a few Khz and about 0.1dB at 20Khz.

Also of note is that even the "good" left channel has increased in level about 0.3dB and has significantly less bass roll off, so lower capacitance values were still affecting the left channel somewhat.

Although I did order a bag of 0.1uF polyester film caps to bypass the electrolytics, in the end after looking at the measurements and doing some reactance calculations I opted not to install them.

So I'm very happy with this fix - at the age these are getting to I suggest people measure theirs as well because it's likely some of the zero bias coupling caps in the input/output circuits have dried out and are compromising the performance. I was shocked at how many had outright failed let alone dropped to extremely low values.
Attached Images
File Type: png DEQ2496 Before repair.png (76.0 KB, 71 views)
File Type: png DEQ2496 After repair.png (74.4 KB, 69 views)
__________________
- Simon

Last edited by DBMandrake; 10th September 2021 at 08:53 AM.
  Reply With Quote
Old 10th September 2021, 08:52 AM   #49
TasoTaso is offline TasoTaso
diyAudio Member
 
Join Date: Oct 2018
Hi, you did a great job! Which capacitors did you fit exactly? I mean part number / make / model / specifications. Thanks
  Reply With Quote
Old 10th September 2021, 08:58 AM   #50
bentoronto is offline bentoronto  Canada
diyAudio Member
 
Join Date: Jan 2008
Location: Toronto and Delray Beach, FL
Another (ultimate?) Behringer mod - what can  do to DEQ2496  (PART 1)
Thanks for very nice presentation of data and good write-up.

How old is your device?

B.
__________________
HiFi aspirations since 1957. "When the toilet paper of Experience is depleted, the backside of Reason goes unwiped"
  Reply With Quote

Reply


Another (ultimate?) Behringer mod - what can  do to DEQ2496  (PART 1)Hide this!Advertise here!
Thread Tools

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

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Behringer DEQ2496 modification tojoko Digital Line Level 3 26th February 2013 03:30 AM
Help with Behringer DEQ2496 pierre77 Digital Line Level 10 28th November 2012 04:42 PM
FS:Behringer DEQ2496 Ryssen Swap Meet 15 16th April 2009 08:24 PM
Behringer DEQ2496//Non-Delay Mod**HELP** syntaxsinner Digital Line Level 5 26th April 2006 03:27 PM
Behringer DEQ2496 into T Amp? joesan Digital Source 7 19th October 2005 10:30 AM


New To Site? Need Help?

All times are GMT. The time now is 01:33 AM.


Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2021 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 14.29%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2021 DragonByte Technologies Ltd.
Copyright ©1999-2021 diyAudio
Wiki