Crosstalk issue with homebrew headphone amps - diyAudio
Go Back   Home > Forums > Amplifiers > Headphone Systems

Headphone Systems Everything to do with Headphones

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 1st July 2014, 07:08 PM   #1
diyAudio Member
 
Join Date: Jan 2014
Default Crosstalk issue with homebrew headphone amps

Not sure this is the ideal location for this post, but here goes:

I work at a trade school that teaches Recording Arts and has a number of studios. The original tech here designed a headphone amp/ cue system which has been working pretty well over the last 7-8 years, with one ongoing issue: there is significant crosstalk between two of the studios' cues. I.e. you can hear fairly loudly the cue send from one room in the cues in another room, and vice versa).

Since our staff noticed this, it's been chalked up to EM-induced signal between the cable runs (all the headphone PSUs and AD/DA converters live in a single machine room that run to wall panels in the studios to connect to the amp boxes), and we aren't allowed to change anything with these runs (union issues), but it's a frustrating issue so I'm wondering if it can be dealt with somehow.

A bit more info: there is +-15V (if I remember) from the power supplies to the wall panels, along with 4 balanced channels (2 cues, 2 channels each) along with a conductor each labeled for "audio shield" and "power ground." The cables between the wall panels are 12-conductor Belkin cables with Minicon connectors-- the shield is not connected on these cables, which I thought may be part of the problem, but these cables are short and it's been assumed this is an induced signal between the machine room and studios.

Any thoughts on what I could try, or even if we're on target in thinking this is EM induction?
  Reply With Quote
Old 1st July 2014, 10:31 PM   #2
jcx is offline jcx  United States
diyAudio Member
 
Join Date: Feb 2003
Location: ..
1st would be seeing if it is in the cable or the amps

have you plugged the headphones directly into the amplifier outputs with the cable disconnected?
  Reply With Quote
Old 2nd July 2014, 02:02 AM   #3
diyAudio Member
 
Join Date: Jan 2014
Quote:
Originally Posted by jcx View Post
1st would be seeing if it is in the cable or the amps

have you plugged the headphones directly into the amplifier outputs with the cable disconnected?
The cable carries power as well as audio, so when unplugged there's no power. And no crosstalk to headphones.
  Reply With Quote
Old 2nd July 2014, 03:20 AM   #4
jcx is offline jcx  United States
diyAudio Member
 
Join Date: Feb 2003
Location: ..
its extremely hard to debug or fix something if you can't or won't change anything - are you by any chance in management?

Last edited by jcx; 2nd July 2014 at 03:33 AM.
  Reply With Quote
Old 2nd July 2014, 10:22 AM   #5
diyAudio Member
 
Join Date: Nov 2004
Location: Germany
Oh great, custom hardware. Is there a block diagram or similar documentation available?

Let me break this down:
On each 12-pin connector, you have 3 pins for +/- DC supplies and power ground, 1 pin for audio shield, and the remaining 8 for 4 balanced channels. Am I getting this right that sort of a ribbon cable is being used? Or is the audio being run over twisted pair (which would be better)? These 2 audio channel pairs both go to the headphone amps? Or theheadphone amps in the same room?

If the crosstalk were inductive or capacitive in nature, it ought to be quite treble-heavy with no bass. Is this actually the case? Is it mono xtalk, and are both channels affected equally?

The root cause may not actually be in these cable runs but elsewhere, especially if the xtalk sounds "normal" or if the two studios in question do not share the same wall panel.

If it sounds normal and they do share a wall panel, I would look very hard at the power ground connection, which may be interrupted somewhere, rendering amplifier audio ground "floating" and relatively high impedance. In this case, left/right xtalk ought to be pretty lousy as well (though with headphones you might not notice that as much). The fault would not necessarily have to be at the cable (though I'd definitely check that first), it could also be hiding within the headphone amps or power supply. In a straightforward implementation, I would expect continuity between power ground and headphone output ground - check with a known-good amp first.

Last edited by sgrossklass; 2nd July 2014 at 10:24 AM.
  Reply With Quote
Old 2nd July 2014, 01:48 PM   #6
diyAudio Member
 
Join Date: Jan 2014
Quote:
Originally Posted by jcx View Post
its extremely hard to debug or fix something if you can't or won't change anything - are you by any chance in management?
No. I'm fine changing something, but we have 14 of these boxes and about the same number of the 12-conductor cables, and rebuilding each cable takes about 2 hours. The boxes are built on perfboards and the 24 wires (2 inputs on each box) running into the boards and the 6 out (2 outputs on each) would mean a serious investment of time in getting to a point where I could even do board-level work. I'm merely thinking in terms of cost-benefit: if I have to invest more than an hour or so on each box, it's probably not worth it since it's an annoyance but doesn't actually stop us from working.

Thanks for your snarky reply.
  Reply With Quote
Old 2nd July 2014, 02:11 PM   #7
diyAudio Member
 
Join Date: Jan 2014
Default R3c4rtsAdmin28

Quote:
Originally Posted by sgrossklass View Post
Oh great, custom hardware. Is there a block diagram or similar documentation available?
Yes, I'll upload the schematic later this morning.
Quote:
Originally Posted by sgrossklass View Post
Let me break this down:
On each 12-pin connector, you have 3 pins for +/- DC supplies and power ground, 1 pin for audio shield, and the remaining 8 for 4 balanced channels. Am I getting this right that sort of a ribbon cable is being used? Or is the audio being run over twisted pair (which would be better)? These 2 audio channel pairs both go to the headphone amps? Or theheadphone amps in the same room?
Right. Not a ribbon cable that goes between the wall panel and the box. It's Belden 14-conductor wire (sorry, I think I said 12 earlier) 9808. Twisted paid would be nice, but selection of 12-14 conductor wire is minimal. And yes, both audio channel pairs go to the amps, there are 2 cues on each box, which you can use for two performers on independent stereo cues, or folded together so that you can mix, say, bass, drums, and guitar on one stereo cue, and vocals and guitar on another.

Quote:
Originally Posted by sgrossklass View Post
If the crosstalk were inductive or capacitive in nature, it ought to be quite treble-heavy with no bass. Is this actually the case? Is it mono xtalk, and are both channels affected equally?
As I was writing my initial post, I realized this would be an important test. I'm pretty sure that yes, it is treble heavy, and yes, it is mono/ neither channel has any more or less of it. But I'm going to double check that today.

Quote:
Originally Posted by sgrossklass View Post
The root cause may not actually be in these cable runs but elsewhere, especially if the xtalk sounds "normal" or if the two studios in question do not share the same wall panel.
They do not share the same wall panel. BUT the power supplies in the machine room share a power strip, and the audio is distributed to the rooms off terminal blocks, and these two share the same rail. I previously figured one of these maybe the crux of the problem.

What do you mean "normal"-- do you mean the actual audio that is going to the cue?
Quote:
Originally Posted by sgrossklass View Post
If it sounds normal and they do share a wall panel, I would look very hard at the power ground connection, which may be interrupted somewhere, rendering amplifier audio ground "floating" and relatively high impedance. In this case, left/right xtalk ought to be pretty lousy as well (though with headphones you might not notice that as much). The fault would not necessarily have to be at the cable (though I'd definitely check that first), it could also be hiding within the headphone amps or power supply. In a straightforward implementation, I would expect continuity between power ground and headphone output ground - check with a known-good amp first.
  Reply With Quote
Old 8th July 2014, 07:41 PM   #8
diyAudio Member
 
Join Date: Jan 2014
Ok, I investigated further. There was 1 bad cable on which the crosstalk with the other studio was really bad. Otherwise, the crosstalk was pretty manageable at all but the highest levels everywhere (mixer to headphone amp in room 1, headphone amp volume in room 2). So I'm wondering if we just had a number of bad cables in circulation.

All the boxes do have continuity between audio ground and power ground. One very strange thing I noticed-- in some boxes and channels, the crosstalk is left-heavy, in some it's right-heavy, in some it's loud, in some it's barely audible. It is not treble-heavy in any combination I found.

So the damage is contained, but I'm still perplexed. I would like to completely eliminate this at some point, if at all possible. Any thoughts on where to look next, or what my clues point to?
  Reply With Quote
Old 9th July 2014, 09:10 AM   #9
diyAudio Member
 
Join Date: Jan 2012
Do the headphone amps in the different studios share a common DC power supply .
Do the headphone amps have balanced inputs .

There are so many possibilities......please post a schematic to at least get us somewhere close to fully understanding the layout .
  Reply With Quote
Old 9th July 2014, 02:01 PM   #10
diyAudio Member
 
Join Date: Jan 2014
They do not share a common power supply-- each room has its own independent supply that is split via terminal blocks going to each of that room's panel inputs. I tried removing the blocks from their shared rail-- it didn't help. I tried moving one amp to another outlet on a different circuit-- it didn't help either.

The headphone amps are balanced input, yes.

Here's a schematic, which doesn't include the input section. The input section is just each Hi/ Low input run through a 10k resistor with a 330pF capacitor between each Hi-Low pair into a single TL074CN for all 4 balanced channels.
Attached Files
File Type: pdf headphone amp schematic.pdf (88.1 KB, 30 views)
  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
Dual TPA6120A2 headphone design to eliminate crosstalk diyinhk Headphone Systems 14 15th July 2014 06:29 PM
MiniDSP for headphone EQ & crosstalk? Soldermizer Headphone Systems 8 25th August 2012 03:51 AM
Help to reduce crosstalk in headphone amp(also rate the design) :) GTrenchev Headphone Systems 15 28th June 2011 04:38 PM
NAD 302 - Headphone jack issue Verve111@gmail. Headphone Systems 0 22nd May 2009 10:33 PM
Headphone cable crosstalk - how to avoid? halcyon Headphone Systems 11 15th March 2003 06:02 PM


New To Site? Need Help?

All times are GMT. The time now is 05:22 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