darkcatalyst
The export I posted is a good example because it's a pretty straight-forward project as of now.
 
Track 1: an audio "guide" which is a Guitar Pro wav export, imported into SONAR.  Muted at mixdown time.
Track 2: a click track I created in Dimension Pro using the Piano Roll View and creating a groove clip
Tracks 3,4,5: Dimension pro string/brass/orchestral tracks.  I generally compose in Guitar Pro, export the midis from GP, import midis into SONAR for use with Dimension Pro
 
So no mic involvement, no actual "recording", just wav/midi manipulation.  I don't believe Dimension Pro or midi is the culprit.  As an even simpler example, the TestProj.zip I posted earlier is a single wav imported into SONAR and panned - I can export that and hear the same bleed.

But here's the crux of the issue. I downloaded your test project. I panned the waveform full left and bounced to track. There was zero signal in the right channel. I then exported as a WAV file, brought the waveform into SONAR, and there was still zero bleed in the right channel. I even did this several times with different dithering types because if the dithering math isn't totally random, that can compromise stereo separation. In all tests I normalized the right channel to bring up anything that might be there, however faint. All it did was bring up the dither noise. There was no click present at all.
 
With your test project I am not able to duplicate your results within SONAR. Therefore the problem does not lie with SONAR. It relates to something unique to your project or setup that it not present in my setup.
 
Regardless of whether you explicitly work with audio or miking, the whole point about leakage is that it is an undesired, unplanned migration of audio from one point to another. There are many ways this can happen, and the fact is you have an audio interface so there are audio inputs and outputs regardless of how you have integrated those into a project. You will not be able to solve this issue by looking within SONAR's audio engine. That is not where the problem lies, or I would have been able to duplicate your results with your project.

But wouldn't you figure that all project settings would be attached to the project itself?  Since the interface is a non-factor here, and it's not the SONAR audio engine, what else could it possibly be?  Are there any global settings unrelated to a project that could be having an effect here?

darkcatalyst
But wouldn't you figure that all project settings would be attached to the project itself?

 
The project consists of only the settings that are within Sonar. This is why your results cannot be reproduced here, because when I load your test project, it reproduces only what Sonar is doing. Sonar is indeed panning 100%, so the problem must be something outside of Sonar. Your test project eliminates Sonar as a possible source of the issue.
 

Since the interface is a non-factor here, and it's not the SONAR audio engine, what else could it possibly be?  Are there any global settings unrelated to a project that could be having an effect here?

We don't know for sure that the interface is a non-factor. It could be coincidence that when the interface is not part of the system, the results cannot be reproduced and when the interface is a part of the system, the results can be reproduced. But I can't think of any global setting that would make a difference.
 
Clearly the interface is attached to the computer via USB or you would not be able to monitor the audio. FWIW a USB bus is not always 100% "clean" from an audio standpoint. The best example of this is when adding a USB hard drive introduces faint clicks into the audio interface due to power supply fluctuations. Remember that all the USB ports on your computer often connect to only a single controller. I eliminated these types of problems completely by purchasing a USB PCI card that's dedicated solely to audio-related USB accessories. The computer's USB ports go to mouse, keyboard, hard drive, dongles etc. where noise is not an issue.
 
Interestingly there's another thread on the forum where someone talks about not being able to pan 100% and he mentions using an RME Babyface. I tend to think it's not the same issue you're experiencing but maybe a misunderstanding of how the stereo/mono options work in Sonar, but still, it was interesting to have two threads show up with problems not before seen in the forums, or reproducible by other users, that mention using the Babyface. Are you sure there's not a stealth setting somewhere in Totalmix that's causing issues? Can you borrow a USB PCI card to see if that makes a difference?

Anderton
The project consists of only the settings that are within Sonar. This is why your results cannot be reproduced here, because when I load your test project, it reproduces only what Sonar is doing. Sonar is indeed panning 100%, so the problem must be something outside of Sonar. Your test project eliminates Sonar as a possible source of the issue.
 

Are you implying that there are project settings that are beyond the scope of SONAR?  Not sure exactly what you're saying here, but that would seem strange to me, being that the project file is a proprietary SONAR format.
 

Anderton
We don't know for sure that the interface is a non-factor.

 
I think we can conclude that it's a non-factor when working within the mixdown scope only.  All of this behavior can be reproduced without introducing any interface into the equation.  As in the test project, a wav was imported, panned, and exported - no interface involvement, yet the mixdown still has the bleed issue.  I think this is the best way to approach the problem since it eliminates a lot of variables.
 
Yet, since it seems no one else is able to reproduce my results, the logical conclusion is that it has something to do with my environment.  I don't think there's anything particular special about my Win7 Ultimate 64bit setup, nor can I think of anywhere else to look - grasping at straws at this point.
 

darkcatalyst

Anderton
The project consists of only the settings that are within Sonar. This is why your results cannot be reproduced here, because when I load your test project, it reproduces only what Sonar is doing. Sonar is indeed panning 100%, so the problem must be something outside of Sonar. Your test project eliminates Sonar as a possible source of the issue.
 

Are you implying that there are project settings that are beyond the scope of SONAR?  Not sure exactly what you're saying here.

 
I'm saying there are a lot of elements outside Sonar, like the interface, USB bus, cabling in the studio, etc.
 

Anderton
We don't know for sure that the interface is a non-factor.

 

darkcatalyst
I think we can conclude that it's a non-factor when working within the mixdown scope only.  All of this behavior can be reproduced without introducing any interface into the equation.

 
If the interface is connected to USB, and there are audio cables involved at all, it's a factor. It just occurred to me (I'm a little slow sometimes!) that you could have a ground loop, and therefore, audio is traveling along the ground lines. You could have all the controls on your interface turned down to nothing and the inputs shorted out, but if audio is on the ground lines, it's going to be VERY difficult to keep out of your project as long as there are any audio connections involved on any level. Presumably there are, because you're monitoring the audio.
 
I think a ground loop is a very likely candidate for the source of the problem. The character of the "hash" noise that's in with the click leak in your MP3 file is characteristic of the type of garbage audio that appears on ground lines through ground loops.

Hi ,

Did you find a fix ?

I have a Babyface and am experiencing the same low audible sound when I hard pan either way ?

I never found a good answer to this in SONAR.  However it was one of the many issues that eventually drove me to switch to Pro Tools, and I haven't looked back.

I never found a good answer to this in SONAR.  However it was one of the many issues that eventually drove me to switch to Pro Tools, and I haven't looked back.
 
Hi thanks for the reply , but its not sonar , i have opened a new Platinum  project with 1 track > then using win audio-Asio4all - laptop headphone  jack > hard panned the track and it is completely silent in the opposite ear !As it should be !
 
However as soon as i connect the snow edition babyface and route the same track to total mix fx and hard pan the same track there it is in the opposite ear ..Also if i open a video in VLC player and then go to total mix fx and hard pan in total mix fx ,the problem is the same again ,i can still hear the movie in the opposite ear ! so its either the babyface or total mix fx ??? ( Also i just want to note that when this bleed or whatever its called  occurs - in total mix fx there is no meters showing it happen !NOTHING)
 
Glad to hear you like pro tools , i started with fl studio then went to ableton then to sonar and have been diving deep now for 1.5 years and still going , my plan was to go to pro tools last but the recent updates have solidly grounded me at cakewalk now forever , there are some things i would like to see added to platinum and at the rate the rolling updates occur i am sure they will fly by soon enough , thanks for the reply i just hope mattias over at rme forum will help because i am thinking time for a new audio unit , however i do want to know if its just some units or all babyfaces that this happens to .
 

Interesting, we must be experiencing different issues with the same outcome.  I tried multiple outputs from SONAR including the Babyface, FireStudio Project, and Alesis MultiMix and always experienced the same issue.  It disappeared with Pro Tools.  Hope you're able to find an answer.

darkcatalyst
Interesting, we must be experiencing different issues with the same outcome.  I tried multiple outputs from SONAR including the Babyface, FireStudio Project, and Alesis MultiMix and always experienced the same issue.  It disappeared with Pro Tools.  Hope you're able to find an answer.

 
This piqued my curiosity so I wanted to find out if SONAR indeed had inter-channel leakage. It does not. Here's the test so others can prove to their own satisfaction there is no "100% panning issue."
 

1. Place a mono signal on an audio track, panned full left.
2. Send the track to the master bus.
3. Insert 10 Channel Tools in the master bus, applying 0 dB gain to the Left channel and 6 dB of gain to the Right channel, for a total of 60 dB of gain applied to the right channel.
4. Set the Master bus meter set to the -90 dB range.

 
Refer to the screen shot. On playback, there is full signal in the left channel and nothing in the right channel. Note that with the -90 dB meter setting and 60 dB of gain, any signal in the right channel higher than -150 dB would have registered. It didn't. There is nothing in the right channel if a mono signal is panned all the way to the left channel.
 
As to why things worked with Pro Tools, there are a number of ways this issue could happen in SONAR. For example bouncing to a stereo track with dithering on would leave a dithered signal in the right channel. I tested this. With an identical setup but with dither applied on a bounce, the level in the right channel was -54 dB. If Pro Tools doesn't dither on bounce, this won't be a problem (note that in SONAR preferences, you can turn off dithering).There are several other possible scenarios, but they all involve a misset parameter. If an equivalent parameter is not misset in Pro Tools, obviously the same problem will not happen. 
 
Bottom line: SONAR itself does not have a 100% panning issue.