Topic: Sound Devices 744t inverted phase issues

[baustin]

I haven't been able to find a thread specifically devoted to this topic. Has anybody gotten to the bottom of this?

This is the only thread I've seen hitting on this subject ( http://taperssection.com/index.php?topic=49528.0 ) but I haven't found a suitable answer. Maybe all I need is a physics refresher course.

Anyway, here's where I stand...

I just made my first 4 channel recordings (2 sets of AUD mics) with my 744t, firmware 1.60. 4 mono files.

Source 01 - Neumann u89i>EAA-PSP2>744t (analog into channels 1+2)
Source 02 - Schoeps mk41>KC-5>CMC6>V3>digi-in>Tascam HD-P2>744t (digital into channels 3+4)

Now I realize this is a bit of a different situation from the above thread started by wboswell (2 very different brands of mics and the Tascam), but the sources synch up identically, except that they are out of phase.

This raises a few questions...

-Which source is out of phase? Theoretically, the Tascam and 744t should only be recording a digital source. If this is correct, does this mean that every 2 channel recording (into channels 1+2) made by a 744t is out of phase?

-Could the V3 have inverted phasing issues like the AD2K+?

-Could synching the Word Clock of the 744t to the Word Clock of the V3 help?


Maybe some of the brillant minds here can explain the whole multitrack phasing thing....


-Bryan

[jeromejello]

-Could synching the Word Clock of the 744t to the Word Clock of the V3 help?


Maybe some of the brillant minds here can explain the whole multitrack phasing thing....


-Bryan

i dont know anything, but if there is a word clock out on the v3 that would be the reference needed for the 744t to sync up i would think.

[chun]

If you use any external digital in - then the 744 will sync to THAT word clock.....

[fsulloway]

Brian- Check out the pic that Damon posted of the wave files in that same thread. I could be wrong but didn't one of the updates for the 744 fix the phasing issue?

[baustin]

Brian- Check out the pic that Damon posted of the wave files in that same thread. I could be wrong but didn't one of the updates for the 744 fix the phasing issue?

I believe that the 2nd pic he posted were files he inverted himself. At least that's what I got.

Also, I just got the 744t this past Tuesday (3/14) so it has the latest firmware.

-B

PS- I'm in NYC right now and I PROMISE I'll send those Tishamingo DAT's as soon as I get home! 

[Dr.FOB]

I have noticed this problem when running V3>digi>744 also.
I queried some other v3>744 users to check into this.

I'm going to assume that the phase of the mic>744 is correct and that the V3 is providing an inverted digi output.
That's how I've been mixing my matrix renderings to date.

[baustin]

If you use any external digital in - then the 744 will sync to THAT word clock.....

Chun,

Is there a difference between synching to the sample/bit rate and synching to the word clock? I honestly don't know.

I was just thinking about this because I remember reading about people synching V3's with broken AD1000's. I think I read that they were able to get AD1000's with dead crystals to work like new. Completely different animal, but it just had me curious.

-B

[baustin]

I have noticed this problem when running V3>digi>744 also.
I queried some other v3>744 users to check into this.

I'm going to assume that the phase of the mic>744 is correct and that the V3 is providing an inverted digi output.
That's how I've been mixing my matrix renderings to date.

Alright....

So this is still an issue and we need to get to the bottom of it.

Chun, I see you have multiple pairs of mics as well as a MiniMe. Maybe you could do some testing when you've got a second. It'd be interesting to see what results an A/D other than a V3 produces.

-B

[Dr.FOB]

Chun, I see you have multiple pairs of mics as well as a MiniMe. Maybe you could do some testing when you've got a second. It'd be interesting to see what results an A/D other than a V3 produces.

John and I both ran multi mic rigs in Philly.  He ran mic direct and mini-me digi in and can check those for us.
I ran V3>line and V3>digi (2 V3's) so that comp won't help.

I will try later a V3>line and V3>digi from the same V3 and see if the digi is inverted there with respect to it's own analog output...

[cleantone]

couple questions... (never used a 744 so bare with me)

Have you imported to a DAW to notice the phase issue?

If so do each of the stereo recordings (you had 2 stereo pairs right?) sounds okay on thier own? Meaning is it only when you combine them that you hear a phase issue.

I think I'm right about this:

suppose you look at each track on its own. If you find something with a lot of low end (big waveforms) you should see the beginning of the waveform go positive before negitive (above the 0 line not below)

This should tell you if any of them are reversed polarity. I assume you mean reversed polarity when your talking about phase here. This is an easy fix.

I have more once you answer this...

EDIT:

I just read some of the thread that was linked here. Heath goes into what I was thinking. I assume it is the same problem.

[baustin]

I will try later a V3>line and V3>digi from the same V3 and see if the digi is inverted there with respect to it's own analog output...

This would be perfect.

Regardless, has anybody tried this just to do a V3 A/D vs. 744t A/D comparison. Actually, that's a stupid question. I'm sure it has.

Anyway, I'm sure V3>744t analog and digital should shed some light.

-B

[cleantone]

Assuming this is the problem (polarity reversed) to tell which one to "fix" look at the compression and rarefaction of a low frequency note. Low because the waveform is bigger. Doesn't have to be low of course. Anyway you in the proper one you should see compression (+) before rarefaction (-). Look at this picture.



I find a lot of the time the highest amplitude peak on a single seciton (snare hit, single note etc...) is compression. That might help you as well in identifying the proper track to fix.

As Heath said in the thread linked in the first post when the kick drum is hit your speaker should go outward and not inward. This can be a little hard to distinguish in some cases though. That is another way of making sure you fix the proper channels.

[baustin]

As Heath said in the thread linked in the first post when the kick drum is hit your speaker should go outward and not inward. This can be a little hard to distinguish in some cases though. That is another way of making sure you fix the proper channels.

Fixing the phase in Wavelab is possible, but is a pain in the ass. Plus I don't want to have paid a couple thousand bucks for a product that kinda works.

IMO, if we can identify a problem, this is a fix for Grace Design or Sound Devices. Thankfully both of these companies have great support and will hopefully address any issues.

[cleantone]

Oh okay. I was more aiming you to saving your recording you had already made. Also info for anyone else who might want to know how to identify the problem track. That other thread didn't fully get that across either.

I agree that it should not happen at all. I would be a really bent myself. You pay out your ass to be able to record 4 channels in the field for a reason right? I am wondering how the hell they can make a 4 channel field recorder that manages to reverse polarity and cost thousands. It seems like this should have been caught by testers or something. Hopefully a firmware upgrade can and will solve this for you.

[wboswell]

When I originally brought it up, here and on the SD forum, no one could explain it.  I'm pretty confident that the kc5's are properly wired and I double checked the interconnects and they were/are properly wired.  That leaves either the V3 or the 744t. 

C. Fox called me one day and told me that he had been on the SD forum and someone was able to figure out the problem, but honestly, I can't remember much of what he said.  I would suggest reviewing that board for some help.