Newbie builds GroupDIY 51X rack

[jsteiger]

According to the Rane notes, the cable's shield is "an extension of the chassis". http://www.rane.com/note151.html

 

[nielsk]

Remember, those Rane notes must be taken with a grain of salt... more of a "tips for beginners" guide than a bible.

 

[dmp]

I've found that Rane article to be very comprehensive and the best article on grounding scenarios that I've seen.
The section "The Absolutely Best Right Way to Do It" shows pin one going directly to chassis and functioning as an extension of the enclosure, which is the pin1 to chassis option on the 51x rack.
I have never found a situation where where better performance happens with pin1 connected to audio ground instead of chassis, but the 51x pcb allows that if necessary for a particular piece of equipment.
If you find buzz or hum, you can experiment and see if it is reduced by changing.
In all the equipment I have built, I have found the "The Absolutely Best Right Way to Do It" is dead quiet and the best. Lifting pin1 on either the input or output increases buzz/hum slightly, and putting a shield through audio ground, which some builds do by design, has created issues that I've gone back later to correct.

 

[nielsk]

So this works for you with phantom powered mics? Using the equipment shield path for the return for phantom does NOT seem like a good idea.
It would appear the the Rane notes are aimed at PA applications (their main market, after all).
In addition to showing examples of tying shields at inputs and lifting at outputs (not a good idea) they do not take patchbays into account.
If one were using 3 conductor cable and only connecting cable shields to the connector shell (not to pin 1) the idea of only connecting cable shields to equipment shields would work. Of course, this is not how the real world functions...

 

[john12ax7]

So this works for you with phantom powered mics?

I was wondering about that. How is pin1 normally connected on mic pre inputs?

 

[chunger]

Not an expert by any stretch of the imagination, but the 2 racks here have pin 1 connected to chassis as suggested "standard".  Everything works in critical use all across the rack (phantom power, no noise problems, etc.).  If problems come up, I have options and can re-investigate.

 

[jsteiger]

I was wondering about that. How is pin1 normally connected on mic pre inputs?

Jensen shows the pin 1 shield connected to chassis on their mic preamp schematics.

 

[[silent:arts]]

Remember, those Rane notes must be taken with a grain of salt... more of a "tips for beginners" guide than a bible.

couldn't disagree more.
you could buy the german broadcast IRT Richtlinien bible for money, but with lesser information / explanation inside

 

[dmp]

So this works for you with phantom powered mics? Using the equipment shield path for the return for phantom does NOT seem like a good idea.

No, it doesn't seem like a good idea - but the standard for phantom power is "grandfathered in" more than an elegantly thought out methodology.
Pin 1 for balanced signals is defined to be the shield and most if not all XLR cables connect the shielding in the cable to pin 1 (and do not have a third wire for pin1). If you wanted to make pin 1 a phantom ground (separate from the cable shield and mic body) you would probably have to rewire your mics so the mic body is not connected to pin1 AND create custom cables.
So it isn't something that you can just change at the preamp input - you effectively have to change the phantom methodology for all your mics and cables.
If you don't change your mics and cables, but connect pin1 to audio ground at the preamp input, you are sending any RFI that was picked up by the mic body and cable shield into the audio ground of your preamp. This is not a good!
Furthermore, I would think you are MORE likely to introduce voltage noise back to the phantom mic because you are reducing the reservoir size of the ground that the phantom current is returning to. What does this mean? The ground plane in a preamp has current flowing through it. The narrower the traces, the larger the tiny voltage changes across the ground traces. These little voltage changes can show up as noise if a sensitive component is grounded upstream of them (like a phantom powered mic). But connect the phantom ground to a big steel case? Current flowing through the big steel case make MUCH smaller voltage wiggles.

 

[Harpo]

Having a look at the jensen-transformers AS016 IMHO shows it quite clear when looking closer to the shielding symbols. There is no need to connect the shield to the preamps for usual 0V reference voltage. The mic preamp exclusively operates the potential difference applied to XLR-pins 2/3. Phantom power return currents (worst case 14mA per mic from a shorted cable) thru XLR-pin 1 close tied to the conductive case keeps picked up induced garbage inside this 48VDC supply loop and away from audio ground.
Just MHO.

 

[nielsk]

Of course, using the case as the return path for phantom power relies on an excellent connection from the XLR case connection to pin 1 and the power supply ground at some point, this is a path subject to poor connection over time (screws loosening, corrosion, and so on. This is why I never use the case as a signal patch, only as a shield).
It's unfortunate that the standard is not 3 conductor shielded cable (as in the classic Gotham cable) as a separate ground conductor & shield would work much better.
The kind of situation where the"Rane" notes  fall short is when you have a 120 input console, several different 24 & 48 track (& beyond) machines, 100's of pieces of outboard gear, multiple rooms interconnected, and so on.
"Audio Systems Design and Installation" by Philip Giddings is a far better resource for this kind of information, but of course it is not reduced to one sheet of paper.
BTW, in an SSL console, the "pin 1" connection to patch for mic inputs goes to the AUDIO ground buss bar... I have yet to see where this approach causes problems.
Has anyone actually experienced a problem caused by connecting XLR pin 1 to the audio P/S star ground node instead of the the case?

 

[[silent:arts]]

the "pin 1" discussions will never find a perfect solution for all.
this is why the GDIY511 Rack has options for pin 1 8)

 

[jsteiger]

the "pin 1" discussions will never find a perfect solution for all.
this is why the GDIY511 Rack has options for pin 1 8)

+1

 

[sisepuede]

Not an expert by any stretch of the imagination, but the 2 racks here have pin 1 connected to chassis as suggested "standard".  Everything works in critical use all across the rack (phantom power, no noise problems, etc.).  If problems come up, I have options and can re-investigate.

I can see in this pic http://studio939.zenfolio.com/img/v22/p1018874573-4.jpg you connected chassis and ground together, if you¿ve sent by 7pin Neutrik connector "that" chassis to your 511 rack probably the whole 511 rack have chassis and ground as the same... in that case doesn't matter if you are using for pin 1 chassis or ground...  Am I wrong?

Cheers.

 

[chunger]

As I understand it, the chassis and ground are supposed to connect at the star ground and only at the star ground in the PSU.  This is how I wired it.

Same with my house. .  . isolated grounds, neutral, and "regular" grounds all connect at 1 point near the grounding stakes. . . probably the same concept.

 

[jsteiger]

As I understand it, the chassis and ground are supposed to connect at the star ground and only at the star ground in the PSU.  This is how I wired it.

Correct 'Chunger. From the rack side, they are fully separate ground return paths to the star ground point.

 

[sisepuede]

As I understand it, the chassis and ground are supposed to connect at the star ground and only at the star ground in the PSU.  This is how I wired it.

Correct 'Chunger. From the rack side, they are fully separate ground return paths to the star ground point.

Sorry guys I am just trying to understand, my PSU is almost done but I am a bit lost here (Ground - Chassis thing).

What I can see here http://studio939.zenfolio.com/img/v19/p670575032-4.jpg (Thanks a lot Chunger for such amazing documentation!) if I am right,  is the chassis of the neutrik connector and XLR's PIN 1 both together, then seems like in this picture http://studio939.zenfolio.com/img/v6/p1016740960-4.jpg chassis from the XLR pannel is going to the PCB chassis block connector so when you plug that XLR in the 511 Rack you'll be passing the ground from the PSU to the chassis of the 511 Rack as far as I can see. Am I wrong?

Shortly: Chassis and Ground in the PSU are the same, but because XLR chassis in all XLR are connected and linked to PSU ground then 511 Rack Chassis is connecetd to the PSU Ground. If I am skipping anything please let me know.

Thanks a lot guys!!

Cheers.

 

[chunger]

Pin 1 is "chassis" on the 7 pin neutrik power connector. . . because I'm using a shielded 8 conductor cable (not these same as in the kit), I decided to sortof make the chassis enclose the entire cable run including the cable end housings. . . that's why pin 1 (wire shield) is connected to the connector body fitting on my cable.

My thinking was it wouldn't hurt to have the chassis shield as much of the system as possible including the cable run, but it's almost guaranteed to be completely unnecessary.  I was just excited to find my marine grade cable at reasonable price that I had to use it  . . . You know. . . just in case you want to use your rack on a boat in the middle of the Pacific.

 

[sisepuede]

Hi guys,

don't know if anyone have had this problem but seems like my phantom power is noisy, like 50hz everytime I turn the phantom power I can hear that horrible noise, I've tried many things, I've removed C9 (0.1uf or 100nf if you like) and then it worked! the noise was gone but after one day recording and working perfectly well today it started again... so I opened up the PSU and the phantom power voltage is 48.1 but it is going up... 48.2, then 48.3, then 48.2, then 48.1 and so... again and again the same all the time and you can't touch diode D2 (1N4002) or resistor R2 (150R) because those are hot like hell... Could be the TL783 Regulator? Any idea? Thanks a lot.

 

[sisepuede]

Good stuff Chunger. Great build.

Third run is in production and planned to be available within four to five weeks.

Hi Cemal,

Here you have some pics, as you can see diodes are right. I couldn't find a TL783C, I'll try on CPC or RS, but definitely I'll remove D2 and C8 to what happens without those.

Cheers.

Diodes:

The whole PSU: