You are here

Beware 'Pin One' trap

All about the tools and techniques involved in capturing sound, in the studio or on location.

Beware 'Pin One' trap

Postby ef37a » Sun Dec 22, 2019 3:52 pm

Several years ago I bought a red and blue set of XLR 3 5mtr cable from Maplin. They went up and over and down from a ZED 10 to a 10k traff box into a 2496. Nay bother.
Yesterday I was in urgent need of two XLR 3 female plugs (seem to have sheds of males?) to make up two shortish XLR to TRS cables and as the Zed rig is no longer setup and used, I chopped the plugs off.

When I came to clean them up I found pin one was tied to the shell. Actually, the braid was soldered to the rather crude cable clamp and then to pin one. The single fixing screw, a la Cannon, then ties in the shell.

Moral is, if you have any made up cables do not assume shell are not connected to pin one, especially if they are of the budget variety.

Dave.
ef37a
Jedi Poster
Posts: 11167
Joined: Mon May 29, 2006 12:00 am
Location: northampton uk

Re: Beware 'Pin One' trap

Postby Folderol » Sun Dec 22, 2019 4:01 pm

Doesn't come as a surprise. Disappointment yes, surprise no.
User avatar
Folderol
Jedi Poster
Posts: 9733
Joined: Sat Nov 15, 2008 1:00 am
Location: The Mudway Towns, UK
Yes. I am that Linux nut.
Onwards and... err... sideways!

Re: Beware 'Pin One' trap

Postby Hugh Robjohns » Sun Dec 22, 2019 7:54 pm

The majority of commercially-made xlr cables will have pin-1 linked to the connectors shells at both ends.

And for systems adhering to the AES grounding recommendations, pin 1 should be connected to the shell. In that way, the cable screen becomes an extension of the equipment's Faraday cage, rather than an aerial to inject noise into the audio reference ground... It also ensures the wires within the connector are fully screened if you extend a cable by plugging another to it... Which could matter in an electrically hostile environment.

At the mic or preamp, or other equipment, pin 1 should be bonded directly to the chassis anyway, so linking internally to the shell makes no difference .

The only time it can become an issue is where the equipment itself has dubious internal grounding arrangements aka a 'pin 1' problem -- in that pin 1 is erroneously tied to the audio ground reference instead of the chassis (which should then be taken to a star earth with the PSU reference ground and the mains safety ground for a class-1 device)
User avatar
Hugh Robjohns
Moderator
Posts: 25823
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: Beware 'Pin One' trap

Postby ef37a » Sun Dec 22, 2019 8:21 pm

Yes Hugh I am aware that pin 1 should go to chassis in the equipment but I get a problem with the point "not audio ground"?

The return for phantom power is the audio ground for a mic circuit so how can it be and 'not' be tied to chassis?

Dave.
ef37a
Jedi Poster
Posts: 11167
Joined: Mon May 29, 2006 12:00 am
Location: northampton uk

Re: Beware 'Pin One' trap

Postby Hugh Robjohns » Sun Dec 22, 2019 8:30 pm

ef37a wrote:Yes Hugh I am aware that pin 1 should go to chassis in the equipment...

Great. Then there's no problem with a connection to the shell is there?

The return for phantom power is the audio ground

Well, strictly, it's not the 'audio reference ground', it's the phantom power supply ground -- the phantom supply being a separate beast to the audio circuitry power supply. It's a small but critically important point when considering current flow along circuit ground paths. The audio power supply reference ground is, of course, normally tied to the phantom supply return ground at the equipment star point.

And, again, the AES recommendation is that the phantom return current should flow via the chassis to the ground star point and back to the phantom PSU that way.

H
User avatar
Hugh Robjohns
Moderator
Posts: 25823
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: Beware 'Pin One' trap

Postby Folderol » Sun Dec 22, 2019 8:42 pm

I must say I'm surprised by this. I always thought that pin one was supposed to be tied to the audio ground, but audio grounds were only supposed to be tied to the chassis at the input.
User avatar
Folderol
Jedi Poster
Posts: 9733
Joined: Sat Nov 15, 2008 1:00 am
Location: The Mudway Towns, UK
Yes. I am that Linux nut.
Onwards and... err... sideways!

Re: Beware 'Pin One' trap

Postby Hugh Robjohns » Sun Dec 22, 2019 9:25 pm

Folderol wrote:I must say I'm surprised by this. I always thought that pin one was supposed to be tied to the audio ground...

No. That is the root cause of 'the pin 1 problem'. It's an almost guaranteed way of injecting noise currents into the audio reference ground, a recipe for disaster if the audio circuitry is a high gain preamp...

In a balanced interface, the cable screen carries nothing relevant to the signal. The balanced signal is not referenced to the screen or ground. And the phantom power supply is wholly independent of the audio ground too.

.. but audio grounds were only supposed to be tied to the chassis at the input.

Are you thinking of the 'one end only' method of trying to break ground loops in systems which have pin-1 problems.
User avatar
Hugh Robjohns
Moderator
Posts: 25823
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: Beware 'Pin One' trap

Postby Folderol » Sun Dec 22, 2019 10:26 pm

Hugh Robjohns wrote:
Folderol wrote:I must say I'm surprised by this. I always thought that pin one was supposed to be tied to the audio ground...

No. That is the root cause of 'the pin 1 problem'. It's an almost guaranteed way of injecting noise currents into the audio reference ground, a recipe for disaster if the audio circuitry is a high gain preamp...
Actually, I was thinking more of the other end. Single ended amps where you could get very large currents in the ground.
User avatar
Folderol
Jedi Poster
Posts: 9733
Joined: Sat Nov 15, 2008 1:00 am
Location: The Mudway Towns, UK
Yes. I am that Linux nut.
Onwards and... err... sideways!

Re: Beware 'Pin One' trap

Postby ef37a » Mon Dec 23, 2019 7:20 am

Folderol wrote:I must say I'm surprised by this. I always thought that pin one was supposed to be tied to the audio ground, but audio grounds were only supposed to be tied to the chassis at the input.

Are you thinking the same as me Will of those old Mullard valve amp circuits? They showed a thick, 16g 'busbar' (heh!) running through the components and only tied to chassis at the input socket. Cannot remember but mains earth must have gone directly to chassis (via a wonderfully retro but potentially lethal! Bulgin socket) and not to the busbar?

Valve inputs were of course very high impedance c 1meg and so had a very high innate CMRR.

I can see phantom power being regarded as NOT part of the signal circuit in theory but in practice it shares earth with the DC supplies, even often produced from a voltage tripler and regulator so it is hard to see how it is isolated?

Bit of a nightmare.

Dave.
ef37a
Jedi Poster
Posts: 11167
Joined: Mon May 29, 2006 12:00 am
Location: northampton uk

Re: Beware 'Pin One' trap

Postby Folderol » Mon Dec 23, 2019 10:40 am

That was where I first saw it, yes. However I've seen the same idea since then with both transistorised kit and with HF (ultrasonics) stuff.
User avatar
Folderol
Jedi Poster
Posts: 9733
Joined: Sat Nov 15, 2008 1:00 am
Location: The Mudway Towns, UK
Yes. I am that Linux nut.
Onwards and... err... sideways!

Re: Beware 'Pin One' trap

Postby Hugh Robjohns » Mon Dec 23, 2019 11:46 am

ef37a wrote:I can see phantom power being regarded as NOT part of the signal circuit in theory but in practice it shares earth with the DC supplies

Yes, the ground (return) of the [phantom, supply is tied to the ground (reference) of the audio circuitry PSU... the critical part is where you tie the two together.

The current received wisdom on this is that it should be done at a star earth point which collects together the grounds from the audio circuitry PSU, any digital circuitry PSUs, phantom power PSU, the chassis itself and, if present, the mains safety earth. All connection socket screens also go direct to chassis.

The core idea is to keep large and (potentially) noisy currents away from the audio circuitry ground references within the audio PCBs.

H
User avatar
Hugh Robjohns
Moderator
Posts: 25823
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: Beware 'Pin One' trap

Postby ef37a » Mon Dec 23, 2019 3:04 pm

Hugh Robjohns wrote:
ef37a wrote:I can see phantom power being regarded as NOT part of the signal circuit in theory but in practice it shares earth with the DC supplies

Yes, the ground (return) of the [phantom, supply is tied to the ground (reference) of the audio circuitry PSU... the critical part is where you tie the two together.

The current received wisdom on this is that it should be done at a star earth point which collects together the grounds from the audio circuitry PSU, any digital circuitry PSUs, phantom power PSU, the chassis itself and, if present, the mains safety earth. All connection socket screens also go direct to chassis.

The core idea is to keep large and (potentially) noisy currents away from the audio circuitry ground references within the audio PCBs.

H

Yes, earthing screens to chassis at entry point is certainly my understanding but this makes the diagram on page 600 of Self's Audio Power Amp Design (ed 6) a bit confusing? The input socket screen is shown taken back to the mains earth chassis point via a thick wire but there is also a screened cable running from input socket back to the signal star point. Two earths! I cannot tell if the input socket screen is tied to chassis locally as it is a line diagram.

This of course is for an unbalanced input.

I have read that if everyone followed the same internal grounding regime (Rane?) that ground loops would be a thing of the past?

Lovely idea! But, 'tis Christmas so you never know?

Dave.
ef37a
Jedi Poster
Posts: 11167
Joined: Mon May 29, 2006 12:00 am
Location: northampton uk

Re: Beware 'Pin One' trap

Postby Folderol » Mon Dec 23, 2019 3:20 pm

Indeed, and if everyone looks towards the light they might come to see the...
One True Way
... whatever that is :lol:
User avatar
Folderol
Jedi Poster
Posts: 9733
Joined: Sat Nov 15, 2008 1:00 am
Location: The Mudway Towns, UK
Yes. I am that Linux nut.
Onwards and... err... sideways!