Ground lift switches

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
i dont recall ever seeing a patch bay with a ground lift switch ....

ground lift switches normally exist behind amplifiers but still these are not necessary to anyone that knows how to install audio equipment properly .

If you focus and manage to understand what causes hum in PA installations you may only use a ground lift switch only in a case of emergency .

Also i think that if anyone installs/adds a ground lift switch in a patch bay is wrong practice ....
 
Patch bay may be a poor choice of words. I don't want to explain this now, so nevermind, I'll find what I need myself. This is more like an intermediate node in balanced lines, and will usually connect the third pin thru, but sometimes I might not want to (which is a wonderful feature of balanced differential lines).
 
I had a patchfield with ground lifts, once, while touring a big keyboard rig. Eighteen keyboards, some mono, some stereo, and four stereo subgroups. The house mix and the monitor mixer could choose individual outs, grouped outs or the entire mix, which also went to the local monitor (but had effects mixed with it, that the FOH guys didn't always want.

Some of the keyboards had three connector mains leads, some two, but all had transformer balanced outputs. Every ground lift switch had a neon round it; if this lit up I knew there were serious problems. Outputs were on XLRs and multipole connectors, separately isolated. Mic preamps (for grand piano, leslie and pipe organ) were brought up balanced in the patch. Sometimes it took a while of flicking switches to achieve optimum performance (and you couldn't start until the lights were rigged and radiating their nasties), but if I ever toured an equivalent system I'd build another. Independent DI boxes might do the job quite adequately, but it's so nice having everything grouped together (isn't that why we invented patchfields in the first place?)
 
I think I found some of what I wanted to know. If you lift (don't connect) the ground pin on one side, the plug and jack shells will still connect. Which is why you're not supposed to connect the shell (or panel or jack) to the ground pin on the female. Does this sound right? I've been out of the business for far too long to remember... This has other implications I need to remember...and I need to check some binding screws on some connectors. Somebody must have the guidelines somewhere...I need to search more.
 
If you're going to do a ground lift on an RTS jack (or bantam jack), you have to have the sleeve isolated from chassis. Most professional patch fields give you this automatically, and let you bus your grounds. With an XLR patch, you generally don't connect the cable screen to the plug shell anyway, so a pin 1 lift works fine.

With a proper balanced signal, there should be nothing passing by the ground, anyway; it's just a reference, irrelevant (except on phantom powered circuits). So you can put a smallish resistor, or paralleled head to tail diodes from chassis to pin one, so if any voltage is generated between apparatuses it's got an escape route.

Is this for fixed installation/studio or a touring rig? All my patch leads have always had ground continuity, and the patch has always been the central star ground (however little this is supposed to count in balanced systems, and in forty years of studio I've never had a ground current problem when the signals were truly balanced, or floating. I can not say as much for outside gigs, live recording or PA.
 
Like I said, this is all XLR. The first thing I'm making (finishing tonight) is a rack panel with a bunch of sections. Each section has one XLR female input and two XLR male outputs and a DPDT switch that reverses pins 2 & 3 of the input female for a "polarity" (or 180 relative phase) forward/reverse switch. This is just for convenience and speed for ad-hoc setup of mixed equipment and snakes to various active crossover inputs and outputs for various mixes, channels, and feeds (plus an organized visual display of the state); and the dual paralled outputs are just instead of "Y" cords, usually to drive both channels of stereo amps with the same program material. Might occasionally use some sections for getting stage mics and di boxes into correct relative phase with each other etc.

Diagrams I've seen show XLR cables (like mic cables) wired with the metal cover shell over the female connector not connected to pin1, and warning that ground lifts are not really SAE standard at all. So if you typically buy mic cables, are the shells (metal cases over the plug or jack on a cable end) connected to anything at all (besides each other when plugged into each other), or is pin 1 only connected to the "shell" on the males?

I'm starting to think I'm just going to leave the gournd lift switches out, as I won't need them immediately at all. I was just kind of thinking ahead and used to having them available on snakes if I ever needed them.
 
Last edited:
Try this. Wire your patches up complete. Then make a couple short (6" or 12", whatever) XLR cords with no ground between ends. Now any time you need a ground lift, insert your short cable in series with the connection. No switches, no special jacks. When I was touring, I would never go out without a few of those in my problem solvers case.

Other items in there are a TRS plug with simply tip to ring jumpered inside. Bypasses insert jack normals. And a shorted plug to stuff in a footswitch hole if needed. And a bunch of adaptors.
 
Yes, I have boxes of that stuff from the 1970's. In-line XLR w/ built-IN T-pad or H-pad, various adapters.

Looks like I had that backwards, they more often connect the female shells to ground or don't connect either shell to ground. And apparently the US and Europe have opposite standards for the absolute phase of pins 2 & 3.
 
I would quibble a bit.

The shell is always chassis. Pin 1 is usually sent to circuit common. You may or may not wire the shell and pin 1 together. Depends upon the rest of the circuit and its needs.

The cable shield only needs to be grounded at one end to be effective, leaving pin 1 vacant at the remaining end.
 
Suppose we are talkin patchfield here, rather than devices. Chassis would be the rack in which it would be mounted. So insisting on chassis grounding for all XLRs means (more or less) that the patchfield becomes our central star gounding point, all ground loops are between it and the power grounds of the various mixers, amps and outboard gadgets wired to it. So far, so good, particularly if all inputs and outputs are truly balanced, or all floating. Now, consider a stage system with a FOH set up, desk, equalisers, inserts, patchfield, crossovers etc. Patch reference ground, no problem. Now, on the other end of the tieline multiway we have two amp racks one each side of the stage, also with patches. All the amp signal grounds to the local patch, or signal cables running through.

But I'm not finished. There is another mixer, with the stage mics split into it, running stage monitors. Own equalisers, crossovers, antifeedback gadgets, power amps, in-ear transmitters; needs ground options to optimise performance with FOH system. Doesn't have to be switches, but some form of removable link.

And, just to maintain the interest, there's a mobile recording truck parked in front; own everything, plus intercom (does anyone build a fully balanced intercom, guaranteed not to generate loops?) Cooperating with a video truck, and video thrives on ground loops; they haven't discovered balancing yet. I want all my tracks to be as free of lighting spike as possible (actually I want them a bit cleaner than that), but I challenge anyone to make a diagram indicating ground currents in this system; everything is balanced (except the insert points on the monitor console, for some reason) but it still generally takes a while minimising interference between systems. And that's without my aforementioned keyboard system…
 
Andrew ... now don't be so sure, don't be so ready to use words like always...

Pin1 does not carry any audio signal.correct
Pin1 goes to Chassis. Always. also correct while the next of interest is the status of the Chassis IE connected to the electrical ground and or via the ground lift switch to 0 volts ( signal ground ) of the power supply ?


My understanding of Ott:
Pin1 goes to Chassis, never audio ground. WRONG ! this what a ground lift switch does connect 0V of power supply ( signal ground ) to the electrical ground which happens to be the Chassis



in extensive installations as described above the number one reason is that you cannot guarantee that the all set up is in ground lift or not mode ... any of the devices connected that is in Ground on mode rather in a ground lift mode will produce buzz

beyond that there is other distribution of power and signal issues that may produce buzz of any kind

Had to operate a similar setup as described above as head technician of the meeting for the European Ministers of Foreign Affairin Athens 2008 and name it was there every Tom Dick and Harry
20.000 sq meters area with cctv , 120 tv,s broadcasting rooms , tv vans telephone interfaces , audio systems , motorized cameras , multilingual translate systems and multilingual CCTV ...

What i ve done ?
asked my people to follow the rules during the installation and got my self 10 Ground loop isolators cost 12 E each and 2 video ground loop isolators that i never used cost 100 E each .

--asked my people to notify me before doing anything in case of any buzz
--got 6 of them produced from the video modulators (audio part ) that feed the CCTV
--one buzz coming from a TV van of unknown cause ( Azerbaijan i think ) asked the guy if he was ok with unbalanced audio he said yes so i cut him down installing isolator between my audio and his
--And got a video loop from one of my motorized camera due to a BNC broken plug ...

That's about it
 
If I had the ability and the Author's permission I would show you the diagrams and the text where H.Ott tells us repeatedly that the screen is never connected to audio ground.
The Pin1 problem is an explanation of exactly the same thing.
Pin1 does not go to Audio Ground, it goes to Chassis.
 
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.