Patchbays.. how do you do it?

[Lee Cardan]

For basic stuff, In > Out, easy: In on top, Out on Bottom

But say a channel strip: Mic In, Line In, Out
Or a PCM 70: Mono In, Stereo Out
Or a gate or comp w/sidechain: In, Key, Out

What is the standard for laying this stuff out?

I ended up going left to right for most things because in the end it was easier to label and still made sense to me- everyone else though I was a wierdo though anyway time is coming that I'm going to have to redo all this stuff again so I'd be curious to know what (if any) is the standard here!

Thanks

[Chinagraf]

Standard is whatever works for you really. When the numbers don't add up you can either leave say one of the top patch points empty in the case of a mono in stereo out fx, or you can diconnect the normalling and use it for something else that you only need one patch point for. It's even more of a pain when hanging outboard off converters because you are leaving one converter output or input sitting there going to waste. In the case of a key input, I use to have them normalled off a buss on the console. So say buss 32 out on the top row normalled to key input on the bottom. Then you just assign to buss 32 whatever you want to key with and no patching neded, but if you want to treat the key signal with outboard you can still patch into an eq or comp.
I reckon the more defining issue is normalling and how it can work best for you in you setup for your workflow.
So for me, I'd have 2 separate normalled patchbays. Top bay slot 1 would be send from console/daw on top row with mono fx input on bottom. Next patchbay would be fx output L/R on top row slot 1 and 2 and console/mix inputs on bottom. Then no patch cords needed unless I wanted to use that particular send with other fx units. Leaves you slot 2 on the top patchbay as 2 patch points that are fully normalled for something else.
ie I have top row - converter outputs normalled to next row down outboard inputs, then next row down is a new patch bacy with top row being outboard output normalled to bottom row which is inputs back into the mix. If just doing usual stuff I don't need to use any patch cords, but when tracking I can patch pres into all my outboard.
The other thing to mention is if you have inputs and outputs above /below each other on the same patch bay disconnect the normalling on those particular points.
Patchbays are like Chinese tile puzzles and just when you think you have it perfect you'll get a new bit of gear and want to reshuffle everything.

[Wiz]

I have outs on the top, ins on the bottom.

PB 1-8. Are preamps out normalled to interface ins.


9 thru 16 are comps and eq set to direct

17 is headphone out normalled, headphones at each end of the room.

Then there are 8 outs set as direct, 2 of which are permanently attached to my VUs , the others are spare outs for things like setting up headphone mixes if I am doing sessions with others.

Mines not labelled, I just have it memorised.

[Drumstruck]

Glad to hear I'm not alone in labouring over patchbay design - and looking forward to hearing more thoughts from you pro guys

I'm down to 1 patchbay now and that's an RCA to TS normalled unit just for that anomaly rack of RCA based gear (CD, cassette, 7.1 amp). In on the top (rear) and out on the bottom (rear) normalled for that one.

I got rid of the rest of my patchbays and am using the digital router in my desk under the "minimised cabling and conections" philosophy.

Outboard rack gear is direct connected to the console via XLR<->DB25s into dedicated analog I/O cards - and then manually connected or sidechained between outboard units as needed.

If I had more than 2 outboard racks I'd probably go back to the patchbay model with ins on top (rear) / outs on bottom (rear) and all normalled.

[audioio]

I have outs on the top, ins on the bottom.

This makes sense in so many ways, for normalling (whether half or full). For example...
* Mic pre out to console line in (or converter/interface in)
* Channel insert send to insert return
* Channel direct out to converter/interface in
* Converter/interface out to "tape" monitor in
* Compressor sidechain send (or point parallelled from comp input) to sidechain return (key or detector in)
<trollbait>
* Studio connector (out) to mic pre in
</trollbait>

See previous discussion (good-natured argument) about whether or not one should put mic-level signals on a TT patchbay. YMMV!

If the out is on the bottom you can't normal.

[Lee Cardan]

All interesting stuff!
I still don't get why outs on top are better?? I totally get normalling but i still struggle to come up with any real 'templates' because certain things just end up all over the place all the time.

So question number 2:
How do you devise your normalling schemes?

[Wiz]

for me, as I said, say outputs of my mic preamps are normalled to my UCX audio inputs.

Another one, might be the outputs of the RME UCX, say outs 3 and 4 are normalled to my bus compressor inputs, and the buss compressor outputs are normalled to the RME UCX inputs 3 and four.

So, if I do nothing, the buss compressor is strapped across the UCX, but If need to acces the RME's outs and ins, I can via the patchbay.

THe sampson S patch is fabulous, as it has a switch you can easily change on the front, to select, Thru, Half normalled and normalled.

I also have headphone out normalled, so if I plug a set of headphones into the patch bay, I can hear it... and when I unplug, it routes to my headphone out, down at my drum kit...

[audioio]

I still don't get why outs on top are better?? I totally get normalling

It's because of normalling. If the the two points are connected via half or full normalling and you want that connection, then it wouldn't matter. But if you plug a cable into the bottom point it cuts off the connection with the top. If that bottom point were the output, you'd be disconnecting it from the input above. With a full-normal scheme this wouldn't matter but half-normalling allows splitting of the signal. With the output on the bottom you can't do that.

It's also about signal flow, I think. Top to bottom, with few exceptions, for consistency. Less mental load (and I need much less mental load!)

How do you devise your normalling schemes?

Probably full-normal on mic to mic pre connections unless you keep them off the bantam bay as some here do.
Half-normal almost everywhere else. This is particularly useful for inserts, especially if you don't have an insert in switch. You can split a signal to a compressor and adjust it to whatever gain reduction you want using the meter, then patch it back to the return at an opportune moment (like not mid-phrase when the singer is doing a run-through and causing a level change in the cans).

What does everyone else think?