Skip to main content

Patchbay "normalled" explained

Can someone explain "normalled"

  1. What is fully normalled? what are the pros and cons and how is this used?
  2. What is half normalled? what are the pros and cons and how is this used?
  3. How is each used respectively and why are pieces normalled and not?



RecorderMan Mon, 09/09/2002 - 08:26
These terms come up when speaking about patchbays.

First, imagine a patchbay.....the one I see in my imagination right now has two rows. The top row are the tie lines from my mic panel in my imaginary tracking room....the row just below that row are the microphone pre-amp inputs into my imaginary NEVE8028 (pretty good imagination...huh?).

OK. If the top to bottom row was "fully-normalled" that would mean that if there were NO patch cables connecting a mic tie from the top row to the mic pre -input on the bottom row, signal flow would still happen, because these two points are "fully normalled" (or connected).
Half-normalled, would mean that they are not connected-in such a way -so that if I plug a patch cable into the top row (mic tie return), the connection to the bottom row (mic pre input) would be broken. I could now, with a patch cable, cross patch into another mic-pre, and not worry about the signal also continuing into the pre just below the the tie line we were speaking about. In the original fully normalled example, a cross-patch would not break the connection, and since the mic would be going into two pre-amps - the original one that is normalled to the tie line, and the one that we were cross patching to - there would be additional loading on the mic and therefore a loss in gain and frequency response.

Doublehelix Mon, 09/09/2002 - 08:44
Yeah, this is a tough one to get your hands around initially, but once you do so, it is pretty easy to understand. It is one of those "a-ha" moments where after hearing it explained about 20 times, you all of a sudden say "Oh...I get it.". It is really not that difficult a concept, but it really troubled me too until recently.

Here is a *very* simplified explaination from a real novice. There are obviously many exceptions and variations, but this should at least get you started. I am also over-simplifying this a bit...

Bascially in a patchbay you have a series of jacks on the front pannel, usually 2 rows, one on top of the other. Each jack on the top row, can be said to be "matched" with a jack on the bottom row.

Now, on the back of a patch bay, there is usally another set of jacks that connect directly to the ones on the front in a one-to-one correspondance.

Now, let's say you wanted to add your mixer's insert points to a patchbay. This is a great idea since it is now pretty easy to patch in a compressor or gate or EQ (or whatever) when you need it, on whichever channel you need to at the time.

Since most budget mixers have inserts that are TRS (but are unbalanced), you need to plug a Y-cable into your insert jack on your board, and have one lead go to the top jack on the back of the patchbay, and one to the bottom jack on the back of the patchbay.

(boy, this is harder to explain that I thought)

Now, in a "normalled" configuration, the top jack and the bottom jack are automatically connected internally (inside the patch bay), so if you do "nothing", these 2 jacks are connected a if you had connected the 2 ends of the Y-cable back together again. If these cables were not connected back together in this way, the signal would not be sent back to the board, and would be interupted at the insert point, and no sound would pass thought that channel on the board.

But now let's say you wanted to add a noise gate to that channel. Well as soon as you plug a patch cable into the top jack on the patchbat, you have interupted that signal path, and are routing the signal elsewhere (in this case it would be to the input jack on the gate, which hopefully has also been wired to your patch bay!). Next, you would take the output from your gate, and patch it back to the bottom jack on the insert point connection, which re-completes the singal path once again to the insert point on your board.

(Wow! This is so simple, but is so difficult to explain!)

Let me try to summarize:

Normalled: The top and bottom jacks on a patch bay are internally connected so that the signal will pass from the top jack to the bottom jack. As soon as you plug a patch cable into the top jack, this connection is broken, and the signal now flows out of the top jack, into the patch cable, but not back into the bottom jack.

Half-Normalled: This is a very similar setup, without any cables, the top and bottom jacks are internally connected and the signal flows from the top jack to the bottom jack. This time however, if you plug a patch cable into the top jack, it does *not* break the signal path between the top and bottom jacks. Now, the signal is split, with signal going *both* to the bottom jack as well as to the patch cable. This configuration allows you to "monitor" the signal for example without breaking the "normallized" connection between the 2 jacks. Now, if you add a second cable to the bottom jack, the normalized signal path will be broken, and *all* of the signal will pass out of the top cable.

Non-Normalled: The top and bottom jacks are not connected internally at all.

Uses: Insert points work well normalled, but some can argue that half-normalled works better here so you can monitor this signal. This is the method that makes most sense to me. For Effects processors sends and returns: Definitely use non-normalled. You don't want your signal going out and then back into itself here! Dangerous!

Most patch bays allow easy switching between these three modes on an individual basis (top/bottom pair). Beware of less-expensive models that force you to switch the entire patch bay at once (I don't even know if these are even available any more).

I hope that helps! I had a really tough time explaining that one!

Kurt Foster Mon, 09/09/2002 - 09:02
The term normaled is used in reference to patchbays / jackfields. It is a way of setting the patchbay up where the signal from one row of jacks, for example the top row, is passed to bottom row of jacks or vice versa. Normals are generally used to pass signal from aux sends to efx processers and console outs to tape/ disk and tape / disk returns to channel inputs. Generally normals are not used on dynamics processers. On a full normal you can plug a patch cord into either jack and it will interupt the signal. A half normal is where only one of the jacks is wired to interupt signal if it is pluged into. It may be the top or bottom jack. Half normals are usually set up so the top row, or output of a reverb for example, is not interupted (or not normaled) if it is pluged into but the bottom row, the aux return, is. This way the operator may split the output of the reverb, routing it to another destination without interupting it's path to the aux return. This is called a mult. But if the operator wanted to patch a different efx processer into the aux return it would be possible because the bottom row, or aux ret., would be interupted by plugging the jack cord in. Many of the more inexpensive 1/4" patchbays available on the market today do not facilitate all of the normal configurations. Usually they can only be set up to preform half normals or no normals. These configurations will work in most applications but if you need the ability to do full normals ( where inserting a patchcord into either jack will break the signal path) Pro Co is the only company I know of that makes these. Patchbays are an expensive but integral part of your studio. Be sure to get professional assitance in setting them up if you feel you need it. In the end it will pay off in savings both in time and money. A patchbay / jackfield that is correctly set up will speed your sessions along making you look proficent but a bad patchbay setup can only reflect poorly on you, crud up your audio and make your recording expierence a living hell...... Fats

Member Mon, 09/09/2002 - 10:17

Your patch bay is like a railroad station....

Normalled is like a drawbridge....

Half-normalled is like a drawbridge with an offramp to the clam place your mother likes.

Normalled is understanding this....
Half Normalled is merely thinking you do.

Not Normalled is chucking the patchbay in the dumpster and going back to crawling around behind the gear with your hair full of cobwebs, an eye out for black widow spiders, a bunch of guitar cords around your neck and a Mini Maglight in your mouth.

Heaven is when digital gets cheaper than potatoes, better than analog, and anticipates your every move. I think I will call that Finally Normalled. I will know it when I get there....

Seriously, the guys who populate RO are saints to be willing to tackle this explanation without pay. It's kinda like God -- if you don't know Him, no explanation is sufficient. If you DO know Him, no explanation is necessary...

Me, I'm too lazy to try this one on, but I'm a regular Googlemeister and I found this link for you:


Matter-of-fact, Google may be the best advice I can give you. It is, by far, the best search facility I have used, as it will search for web content AND newsgroup/bbs content on any subject under the sun. I just went to Google and typed in "patchbay basics" and hit the WEB button and got like 300 hits. Try Google for all your brainbuster needs.

Good luck !

:cool: RW :cool:

e-cue Mon, 09/09/2002 - 21:06
Normalizing means to adjust the peak volume of a selection to a known value. Generally the recommended maximum is -0.5 dB (I think that's 94.49% or something)... oh wait, we talking about patchbays... I really should read before I type.

This can also refer to mixers, such as the mackie 32 8's.

This message has been edited for druken typos.

Member Tue, 09/10/2002 - 06:18
Thanks all for the info. The way it works is easier to understand. But its the application that makes it harder to grasp since it depends on how the user wants to set up his patchbay.I will show how I think I want my patch bay set up and maybe 'you' can tell me what I need. I want to plug all my compressors/gates into the patchbay so when I need one I just plug a 1/4 trs cable from the insert point on the board to one point on the patchbay. would a signal chain work if I wanted to put an EQ into a compressor into a gate and have it all connected back to the insert point on the patchbay? I am using the 4 aux sends to power a headphone amp. But if I need to add reverb or something like that I would need my aux sends, I guess i will have to plug the headphone inserts and reverb etc into the patchbay also. So after all this what kind of patchbay would i need?

Kurt Foster Tue, 09/10/2002 - 07:51
A full patchbay would include all connections in the system. If you want to only have your compressors, reverbs and headphone amps wired to a bay that would work also. This will save a ton of money because you will save in all the cableing and patchbays.
The compressors should not be normaled. Run the inputs L / R to the top row of jacks on your patchbay for example jacks 1 & 2 (top). Compressor outs would appear directly under the inputs, jacks 1 / 2 (bottom). The aux sends may be 1/2 normaled to the reverbs or the ph amps, top row sends, bottom row amp and rev inputs, with the path being broken by inserting in the bottom row (or ph amp or reverb inputs).
Reverb outs would be wired to the top row, being returned to aux returns or channell inputs 1/2 normaled with the path being broken by inserting in the bottom row. Usually aux's that are pre fader would be normaled to phone amps and post fader aux's would go to reverbs but either may be patched for what ever application your performing at the time. . . . . . .Fats

knightfly Fri, 09/13/2002 - 10:16
Stephen - Fats and others have pretty well covered the facts, other than I've seen several bays that allow switched normalling such as Neutrik, Furman, etc - Here is yet another viewpoint which may simplify things for you.

I think of patch bays as merely a way of extending all those pesky rear panel connections that are nearly impossible to get to, so you can change things around temporarily to accomplish a task that is not "normal" - as Fats pointed out, there are some pieces of gear you should not normal, some that should be half-normalled, and some that should be full-normalled.

The way I go about deciding what to bring to a bay and which type of normal to use, is to make a drawing of each piece of gear you have/plan to buy; you don't need to be Michaelangelo, just a rectangle for each piece of gear, big enough to show in's and outs will do for audio routing. I set up the drawing according to proposed signal flow from input to output(s), leaving enough room for wires to be drawn between boxes. (I do this in a basic CAD program, makes it easier to do changes) Once you have a drawing of all your gear, and the way it "normal"ly will be connected, you can then decide which connections you will want to change on a regular basis, which you may want to change occasionally, and which ones you'll only change if a connection goes bad. These last types are a waste of money for patch bays...

If on a budget, you would use this info to prioritize and to figure out how many bays you need for your system.

From the drawing, I then use different colored double lines drawn perpendicular to the connection lines, depending on whether the patch is to be half, full, or non-normalled. For example, you could use red, blue and green for the 3 main possibilities. These marks would look like a schematic drawing of a capacitor, except that the line continues thru the two perpendiculars.

Once you have decided which connections will be done which way, just count the red, blue, and green hash marks and that's the number of PAIRS of patch points you need, minimum.

The next drawing to make is a pictorial drawing of the actual bay, from the front. Draw circles for each jack, leaving room between for printing source and destination information. As mentioned above, usually top rows are gear OUTPUTS. If you, for example, were using an OctoPre "normally" connected into the line in's of the first 8 channels of your board, you might label the first 8 upper jacks as "OctOut1",......."OctOut8", or more generically "PreOut1", etc. Then the second row (wired to the first 8 line in's of the board) would read "LinIn1", etc - Bays usually don't give you much room for labeling, so you need to find a shorthand that works for you - growing up with DOS instead of Windows helps you learn to be concise here. I've had fair luck using laser labels with 6 point type in Word, then trimming them small enough to fit, then finally covering with trimmed lengths of 3M magic mending tape (low glare, better readability)

Some of the bays I mentioned allow you to just flip a small circuit card over after removing the front or rear of the bay chassis in order to change the normalling. Others require going into the connectors and re-soldering. For more permanent installs, I would recommend the solder type, which also tend to have soldered rear connections. These take longer to set up, but the less plug-in connections you have in a patch system, the less likely you will have connector corrosion problems later. I would STRONGLY recommend that you wire the solder type bays with wires routed to the same side of the rack and dressed neatly so that you can open the bay either to the side, top or bottom (choose one, depending on accessibility), this way you can open the bay and get to connections after it's installed, and change or add connections easily.

I would also recommend that you use a consistent labeling method for all cables, so you can tell without having to trace a wire physically just where it goes to and from. There are electician's labeling systems available thru Newark electronics (one example) - one type comes in a red dispenser and has peel/stick labels that have one section frosted white with the rest clear. You write thru a window in the dispenser onto the white part, peel the label off and wrap it around the cable. The clear part covers the written part to protect it. The advantage of this type is that you can label AFTER you put connections together.

I find that the best arrangement of bays for ME is to follow signal flow. I put all the "normal" connections, as in the usual way things are connected, at the upper left. Compressors and other gear that isn't (almost) always in the signal path, get their own bays (usually non-normalled ones)- I also try NOT to have gear that needs to be adjusted frequently placed below the bays . Just too many cables in the way.

Hopefully I managed to stir the mud even more, making an already confusing issue totally indecipherable... :=) Steve

Kev Fri, 09/13/2002 - 11:55
Hi Steve, :w:

I'll keep this short and simple

Patch bays are good and worth as much money as you can spare.
Some have switchable Normal/ Halfnormal ( I prefer the term SNIFF and BREAK)
IN on the bottom row
OUT on the TOP row
Use the best wire you can afford. Canare MR202-24AT is good to work with

Neutrik have a Patch Panel that has 4 elco connectors. I use one 24 way cable per connector 12 to female XL3 and 12 to male XL3. In other words 12 coming and 12 going. Slide in labelling. Totally modular and when you are growing fast nothing remains the same.

I set this up for one friend and now they all want it. I will be in patch panel hell for at least the next year. It does take time.

Member Wed, 09/18/2002 - 08:21
Thanks for the help. I am still trying to figure out which 'normal' to use for each piece of gear.
Maybe what would help me to finally understand and others for that matter if they are having trouble with this to is...what happens when a device is hooked up incorrectly such as a compressor into normalled and a reverb into a half normalled.