Crazy idea? Phantom supplied patchbay

Help Support GroupDIY:

DaxLiniere

Well-known member
Joined
Aug 10, 2004
Messages
163
Location
London, UK
Hey guys!
I run Puzzle Factory Sound Studios, once in Australia, now in London. I have B-gauge patchbays in my control rooms that I use to connect the wallboxes in my live rooms to my preamps/console.

I have been thinking it would be useful to build a 1RU strip with a whole lotta switches which can deliver phantom power to each of the individual B-gauge sockets, rather than using the phantom supplies in the preamps themselves.

But why? I hate patchbays in general, even more so when it comes to sending phantom through them. Even when properly cleaned and using brand new quality patch cables, a little bump can sometimes cause a loud crackle.
If phantom was wired directly to the back of the patchbay (at the cables that run to the XLR wall boxes), it would eliminate this PITA.

A question I have; would this affect crosstalk and, if so, is there a way to eliminate that?

I believe the phantom power spec is 100mA max per device, so with that in mind, a 5A linear regulator will do the job easily. To be honest, 3A is probably enough, knowing that many mics won't need phantom and very few will draw the entire 100mA.
There are a few low-noise regulators that will do 48v output such as https://www.ebay.co.uk/itm/Assembled-LT1083-Adjustable-HIFI-Linear-Regulated-Power-Supply-Board-Dual-0-48V/253477392391?hash=item3b046e1007:g:iowAAOSwa2lZfOFZ

I'm happy to buy expensive 6k8 resistors and carefully match them for each output.

Any thoughts on how best to approach this project?


All the best,
Dax.
 

abbey road d enfer

Well-known member
Moderator
Joined
Jan 22, 2008
Messages
12,621
Location
Marcelland
DaxLiniere said:
A question I have; would this affect crosstalk and, if so, is there a way to eliminate that?
No issues with x-talk, however that implies the patchbay "grounds" need to be connected together. Are the grounds floated or tied together at the patchbay?
Anyway, they have to be connected somehow (usually via the preamps' safety ground).
It's just a warning that you will have to make a choice regarding where you connect the 0Volt of the 48V supply.

Any thoughts on how best to approach this project?
Just check that you have no issues with tiing the grounds together at the patchbay first. Then experiment with two channels that are powered from the most distant plugs.
 

Rob Flinn

Well-known member
Joined
Jun 3, 2004
Messages
4,501
Location
Sussex, UK
I used to run a tie line directly into each mic amp to avoid running mics through the patchbay.  I selected the  mic amp by plugging into the right socket in the live room, thus eliminating that flakey connection.  I had a piece of 48 pair cable to do the job.
 

abbey road d enfer

Well-known member
Moderator
Joined
Jan 22, 2008
Messages
12,621
Location
Marcelland
Rob Flinn said:
I used to run a tie line directly into each mic amp to avoid running mics through the patchbay.
I agree. Although I reckon it is possible to use a jackfield, it's too risky. In the heat of the moment too many things can happen.
In my studio I have an XLR patchbay. It's not pretty, it's bulky, it's not as easy as jacks, but it gives you time to pause.
 

JohnRoberts

Well-known member
Staff member
Moderator
Joined
Nov 30, 2006
Messages
20,484
Location
Hickory, MS
Inadvertent shorts while patching mics in a patch bay can dump amps of current into preamp electronics from charged up phantom caps.

Low level mic signals don't need to go through extra switch contacts.

As Abbey already shared, any external phantom supply needs its ground bonded to pin 1.

JR
 

Rocinante

Well-known member
Joined
Jan 17, 2014
Messages
1,132
Location
Minnehopeless
Yeah this reeks of a bad idea. If you have a new intern or tech or like me are babbling while doing stuff things could get ugly. Why do you need this?
 

mjrippe

Well-known member
Joined
Dec 27, 2011
Messages
2,514
Location
Brooklyn, NY
Too many possible problems.  Like what happens if phantom is tuned on at the preamp as well?  Or you plug phantom in to the sidechain input of a compressor, or output of a digitial fx, or...
 

DaxLiniere

Well-known member
Joined
Aug 10, 2004
Messages
163
Location
London, UK
Hey everyone!
Thank you for your replies. The forum didn't send me any after Magneto's reply.
I'll take a read through and get back to you.

Thanks again!
Dax.
 

gyraf

Well-known member
Staff member
Moderator
Joined
Jun 4, 2004
Messages
10,205
Location
Aarhus, Denmark
In our calrec UA-8000, everything was floated balanced, ground free. Which meant that supplying P48 was a specialized task for specialized "mic lines". All these met at the top of the patch bay, where the phantom switches were also located. So I kinda get your idea, I think.

On the other hand, our SSL4044 only had one phantom power switch, located at the PSU in the basement. Was never switched off during the 15 years we had it there. This why I know the ribbon-mic-and-phantom scare is exaggerated heavily. Regarding P48 potentially killing active-balanced line outputs, you simply do not connect a line output to a mic input: this is what line inputs are for.

/Jakob E.
 

DaxLiniere

Well-known member
Joined
Aug 10, 2004
Messages
163
Location
London, UK
Thanks again everyone for your reponses.

MagnetoSound said:
Spec is for 10mA per channel, not 100mA.
Might save you a few quid 😉

Oh, yes! Thanks, good catch.

abbey road d enfer said:
No issues with x-talk, however that implies the patchbay "grounds" need to be connected together. Are the grounds floated or tied together at the patchbay?
Anyway, they have to be connected somehow (usually via the preamps' safety ground).
It's just a warning that you will have to make a choice regarding where you connect the 0Volt of the 48V supply.
They are all currently floating. The top rows are normalled to bottom row through switching contacts, but the grounds are all currently separate. I know I need to solder in a big ol' length of wire to bridge them all.
Connecting input grounds of all preamps/console seems like the right idea anyway, right?

Rob Flinn said:
I used to run a tie line directly into each mic amp to avoid running mics through the patchbay.  I selected the  mic amp by plugging into the right socket in the live room, thus eliminating that flakey connection.  I had a piece of 48 pair cable to do the job.

I used to, too, but it's not practical with a console. I truly hate patchbays and all the reliability problems they bring.

abbey road d enfer said:
I agree. Although I reckon it is possible to use a jackfield, it's too risky. In the heat of the moment too many things can happen.
In my studio I have an XLR patchbay. It's not pretty, it's bulky, it's not as easy as jacks, but it gives you time to pause.

a) I have 40 lines and b) B-guage patchbays are already installed. See next point for risk.

Rocinante said:
Yeah this reeks of a bad idea. If you have a new intern or tech or like me are babbling while doing stuff things could get ugly. Why do you need this?

I'm planning to use locking switches, the type where you have to pull the 'stem' before it will move position.
Plus, this is a technical environment, but public playhouse. There are always details to pay attention to. :) I'm okay with this.

mjrippe said:
Too many possible problems.  Like what happens if phantom is tuned on at the preamp as well?  Or you plug phantom in to the sidechain input of a compressor, or output of a digitial fx, or...

Mic signals are on B-guage, line signals are on TRS so this is not easily possible. I don't think there is a problem if phantom is on at the preamp as well because a) there is 13k ohms between the two PSUs and b) they would be in parallel, not series.

gyraf said:
In our calrec UA-8000, everything was floated balanced, ground free. Which meant that supplying P48 was a specialized task for specialized "mic lines". All these met at the top of the patch bay, where the phantom switches were also located. So I kinda get your idea, I think.

Hey Jakob, long time, no speak. Yes, you've got it in one. A bank of 24 switches on a 1U rack plate directly above each of the two B-gauge patchbays. Switches are vertically aligned so there's no confusion and locking switches to prevent accidental bumps. Carefully matched pairs of Vishay 1% 6k81 resistors and a yet-to-be-determined 1 amp super low noise PSU.
(If I had thought earlier that I would ever install a patchbay, I would have allowed for LEDs at the mic wallplates I had laser cut to indicate phantom present. Too hard to drill now as they're stainless steel.)

gyraf said:
On the other hand, our SSL4044 only had one phantom power switch, located at the PSU in the basement. Was never switched off during the 15 years we had it there. This why I know the ribbon-mic-and-phantom scare is exaggerated heavily. Regarding P48 potentially killing active-balanced line outputs, you simply do not connect a line output to a mic input: this is what line inputs are for.

That's really interesting to hear. And were those mic lines on bantam or XLR patchbay?

All the best,
Dax.
 

DaxLiniere

Well-known member
Joined
Aug 10, 2004
Messages
163
Location
London, UK
Okay, that's great to know. Thanks Jakob.
Do you think there is any merit in the idea that sending phantom from here would be to the detriment of the preamps? Will I definitely need to connect all the patchbay grounds or, since the signal from a mic is balanced anyway, could I avoid creating a path for the phantom to get to the preamps/console in the first place by omitting the shared ground?

Lastly, I'm thinking I might set up banks of 8, in case I ever wanted to use a mic line as a tie line (for reamping with XLR gender-benders, etc).

Another Q: I need to use 6k81 resistor pairs on every mic line, right? It's bad practice for all of them to share a single pair I guess.


Thanks again,
Dax.
 

Rob Flinn

Well-known member
Joined
Jun 3, 2004
Messages
4,501
Location
Sussex, UK
DaxLiniere said:
Quote from: Rob Flinn on October 05, 2020, 08:25:27 AM

 
I used to run a tie line directly into each mic amp to avoid running mics through the patchbay.  I selected the  mic amp by plugging into the right socket in the live room, thus eliminating that flakey connection.  I had a piece of 48 pair cable to do the job.

I used to, too, but it's not practical with a console. I truly hate patchbays and all the reliability problems they bring.

I have never had a studio without a console, & it was very practical when I was doing it. 
 

abbey road d enfer

Well-known member
Moderator
Joined
Jan 22, 2008
Messages
12,621
Location
Marcelland
DaxLiniere said:
Another Q: I need to use 6k81 resistor pairs on every mic line, right? It's bad practice for all of them to share a single pair I guess.
Think it over; if they shared the same resistors, they would all be in parallels!
 

DaxLiniere

Well-known member
Joined
Aug 10, 2004
Messages
163
Location
London, UK
abbey road d enfer said:
Think it over; if they shared the same resistors, they would all be in parallels!

Oh geeeeez... never type while asleep folks. (which is the mortified embarrassment emoji?)
 

DaxLiniere

Well-known member
Joined
Aug 10, 2004
Messages
163
Location
London, UK
I've decided I'm going to go ahead with at least trying this out.

Anyone have any suggestions where to look for a schematic of the absolute quietest 48v supply? I figure 1 amp will be ample, if you're pardon the pun. ;)

Thanks again!
 

gyraf

Well-known member
Staff member
Moderator
Joined
Jun 4, 2004
Messages
10,205
Location
Aarhus, Denmark
G9 supply works well and is very simple. Also, there are some TL783 examples around.

You don't need that quiet, if you just match your resistors :)

/Jakob E.
 

DaxLiniere

Well-known member
Joined
Aug 10, 2004
Messages
163
Location
London, UK
G9 supply works well and is very simple. Also, there are some TL783 examples around.
Cool! Any advantage of going with a 25+25v toroid instead of the voltage tripler? I always figured 'you don't get something for nothing' and that voltage doublers/triplers would probably have some disadvantage since you don't see them all the time.

You don't need that quiet, if you just match your resistors :)
Ahh, gotcha. Interesting. Let the mic amp's CMRR take care of anything. I assumed that if the mic circuitry itself received unstable/noisy power it would affect the entire preamp state (inside the mic).

TIP121 datasheet says Collector Current is 5A DC (continuous) but Collector Cutoff Current is 0.5A. Have I read the datasheet correctly that this could provide 5A? I'll only need 1A max, but 0.5A is probably cutting it a bit fine.

I'm comfortable building either TIP121- or TL783-based design and would love to know if there is a quality difference between the two.

Thanks again!
 

Latest posts

Top