Film-Tech Forum ARCHIVE: Balanced outputs to unbalanced amps

This is topic Balanced outputs to unbalanced amps in forum Film Handlers' Forum at Film-Tech Forum ARCHIVE.

To visit this topic, use this URL:
https://ft-forum.com/ft/cgi-bin/ubb/ultimatebb.cgi?ubb=get_topic;f=1;t=009141

Posted by Sebastian Binz (Member # 4567) on 02-25-2008, 06:37 PM:

Since the search feature didn't bring up sufficient results, I would like to ask this question here again:

The outputs of our processor (JSD-80) are balanced, but our amps (Dynacord PAA series) only have unbalanced inputs.

What is the best way to wire this type of setup?

thank you for your comments

[ 02-26-2008, 05:53 AM: Message edited by: Sebastian Binz ]

Posted by Brad Miller (Member # 2) on 02-25-2008, 06:49 PM:

Download this - page 29

Posted by Steve Guttag (Member # 268) on 02-25-2008, 08:30 PM:

In order to know how to best wire it up, one needs to know the output design of the USL device...is it true balanced or "quasi-balanced." A true balanced output can have one leg tied to reference and merely shift the entire signal above reference...just like a transformer output.

Quasi-balanced (all too common by midiocre designers) merely provide an inverting opamp to create the "-" terminal from the plus terminal.

If you connect a true balanced output to an unbalanced input, your signal level should remain constant. If you neglect to connect the "-" terminal of the output, the signal level should drop considerably.

If the output is quasi-balanced, then connecting the "-" terminal to the low/E of the unbalanced intput shorts it out, in effect. You are better off just ignoring the "-" on the output in this case and treat the output of the quasi-balanced output as nothing more than an unbalanced output with an extra terminal. It will play 6dB lower than if you had a balanced input but should work otherwise.

Ideally, one should place something like a high-quality audio transformer in line to effectively make the inputs of the amps balanced. You will get the maximum signal transfer with the maximum noise isolation. The only potential downside is if low quality transformers are used you will color the sound some.

Steve

Posted by Sebastian Binz (Member # 4567) on 02-26-2008, 05:59 AM:

Thanks for your replies,
maybe Clint Koch can answer the question of the real or quasi-balanced outputs?
A tech I talked to just recommended to connect + to tip, - to ring and leave the shield floating on the amp's end. Is there a chance to damage anything by doint this?

*I take the phone joker*

Posted by Steve Guttag (Member # 268) on 02-26-2008, 08:29 AM:

Again, unless you know what the output design is, the advice you received was, at best, questionable. If the USL output is quasi-balanced, then you are shorting out one opamp and not providing a proper reference the + side of things so you will likely get oscillation and other unpredictable results that will vary over time.

Steve

Posted by Sebastian Binz (Member # 4567) on 02-26-2008, 05:02 PM:

I actually found something in the XD10P's manual:

In the drawing F208 on page 75 of the manual the drawing recommends to connect + to + and - in connection with the shield on the amp's end to "com". (whatever is meant by "com"?)

Posted by Marin Zorica (Member # 1716) on 02-26-2008, 05:36 PM:

quote: Sebastian Binz
A tech I talked to just recommended to connect to tip, - to ring and leave the shield floating on the amp's end. Is there a chance to damage anything by doint this?

That's correct, just do that way, you won't have any problems!

Posted by Clint Koch (Member # 1179) on 02-27-2008, 11:40 AM:

The outputs are balanced, not quasi balanced.

Posted by Nathan Tarpley (Member # 4476) on 02-27-2008, 11:54 AM:

Taking this question a step further...

Lets assume you have a processor with true electronically balanced outputs, such as the DFP3000, and all of your amps have balanced inputs. How do you best interface a monitor that only has unbalanced inputs into this system? Ideally you would like to maintain the balanced interface between the processor and amps, but wouldn't grounding the "-" terminals effectively unbalance them?

Nathan

Posted by Steve Guttag (Member # 268) on 02-27-2008, 12:12 PM:

Yes, you can not introduce an unbalanced monitor in this situation. Your choices are to replace the monitor (much preferred) or to add transformers (or suitable active circuitry) to the inputs of all of the channels of the monitor, or use a distribution amplifier on the various channels to feed each independantly.

Steve

Posted by Nathan Tarpley (Member # 4476) on 02-27-2008, 02:39 PM:

What about leaving the "-" unconnected on the monitor input and only using the "+" and the shield?

Nathan

Posted by Steve Guttag (Member # 268) on 02-27-2008, 10:00 PM:

That won't necessarily work...it all depends on the output design. If the output truely emulates a transformer type balance, then connnecting only one terminal will not provide a complete circuit, tieing the shield or no.

Odds are, you will get SOME signal flow as the gound is in there, somewhere and if it is "balanced" then it is likely in the middle. But truely, hooking balanced out to unbalanced in should just be plain avoided.

Steve

Posted by Sebastian Binz (Member # 4567) on 02-29-2008, 05:16 PM:

Thanks for your help guys, the system is running with great performance.
The only issue we are experiencing (which is more annoying than an actual issue) is sound from the RS232 connection between a desktop PC and the processor. It only appears upon hitting the "connect" button in the software until you tell the program to disconnect. You can literally hear the bits and bytes.
We have tried several approaches to this issue, including different wire routing, unplugging the non/sync inputs which are coming from the same pc, and additional grounding of both the processor and the pc, with zero results.

Posted by Sam D. Chavez (Member # 1841) on 02-29-2008, 06:53 PM:

Begging the question, but is the desktop always connected to the JSD-80? Is this only for setup, or for alternate content video?

I have seen laptops inject noise when plugged into the wall to keep the battery charged while running a DVD. Once running on batteries alone the noise stopped. You might try an experiment with a UPS to provide isolation from the power plug.

Posted by John Hawkinson (Member # 1135) on 03-01-2008, 12:18 AM:

Also, remember that all peripherals attached to the computer can inject noise. I recently had a ground loop from a video projector plugged into the VGA out of a laptop... unplug the VGA, hum went away.

--jhawk

Posted by Sebastian Binz (Member # 4567) on 03-01-2008, 03:15 AM:

Yes, the desktop is always plugged in. We are using it to run a powerpoint pre-show through non/sync (analog). The RS232 cable is also connected at all times, which doesn't cause any trouble.
If there was a ground loop somewhere, the noise should be audible at all times, shouldn't it? Instead, it only appears as soon as the software connects to the processor and quits, when I disconnect.

The noise is not the usual ground humming, you can hear beeps of different frequencies, depending on the amount of data transmitted. When the connection is idling you can hear a pattern that some describe as a "techno rythm", which is especially annoying when taking RTA/SPL measurements.

Regarding the ground loops of computers, I can confirm that a laptop can cause a serious hum through the audio output of my laptop at home, but only as long as the power supply is plugged in. But that is a different setup and doesn's seem to be the problem here.

Posted by Richard May (Member # 2627) on 03-02-2008, 12:31 PM:

That is the same sound I get when I connect to the proceesor with the software. Goes away when program disconnects.