Hello brain trust!

I'm getting myself wrapped in confusion trying to figure out the best way to integrate an unbalanced tape machine with my balanced desk and audio interface.

Here's the low-down:

I have an Antelope Orion Studio Synergy Core interface as the heart of the system. It has DC-coupled outputs (non-cross coupled). As I understand it, when integrating unbalanced connections with DC-coupled outputs, you must float the ring (see the Rane note attached, no. 10). DC-coupled outputs are somewhat unique in that you can't simply use an unbalanced TS cable, because you don't want the Ring shorting to ground, possibly because of DC..? Not sure.

Anyway, I tried this out with my tape machine by replacing some of the RCA>TS cables with a TRS on the end without the Ring connected. Sure enough, it sounded noticeably better when connected straight in and out of the Antelope interface, less stressed. From what I understand this is because the quirk of a DC-coupled (non-cross coupled) balanced output is that if it is trying to push the inverted signal/impedance to ground then it unduly stresses the output amp (putting DC on the ground...?). Leaving the Ring floating means it just doesn't try, keeping the main driving output (on the Tip) staying in it's happy place and not chucking DC all over the place.

For those unsure about this, there are lots of articles around the place, like the well known Rane article (image attached) about balanced<>unbalanced connections. Like I said, DC-coupled interfaces are rare and require this "Floating Ring" connection to get the best signal. I believe there's even a chance that sending the Ring to Ground could actually cause damage in the long run because of extra stress/heat on the output of the interface... Maybe someone could confirm that. I certainly don't want to take the risk obviously.

I have a new desk coming to the studio this week (), a Toft ATB 32. It has electronically balanced inputs and outputs.

So, for the Toft, the "correct way" to interface the unbalanced tape machine to the desk for electronically balanced cross coupled ins/outs is to wire the Ring to Ground (attached image no.6). But for the Antelope the "correct way" to do it is to wire Ground to Ground and Float the Ring.

Now, my question is:

How do I interface "Floating Ring" non-cross coupled unbalanced connections with regular electronically balanced or impedance balanced connections? Or to put it another way, referring to the attached Rane note, how do I interface cross coupled (image no. 6) and non-cross coupled (image no. 10) unbalanced to/from balanced equipment?

Anyone cleverer than me have any insights into this? I should say as well that I'll be going through a balanced patch bay. So really the question is simply, how should I wire the looms to/from the tape machine? Floating Ring? or Shorted Ring?

Thank you all very much for any assistance you can offer.

I've not floated a ring before, but have lifted a ground. +tip - ring lift ground off sleeve (not that it has one) kind of floating a sleeve, which would connect with a regular ts cable, put + to + and minus to minus on a trs you should be good to go.

Cross reference both manuals.

Back in the day I used Henry Engineering Matchbox's for this.

Thank you for your response. The issue is the dc non-cross coupled outputs require not shorting the ring to the sleeve. It all depends on the amplification topology of the device, and unfortunately basically no devices give this information.

Thank you though 👍

Yes good idea. I have some boxes like this. The problem is I need to do it for 48 channels so I’m trying to find a “cable only” solution.

Just to update the thread: the Antelope can be configured to be cross coupled or non-cross coupled on the INPUTS. Only non-cross coupled on the OUTPUTS.

So considering everything else is cross coupled I think the current best solution will be to wire inputs to the tape machine as Ring floating and the outputs from the tape machine as Ring shorting (or more trad Ring >Shield and unbalanced shield disconnected).

Will report back after I make some tests.

will not work if the tape machine inputs are transformer coupled.

Thank you, but the tape machine inputs are not transformer coupled. If they were it would save a lot of headache!

I'm essentially trying to find a solution for both interfacing directly to/from an audio interface and to/from a mixing desk. The balanced/unbalanced thing isn't the issue; only that the audio interface has DC coupled outputs which means a very esoteric way of un-balancing, but only in that use case. I'm trying to find a single solution that works in all situations.

I have a similar problem interfacing with a pair of API 550A clones, which have unbalanced inputs, as the originals did. I haven't gotten around to doing it yet, but I'm planning to drill some holes in my patch panel and install switches that let me either float the cold signal or dump it to ground, depending on whether I want the input to come from my D/A converter or a transformer-coupled preamp.

I still haven't been able to get a straight answer on whether the topology of my converter's outputs (it's a Ferrofish Pulse 16) truly require floating the cold, but it seems like at least a safe way to do it.

That's a good idea. Yes, if you have the CV version of the Ferrofish then this is the same scenario. I considered some kind of elaborate switching/routing system but I'll need it to do 48 channels, so I'm trying to find a cable situation.

I think you could potentially do it with patch leads. Unbalanced to balanced isn't an issue. For balanced out from Ferrofish to the API clone, you could have a patch lead for the patch bay with no ring connected. I think that would do trick. The important thing is to float the ring, not short it to ground. To be frank, your situation is probably a perfect situation for just chucking a couple of transformers on the inputs. Again, not really an option for me because I'd need 24 or 48 of them...!

Use 10 from desk to tape.
6 is good for a path from tape to desk. (unbal to bal)

BTW, leaving negative leg floating at output is not a special thing only in dc-coupled outs. It is a good thing to do with almost all active electronically balanced outputs, except those "electronic transformer"-style outputs, aka "Quasi-floating balanced output"... and except impedance balanced, which is why they are pretty popular (among other good features mentionedd in engineering textbooks).

You're confusing two completely different things; "DC Coupled" means a totally different thing from "Cross Coupled". A DC coupled in/out just means that it can pass DC on the ins and outs (for use in Control Voltage applications, or to extend LF response all the way down). A DC Coupled in/out can be balanced, unbalanced, cross-coupled, not cross-coupled. Since the presence of DC in an audio chain may cause headroom/distortion issues in *some* signal chains it's generally considered good practice to block DC in your I/O design, but opinions/applications vary.

"Cross-coupling" refers to an output topology which attempts to emulate the behavior of a true transformer output which responds to the termination/non-termination of one side of the balanced output. There's no obvious way to know if an output is cross-coupled or not unless: the user manual states it, you've got a schematic, or you can actually test. True cross-coupled outputs are less common so the usual way to connect an electronically-balanced output to an unbalanced input is to leave one side floating, but if the output is truly cross-coupled that unused side (minus) should be terminated for correct operation. It would be nice if manufacturers actually specified this information in their product info/user manuals but they rarely do anymore. In non-cross-coupled balanced outputs most designers will add output resistors to current limit the output driver just in case it's shorted to ground, but you never know... If I had to *guess* I'd guess that an output is not cross-coupled, but again, that's a guess without actual documentation/testing.

And as an aside; if an output *is* a real transformer out then the ring/minus side should most definitely be terminated to ground when driving an unbalanced input.

In short, no way to really know without confirmation from the manufacturer that outputs are/are-not balanced via a true cross-coupled topology - and again, this has nothing to do with whether or not the outputs are DC coupled. Bit of a mess, but if you want to do it right you need to know this info.