Ribbon cables in Patchbays?

[Murdock]

Hey folks,

I'm about to wire my console to a patchbay and need someone to calm down my worries...
I have some really nice Lemo Patchbays that I wanted to use but now I'm unsure if they would degrade my signals.
They are 32 channels Lemo to XLR. The thing that I'm unsure of is the way they are interconnected. It is via ca. 60cm long unshielded ribbon cables...
They are wired so that every channel is next to a shield terminal. So it is shield, hot, cold, shield, hot, cold, shield and so on.
I think they came from a professional radio station or studio so they should be high quality. But I've read some different opinions about ribbon cables concerning crosstalk and lack of shielding...
Can anybody tell me that it is okay to use them withouth worry to do harm to my precious signals?
I would probably use this one for my balanced inserts. So it should be line level with an output impedance of around 30 Ohms and input impedance of about 5k Ohms.

Here are some pictures of one patchbay:

 

[Bo Deadly]

Actually I would be a little concerned. That is undoubtedly supposed to be inside a metal enclosure like a console or totally enclosed metal rack or an encloser that can be racked. If they're in a metal enclosure and the cables are not folded back on each other in an unfortunate way, then cross talk should not be bad. But I would test it. More specifically, I would decide how the panels are going to be arranged such as adjacent on the same face or front to back or what, find a suitable metal enclosure that does what you want, mount the panels and then either get new cables that are the correct length so that they're not running back and forth on top of each other or, if those IDC connectors are reusable and they come apart nicely and you have a vise to reset them at a shorter length then maybe shorten the cables you have. Then run hot signals through and check for crosstalk on a scope or with a USB interface and spectrum analysis software.

 

[abbey road d enfer]

Can anybody tell me that it is okay to use them withouth worry to do harm to my precious signals?

It all boils down to the actual impedance of the connection and the nominal level.
The connection impedance is generally dominated by the source, since good practice says use bridging, which means the source Z is at least 10x lower that the receiving Z.
Typically a line source is about 50-100r and an input is 10-20k, so the resulting Z is that of the source.
Beware that some domestic sources (CD player...) have a quite higher source Z, up to a few kiloohms.
Microphone connections have a typical Z of 150-300 r, but due to their sensitivity, they are much proner to x-talk.
The way your ribbon cabls are wired, assuming 20pF/m, 60cm result in 12p for adjacent conductors, and 4pF between two adjacent hots, worst-case* would be -85dB x-talk at 20kHz for a 100 ohm connection between lines of identical nominal level.
Now, between two mic-level connections of identical sensitivity (same gain) would be 6-10dB worse.
A worst scenario is when you have two microphone connections, one with a very hot signal and the other quite low, where the X-talk figure falls by the same amount as the gain difference between channels.
Broadcast companies do their X-talk tests with one channel driven close to clipping with its gain at minimum (typically that would be with about +20dBu) and the other at max gain connected to a dummy load (200r here, 150 across the pond), which means you have to chalk off about 70dB to the X-talk figure, for a miserable -10dB @20kHz!

To sum it up, don't use your patchbay for microphones. For CD players, turntables..., make sure their outputs are low-Z. If not use a balanced line driver, which will also prevent any potential hum loop preoblem. For the rest it is absolutely fine.

*Worst-case neglects the shielding effect of the intermediate grounded conductor and possible reduction due to balanced connection.

 

[Rob Flinn]

It is or at least used to be the case that you could get shield ribbon cable. The cable has a wire mesh on one side that you ground & because of the close proximity to the wires in the cable creates a ground plane which shields the wires. I haven't used it for the connectiosn that Abbey says aren't so good, but it worked very well for between interfaces & a Console.

 

[abbey road d enfer]

It is or at least used to be the case that you could get shield ribbon cable. The cable has a wire mesh on one side that you ground & because of the close proximity to the wires in the cable creates a ground plane which shields the wires. I haven't used it for the connectiosn that Abbey says aren't so good, but it worked very well for between interfaces & a Console.

Shielded ribbon cable provides extra protection against EMI/RFI but does not do much for x-talk between conductors in the same ribbon.
However it would add X-talk reduction between adjacent looms.

 

[Rob Flinn]

Shielded ribbon cable provides extra protection against EMI/RFI but does not do much for x-talk between conductors in the same ribbon.
However it would add X-talk reduction between adjacent looms.

I hear & understand what you're saying. All I can say is in this particular case it worked very well. I was hooking up to a Harrison series 12. The whole of the back of the racks were hooked up with ribbon cable too.

The main reason we switched to the screened ribbon for these hook ups was because the owner of the studio managed to put a core cutter through the 6 x 24 pair cable we originally ran & the the ribbon cable was the only way to get him up and running for a session in time.

 

[pucho812]

if cross talk is a concern, just remember a lot of analog consoles are wired with ribbon cables under the hood

 

[abbey road d enfer]

if cross talk is a concern, just remember a lot of analog consoles are wired with ribbon cables under the hood

True, but for the main loom, good designers make sure there is no high-level signal in the conductors.
Typically they will have grounds, rails and bus.
In particular, X-talk between bus is minimized by the fact they're very low Z and carry almost no voltage.
When ribbon cables are used for interconnect between the channel and connectors, they usually take precautions of separating sensitive connections with guards and in principle make sure all connections are low-Z.

 

[Newmarket]

Check the physical condition of the cables. Depending on the cable the insulation (often PVC) can go brittle over time and cause the conductor wires to fracture.

 

[Brian Roth]

I figured I'd toss this into the discussion. I was involved with the installation of two GIANT Amek 9098 series desks ("By Rupert Neve, the Designer"), originally made in the 1990's.

The first was in 2003 and was a 72+ input inline, a 9098i. The second was in 2009 and an earlier version which was a "split" monitor desk with 48 full function modules and 48 monitor modules. I also re-installed that second desk when it was sold to a new studio in Mexico City.

Both desks had a large external patchbay system which tied to the desk via a "bazillion" ribbon cables. I am guessing each ribbon was 30' in length. My memory about those cables is a bit fuzzy, but I believe each ribbon had an overall shield beneath an outer sheath.

I've attached a couple of snapshots. The first one shows the control room in Mexico City with the patchbays in the rear wall. Gives an idea how long the ribbons had to be to route through the walls/floor. The second pic shows the ribbons sitting on the floor of the tracking room as I pulled them all out of the control room here in Kansas before packing them up for shipment to Mexico City.

All I can say is this was the Amek factory design for both desks which originally sold for well into the six figures (US$) and both desks seemed to work quite well. I also attached a sales brochure for the 9098i but it doesn't mention how long the ribbons were....perhaps an option ordered at purchase.

Bri

 

[Bo Deadly]

Both desks had a large external patchbay system which tied to the desk via a "bazillion" ribbon cables. I am guessing each ribbon was 30' in length. My memory about those cables is a bit fuzzy, but I believe each ribbon had an overall shield beneath an outer sheath.

Wow. I would not have thought that that would work well.

 

[Jeff Goodman2]

To the OP, Belden used to make twisted pair ribbon cable. Shielded and not shielded. I apologise I do not know the reference numbers or if they still make it. But I do have a small amount at home and when I get back I will try to remember to post a pic and also see if there is a ref number on it.

All the old (post SoundWorkshop) Otari analog consoles used ribbon cable across the bottom/back of the console for signal routing. Not the best example, but also not very noisy consoles.

Best, Jeff

 

[FIX]

I've been using ribbon cable for patchbay connections for about 20 years, I use the DA-88 pinout, and of the 30 some consoles I have sold, I have NEVER had any issues with cross talk, hum pickup, noise, ect. I even use them for busses. Of course, 100% of my lines are balanced, in and out, and the CMRR is very good. All sources are no higher than 75 ohms, and they are true balanced, even the busses. One console has about 35 feet of ribbons going to the patchbay, with ZERO problems. He mixes in immersive for many Universal, Sony and other clients. You can turn the CR volume up all the way and you hear nothing. Of course, all outputs are +4, including the busses.

So, if it's done right, it can be done.

 

[Bo Deadly]

At this point I think CAT6 would be better for a variety of reasons.

 

[Brian Roth]

To the OP, Belden used to make twisted pair ribbon cable. Shielded and not shielded. I apologise I do not know the reference numbers or if they still make it. But I do have a small amount at home and when I get back I will try to remember to post a pic and also see if there is a ref number on it.

All the old (post SoundWorkshop) Otari analog consoles used ribbon cable across the bottom/back of the console for signal routing. Not the best example, but also not very noisy consoles.

Best, Jeff

Now that you mention it, I recall seeing some ribbon cable that had twisted pairs, but every foot (?) or so it would revert back to "flat" so that a standard IDC connector could be crimped onto the cable.

I do recall the Otari desks with a bunch of ribbon cables inside. One minor-ish problem happened when an unbalanced source (like from a synth) was fed into a balanced line input. Relatively low level crosstalk from other sources could be heard.

Bri

 

[moamps]

Flat cables are widely used for wiring inside the device, for example in analog consoles that have incorporated patchbays (D&R etc.). The picture shows an example of a twisted cable that connects the input XLR connectors to the channel strip of a broadcast Soundcraft console.



The first post shows that 16 channels are connected with a flat cable that has 64 wires, which means that most likely each channel is separated from the adjacent one by one or two ground wires. This procedure contributes to the reduction of X-talk. The Tascam pinout standard for the DB25 connector is designed so that when crimping on a flat cable, the channels are separated from each other with ground wire (which is not the case with Yamaha pinouts).
In my experience, if flat cables are used outside the device that protects them from external EMI, they must be in some way shieded from power cables and all other installation cables.

 

[Newmarket]

As I've experienced it, the problem with any shielded type (or 'DIY' shield made from Aluminium Foil etc) is how to effectively terminate the screen to Chassis / GND

Twisted Pair Ribbon Cable example:
https://docs.rs-online.com/bb0a/0900766b814c078e.pdf
As pointed out it has flats at intervals to get an IDC connector on it.

Round Shielded type
https://docs.rs-online.com/4e2b/0900766b8002cb77.pdf

 

[Murdock]

Thanks alot for all the valuable Information!
People got mixed experiences as it seems...
But as my whole Console is balanced and most of my other gear also I‘ll give it a try.

Concerning the twistet pairs ribbon cable. I think I can not use it as the pin out is so that each third wire is a shield wire...

And concerning the rounded ribbon cable. Wouldn‘t it be more prone to crosstalk as all the wires are closer together than in a normal ribbon cable where only the two adjecent wires are close?

 

[Newmarket]

Thanks alot for all the valuable Information!
People got mixed experiences as it seems...

Well it's about the whole system (like most things) as outlined in Abbey's and Paul's posts above.

But as my whole Console is balanced and most of my other gear also I‘ll give it a try.

Concerning the twistet pairs ribbon cable. I think I can not use it as the pin out is so that each third wire is a shield wire...

Yes. It's not easily compatible with normal wiring conventions. You need to start from the point of view of the cable really.

And concerning the rounded ribbon cable. Wouldn‘t it be more prone to crosstalk as all the wires are closer together than in a normal ribbon cable where only the two adjecent wires are close?

I think you have a point there. And there's no twisted pairs to reduce X-Talk with a balanced signal (with equal and opposite signals in addition to impedance balanced). But you get better overall screening from external sources - screen termination permitting.

 

[FIX]

Rounded cable is only for hard drives!!! One tip, don't stack your ribbons neatly, the work best when messy...