Can I use digital cable for analog cable?

[gabler]

I have some Tdif snakes I was going to resolder the Dsub and put TRS 1/4
on the other end. My cable company CBI said I could use this cable but not vice versa.(analog cant be used for digital of course) What do you guys think. Will they sound just as good since it is a different capacitance.

[bob katz]

Originally Posted by gabler

I have some Tdif snakes I was going to resolder the Dsub and put TRS 1/4
on the other end. My cable company CBI said I could use this cable but not vice versa.(analog cant be used for digital of course) What do you guys think. Will they sound just as good since it is a different capacitance.

Absolutely. In fact, cable designed for AES/EBU is automatically low capacitance and likely to produce superior analog audio as well. Some AES/EBU cables are extremely stiff and hard to deal with in portable applications, but Mogami makes a nice flexible version.

BK

[Jim Williams]

I've been recommending the AES cable for many years over the "analog" versions. No one wiring a control room or studio should ever use stuff like the Mogami snake cables for audio, use the AES digital version, much lower capacitance and fog.

Jim Williams
Audio Upgrades

[InGear]

I agree with Jim Williams and Bob Katz; here's some more detail and a caution about reading cable specs carefully:

When making studio cables, if you read the specs in some catalogs, you might conclude that analog snakes have lower capacitance than digital snakes:

Mogami's spec for 2932 analog 8-pair snake lists capacitance of 3.7 pF/ft interpair "partial", and 40 pF/ft wire-to-shield "partial". (Catalog page 24)

Mogami's spec for 3162 digital 8-pair snake lists an "effective capacitance value between inner twin" of 14pF/ft, and doesn't mention a wire-to-shield value. (Catalog page 50)

So at first glance it might appear that the analog cable is *way* better: 3.7 pF/ft vs. 14 pF/ft!

But not necessarily so -- 3.7 pF/ft is an engineering calculation for JUST THE WIRE PAIR. When you add shield capacitance it gets bigger (at least to 24 pF/ft by calculation of "wire-to-shield-to-other-wire").

To resolve this for real, I tested 30 feet of Mogami 2930 analog snake against 30 feet of Mogami 3162 digital snake. Here's the results on a Fluke multitester. Not guaranteed NIST-traceable accuracy, but hopefully with decent relative accuracy:

Mogami 2930 ANALOG snake: about 34 - 35 pF/ft within a pair in 1 channel

Mogami 3162 DIGITAL snake: about 25 - 26 pF/ft within a pair in 1 channel

Of course, your mileage may vary (and it would be interesting to see this comparison reproduced on laboratory-grade equipment).

In conclusion, at least by the above measure it IS better to install digital cables even if you'll be running analog signals over them.

[SiliconAudioLab]

Did the whole studio (except for mic cables) w/ Mogami 3080 AES digital for audio and digital.

Never looked back since!

[gabler]

Thanks guys! Has anyone done a a/b blind test? I just got a new Midas Venice 320 (can any mods be done to the 320? (Jim Williams)?

[NathanEldred]

I use all large gauge Canare 110ohm 206 cable when I'm doing any analog mastering. It sounds more open to my ears on both ends of the spectrum...or the regular Canare is less open than the 206 digital stuff, however you want to look at it.

[ethan_c]

Ah, but here is a potential curve ball...

I have a D-Sub to (4) AES xlr(m)/(4) AES xlr(f). Now seeing as how these XLR's are wired for 2 channel (hence only 4 of each instead of 8) how could I use this in a DSub situation when it would be looking for 8 channels?

Anyone?

[XSergeantD]

Originally Posted by ethan_c

Ah, but here is a potential curve ball...

I have a D-Sub to (4) AES xlr(m)/(4) AES xlr(f). Now seeing as how these XLR's are wired for 2 channel (hence only 4 of each instead of 8) how could I use this in a DSub situation when it would be looking for 8 channels?

Anyone?

Most AES D-subs on equipment have I/O on the same d-sub, therefore your 4M and 4F XLRs

[BLT]

My friend and I did some listening tests between mogami 2549 (regular mic/line cable) and 3080 (110 ohm AES cable), using them both for mic and line level material. We were surprised by how different they sounded. We determined that digital cable sounded quite a bit better for line level interconnects (as many people have mentioned before, it sounds much more "open" and transparent), but for mic level, it was a matter of taste, just the same as how you'd use a different preamp or mic for a certain sound or color. I'm going to rewire my studio with all digital cable for line level interconnects, and run a few 3080 cables into my tracking room along with the 2549 so I have the choice.

My question is, before I go and buy a couple hundred feet of mogami 3162 (8-pair digital snake) to hook everything to my patch bays, is there any reason I'm not thinking of to NOT go the digital cable route? Any chance it's more susceptible to noise, hum, or interference for example? More crosstalk between channels? I was wondering because when looking at bantam patch cables, a lot of the pre-made ones use QUAD digital cable. Made me worry that my whole studio will hum because I didn't do enough research! Are the digital snakes ok or should I stick to individual 3080 cables (better shielding maybe?). Any logic to this or am I just worrying? Thanks for your help!

-Brett

[Duardo]

I have some Tdif snakes I was going to resolder the Dsub and put TRS 1/4
on the other end.

If they're actually TDIF snakes I'd make sure that the pinout will work...

I have a D-Sub to (4) AES xlr(m)/(4) AES xlr(f). Now seeing as how these XLR's are wired for 2 channel (hence only 4 of each instead of 8) how could I use this in a DSub situation when it would be looking for 8 channels?

If you're wanting to use these for analog connections, you can just use adapters or remove four of one gender and solder the other one on. Although again, I'd check the pinout...there are two different standards for wiring up AES/EBU snakes...

-Duardo

[beyarecords]

May I ask what 'Frame Rate' you guys are using for your Mogami 3080 cables? 48, 96 or 192KHz? Do the differing frame rates make a difference to the overall sound quality?

[SiliconAudioLab]

Originally Posted by beyarecords

May I ask what 'Frame Rate' you guys are using for your Mogami 3080 cables? 48, 96 or 192KHz? Do the differing frame rates make a difference to the overall sound quality?

We use a frame rate of 48K and a sample rate of 30fps.

[peeder]

Originally Posted by gabler

Has anyone done a a/b blind test?

You can't rely on a blind A/B test for something as subtle as cable...there are too many variables you can't control for.

Do nulling on generation tests (i.e. going in and out of a converter pair over and over) to see if you can spot a difference using a good set of test signals (test tones/noise, digital black, and program material). Go ahead and stick anything in the signal path you might want to test. You can apply equal gain to the nullfiles but do not normalize them (as noise is random). Do at least three complete test sequences, alternating from one cable to the other, plugging and unplugging and coiling and uncoiling the cables in order to ensure that the cables' seating and positioning is not a factor, and to be sure, do this for three separate instances of each brand/model of cable, of equal lengths, to ensure their connectors aren't corroded or something.

My bet is that with properly balanced signals and normal impedance differentials and standard levels (and in an environment that isn't pathological for audio) you will not be able to characterize any difference at all for any audio signal. If you could do so repeatably under these strict testing guidlines, everyone would post the samples and charts when selling their cables, just as they might for mics. For digital signals you will want digital cable of course.

[rodrigo]

I have read about this before here at Gearslutz. So when I was shopping for digital cable for analog interconnects, I got Mogami 3159 ( 1 Pair 110ohm AES/EBU Console Cable) instead of 3080 (2 Conductor 110ohm AES/EBU Digital Audio Cable). Is there any difference? Would I be better off using 3080?

I also have a question about telescoping shield cable for line level interconnects. As I seem to understand, I should leave the shield at the output of one device (e.g. preamp) and cut the shield at the input of the other device (e.g. audio interface). But what if I'm using a patchbay between device 1 and device 2? should I cut the shield at both connections? (preamp-patchbay, patchbay-audio interface) or just at the input on the audio interface?

On a different application, is it ok to upgrade internal wiring on pickup enabled instruments such as D6 Clavinet or Rhodes with digital cable? or is this not considered line level interconnects? Thanks!!

best regards,

ps. Jim Williams: I'll be doing a circuit upgrade to my clav, thanks for your contribution to clavinet.com!

[peeder]

Originally Posted by rodrigo

I also have a question about telescoping shield cable for line level interconnects. As I seem to understand, I should leave the shield at the output of one device (e.g. preamp) and cut the shield at the input of the other device (e.g. audio interface). But what if I'm using a patchbay between device 1 and device 2? should I cut the shield at both connections? (preamp-patchbay, patchbay-audio interface) or just at the input on the audio interface?

My understanding of telescoping shields is the opposite: the way they work is they use a standard balanced input to subtract noise from a balanced cable, even though the balanced cable is connected to an unbalanced output. If your output is balanced there is no benefit from disconnecting the common wire at the output (not on the input end as you suggest)...you will just get your signal level cut in half (if you cut the common wire--you will lose grounding if you cut the shield). But with telescoping shields, you get some of the benefits from your balanced input cancelling out line noise picked up in transmission.

If you have an unbalanced XLR output (rare, but it happens) you shouldn't need to alter your cable in any way to have the cable function as telescoping shield (although you can if you want the shield wire to be subtracted rather than the common wire). If you have an unbalanced 1/4" phone jack (tip-sleeve) output, then you use a 3-element cable (hot-common-shield) just as you would for a balanced cable, but you don't connect the common wire (some say the shield, but the shield wire is different than the hot, and in order to subtract properly I think you ought to use identical, tightly paired wires to subtract identical noise) to the unbalanced end. Simply using a balanced (TRS) 1/4" cable will work as the ring won't make contact within the output jack.

You will want to use a balanced patchbay and have all connections to the patchbay balanced. You only want the common wire disconnected at the unbalanced output plug.

[svart]

one thing that isn't mentioned here that should be is that it's NOT the cable that is necessarily causing the audio to change tone, it's the interaction between the driver and receiver circuits on either end of the cable that makes the difference. You are driving a length of wire that has impedence, resistance and capacitance and is thus reactive. Slewing into a capacitance is notorious for degrading certain qualities of both small and large signals usually by straining the drive circuit as it tries to dump current into the wire to "charge" that "cap". It's much like trying to drive a MOSFET gate or some such, it takes a lot of OOMPH.

With that being said, this is one of the reasons people blindly go for lower capacitance cable, it's easier on the driving circuit and thus gives the impression of "sounding" better.

You could improve the output driver of your source and you would get the same effect with the "lower" quality cable.

[Jim Williams]

You can improve your output driver and get better results with a great cable than a good driver and average wire.

Using Mogami digital wire in the Rhodes won't hurt, but the passive volume/tone controls suck most of the air out before you can hear it. Active Mk 1's have a 33k input impedance, the Mk 2's it's only 10k ohms, too low to hear the higher harmonics of the tines. This is why most users have the treble cranked up all the time.

Redesigning the Mk 2 input to a non inverting config with a 1 meg input impedance allows all the tones to be heard. The new pianos have this design with a discrete FET transistor input, I'm measuring -115 db noise on the new pianos.

Jim Williams
Audio Upgrades
Rhodes Music Corp.

[rodrigo]

Originally Posted by peeder

My understanding of telescoping shields is : use a standard balanced input to subtract noise from a balanced cable, even though the balanced cable is connected to an unbalanced output.

Thanks for your response, peeder; it makes perfect sense. I had a different idea based on the following recommendation stated on a different thread:

Originally Posted by David Kulka

In my opinion, the best thing you can do when wiring a studio is to telescope your grounds. The rule for that is: except for mic lines, when connecting the output of one balanced piece to the input of a balanced piece, cut the shield on the input side. (You end up with a lot of male XLR's and TRS input cables with no shield connected.) This will prevent and solve a huge number of problems and it speeds up wiring, since there's one less wire to solder.

If you do this, and don't have a lot of unbalanced stuff in your room, and still have a lot of problems, it can be cleaned up, but you'll probably likely need a tech to help out.

So based on your input, I would only use telescoping shields when wiring unbalanced gear connected to balanced gear. I'm definitely planning on a balanced patchbay, and most of my gear is balanced. My unbalanced gear: (rhodes, clavinet, nord lead, guitars and bass) won't be connected to line level inputs, but rather D.I. or miked amps. So in this case, does that mean I won't be needing telescoping shields at all? Thanks!

regards,

[rodrigo]

Originally Posted by Jim Williams

Using Mogami digital wire in the Rhodes won't hurt, but the passive volume/tone controls suck most of the air out before you can hear it. Active Mk 1's have a 33k input impedance, the Mk 2's it's only 10k ohms, too low to hear the higher harmonics of the tines. This is why most users have the treble cranked up all the time.

Redesigning the Mk 2 input to a non inverting config with a 1 meg input impedance allows all the tones to be heard. The new pianos have this design with a discrete FET transistor input, I'm measuring -115 db noise on the new pianos.

Jim: I have one of the old pianos; a Mark II stage piano with only BASS BOOST and VOLUME controls. In your opinion, is there something I could do to improve its tone response without shipping it to the US for service? I was thinking maybe an active circuit such as Creation Audio Redeemer would help. What do you think?

I have no background in electronics, but I could handle a simple installation such as the redeemer. If you have other suggestions I could hire a local guy to do the job. BTW, Kudos on your new Rhodes design, I've seen wonderful reviews and you're the man behind such a great new instrument!

best regards,

Rodrigo Montfort

[peeder]

Originally Posted by rodrigo

So based on your input, I would only use telescoping shields when wiring unbalanced gear connected to balanced gear. I'm definitely planning on a balanced patchbay, and most of my gear is balanced. My unbalanced gear: (rhodes, clavinet, nord lead, guitars and bass) won't be connected to line level inputs, but rather D.I. or miked amps. So in this case, does that mean I won't be needing telescoping shields at all? Thanks!

The quote refers to grounding while I'm referring to balancing. Those are different issues.

I highly recommend maintaining an effective ground throughout your system, and only interrupting it when you get hum issues. However, I do NOT recommend breaking it via snipping cable! That has never worked for me, anyway. Instead, I would use an isolation transformer, such as the EBTech hum eliminator. EBTech also makes a product called Hum X which doesn't get placed in the signal path at all, but into the power line. Both work very well in the only place I have any such issue, which is using two guitar amps simultaneously.

There is a theory that all equipment should have one and only one path to ground. Beyond that, that all paths to ground should have the same length (this is where "star grounding" and "balanced power" come in). I imagine there are slight benefits to electrical noise floor as a result of such measures, and with unbalanced gear, the benefits are often greater. However, going around snipping grounds out of cables can give you serious noise problems! These will be a major hassle to debug and you'll be resoldering all your cables shortly and swearing never to pay attention to those morons online again!

And come to think of it, you might not want to pay attention to me. YMMV...

[nandoanalog]

This thread proves that digital owns analog.

[Dan Lavry]

Originally Posted by InGear

I agree with Jim Williams and Bob Katz; here's some more detail and a caution about reading cable specs carefully:

When making studio cables, if you read the specs in some catalogs, you might conclude that analog snakes have lower capacitance than digital snakes:

Mogami's spec for 2932 analog 8-pair snake lists capacitance of 3.7 pF/ft interpair "partial", and 40 pF/ft wire-to-shield "partial". (Catalog page 24)

Mogami's spec for 3162 digital 8-pair snake lists an "effective capacitance value between inner twin" of 14pF/ft, and doesn't mention a wire-to-shield value. (Catalog page 50)

So at first glance it might appear that the analog cable is *way* better: 3.7 pF/ft vs. 14 pF/ft!

But not necessarily so -- 3.7 pF/ft is an engineering calculation for JUST THE WIRE PAIR. When you add shield capacitance it gets bigger (at least to 24 pF/ft by calculation of "wire-to-shield-to-other-wire").

To resolve this for real, I tested 30 feet of Mogami 2930 analog snake against 30 feet of Mogami 3162 digital snake. Here's the results on a Fluke multitester. Not guaranteed NIST-traceable accuracy, but hopefully with decent relative accuracy:

Mogami 2930 ANALOG snake: about 34 - 35 pF/ft within a pair in 1 channel

Mogami 3162 DIGITAL snake: about 25 - 26 pF/ft within a pair in 1 channel

Of course, your mileage may vary (and it would be interesting to see this comparison reproduced on laboratory-grade equipment).

In conclusion, at least by the above measure it IS better to install digital cables even if you'll be running analog signals over them.

Good post.

Let me add to it: The capacitance between the conductors pair and the capacitance between a conductor and a shield are separate issues, and should be specified and viewed separately, because they impact the signal differently in different cases.

That is so because the impact of the capacitance is there only when one applies voltage change between conductors (or conductors and shield). In the case of balanced drive, the 2 inner conductors see the full differential voltage between them. But if the conductors pair are very close to each other (say tightly twisted pair for example) relative to the distance to the shield, the combined action of the pair tends to cancel much of field as it relates to the shield, making the cable to shield capacitance less important.

But take the same cable and use it in unbalanced mode. Now the impact of capacitance between the pair can be the same, but it is higher when one views the electric field between the signal conductor and the shield.

Generally, regarding analog vs. digital cable - There is no such thing as a digital cable. All cables are analog, as all signals are analog. An analog signal can be "anything", and a digital signal is "closer" to being restricted to having one of two voltage states, say 0V and 5V (or other levels). But in reality, the transition between states is not sudden, there is some time for the signal to rise and fall, the signals may have ringing during transitions, or exponentially rising, or.... all very analog stuff. In fact the whole idea of running curent on a conductor is very analog (unless one wants to get into quatum physics, examinimg an electron at a time, which has no place here, because we are dealing with a flow of huge numbers of electrons, and the current is the avarage of all the electrons, thus continues analog not discrete behaviour).

So the concept of analog cable vs. digital cable is an outcome of marketing and sales. Perhaps such a distinct division was done to help a customer buy the right product, but the fact remains, there is no such thing as digital cable. There are cables suitable for say digital audio, but the same cables may not work for say digital video... A cable for digital audio tends to be capable of higher frequencies (more bandwidth) then a cable for signals for analog audio bandwidth. Both cables are analog cables, they both carry analog signals.

The more in depth way to view cables is to view their characteristics, and find a cable that is well suitable for a given application. That could require both mechanical and electrical considerations.

Regards
Dan Lavry

[Jim Williams]

Originally Posted by rodrigo

Jim: I have one of the old pianos; a Mark II stage piano with only BASS BOOST and VOLUME controls. In your opinion, is there something I could do to improve its tone response without shipping it to the US for service? I was thinking maybe an active circuit such as Creation Audio Redeemer would help. What do you think?

I have no background in electronics, but I could handle a simple installation such as the redeemer. If you have other suggestions I could hire a local guy to do the job. BTW, Kudos on your new Rhodes design, I've seen wonderful reviews and you're the man behind such a great new instrument!

best regards,

Rodrigo Montfort

Those retrofit preamps have some of the same dirty, noisy electronics you don't want to hear. 9 volt powered preamps are current limited and therefore will not be as quiet as a design without power limitations. A better way for recording is to take the signal off the harp at the RCA jack. Run this through the best active DI you can find. This will break the connection to the passive 10k audio volume pot that loads down the hf response of the tines. You will have a 1 meg ohm input impedance and will be able to hear all the Bob James subtle harmonics of this instrument.

Down the road I will be designing some "retrofit" preamps for the older Fender Rhodes passives. I might even do a retro design of the Mk I preamp, with all it's glorious dirt and noise! These will be "official" Rhodes products, not 3rd party designs. Stay tuned!

Jim Williams
Rhodes Music Corporation

[DONNX]

Originally Posted by BLT

My friend and I did some listening tests between mogami 2549 (regular mic/line cable) and 3080 (110 ohm AES cable), using them both for mic and line level material. We were surprised by how different they sounded. We determined that digital cable sounded quite a bit better for line level interconnects (as many people have mentioned before, it sounds much more "open" and transparent), but for mic level, it was a matter of taste, just the same as how you'd use a different preamp or mic for a certain sound or color. I'm going to rewire my studio with all digital cable for line level interconnects, and run a few 3080 cables into my tracking room along with the 2549 so I have the choice.

My question is, before I go and buy a couple hundred feet of mogami 3162 (8-pair digital snake) to hook everything to my patch bays, is there any reason I'm not thinking of to NOT go the digital cable route? Any chance it's more susceptible to noise, hum, or interference for example? More crosstalk between channels? I was wondering because when looking at bantam patch cables, a lot of the pre-made ones use QUAD digital cable. Made me worry that my whole studio will hum because I didn't do enough research! Are the digital snakes ok or should I stick to individual 3080 cables (better shielding maybe?). Any logic to this or am I just worrying? Thanks for your help!

-Brett

Can anyone answer this guy's question. By using AES/EBU cable like mogami, which I don't believe is shielded like canare aes cable, will you get more hum, efi, rfi going on?

I notice all the analog cable are shielded with mogami. But their aes cable is not.

Anyone out there got their entire studio wired with AES cable? except mic lines of course. Any issues with hum? or noise?

THanks

[DONNX]

[daniel c]

Mogami 3162 uses shielded twisted pairs with a drain wire for each core. You will be fine.

Look right down the very bottom of this link;

MOGAMI® - AES/EBU 110 Ohm Digital Audio Cables

[KristianA]

Hi all of you! If I decide to go for the Mogami AES/EBU 3080 when soldering the new linecables I need for my studio... Would there be a problem keeping the older analog-cables (Mogami 2552) I already have in the signal-chain or do I have to change them all?

Any pitfalls when soldering?

thanks for your help /Kristian

[DarkSky Media]

Originally Posted by KristianA

Hi all of you! If I decide to go for the Mogami AES/EBU 3080 when soldering the new linecables I need for my studio... Would there be a problem keeping the older analog-cables (Mogami 2552) I already have in the signal-chain or do I have to change them all?

Any pitfalls when soldering?

thanks for your help /Kristian

The only down side is that then some of your connections (the AES ones) will be of better quality (eg will pass wider bandwidth) than others.

If you can live with that, then there is no urgency to upgrade the existing cables.

[ezrecords]

I don´t recommend to use digital purposes cable instead of analog cable. In any case, everybody is free to try it.