Connecting line level instrument to a mixer

[gogi]

For a 25 foot long cable carrying a line level instrument (keyboard L/mono output or electric guitar ); what are the advantages and disadvantages of connecting to a mixer via a conventional 1/4 " mono jack cable terminated with an appropriately wired XLR or via a passive DI box (I have the Behringer Ultra 400P and Live Wire SPDI in my inventory currently)?

I always assumed that it is preferred to use mic level signals into a mixer but ran into a keyboardist who insisted on using his 1/4" mono jack cable that was terminated with an XLR connector.

PS: I have read the interesting conversations recently on DI boxes but want to get your opinion on my specific situation.

[pk.roberts]

There are a number of issues to watch out for, but the first two I would consider are; (1) is the XLR input on the mixer able to accommodate line level signals? If not, distortion might be an issue? Depending on the specific mixer, each channel may have an additional jack input intended for line level. (2) Depending on whether the mixer has a 'global' phantom power switch (and that it it on) and how the jack to XLR lead is wired, you might be putting 48 volts across the output of the keyboard which may make it very unhappy.
There are other issues as well, but I'd look at these two first.

[gogi]

The mixer is a Yamaha 5016 CF . Each channel input can accept either XLR or jack inputs. Yes; and the mixer has a global phantom power switch that is user selectable

[Mike Stranks]

For reasons including those outlined by pk roberts it is a BAD idea to connect a line-out of a keyboard to an XLR mic in of a mixer.

In addition to what's been said, the link will be vulnerable to picking up all sorts of electronic mush that could well be present on a stage.

You're in a tricky situation with the insistent keyboard player unless there's a remote possibility that the line out of the keyboard is not actually at line level. Remote in the extreme I would imagine. So they're probably simply WRONG. Maybe say that you won't use his lead 'cos there's a serious risk of either compromising the sound or damaging the keyboard or even the mixer.

A short (3m or less) jack lead into a DI is the answer - but preferably not that Behringer. Sorry! Orchid Electronics do a great range of DIs which are very economically priced.

[Gronow]

Apologies for being simplistic but why not plug the keyboard into one of the line level stereo pairs? It's what they are there for surely. Regarding guitars which you also mention, I've been having a look at this this week (with plenty of welcome advice from the forum) and the answer is that guitars need to go through a DI box which can be powered from the mic input it's going into. Even if you get sound from plugging straight in, the guitar will likely sound horrible. I've blown the cobwebs off my Pod2 which solves the problem nicely for me (and could also go into one of the stereo pairs). The Orchard electronics one which everyone seems to love is about £25 so a bit of a no brainer.
Happy new year!

[The Elf]

You keyboard player's insistance is misplaced. He needs a DI box. But ditch the Behringer - IME it is unlikely to be doing you any favours.

[Mike Stranks]

Apologies for being simplistic but why not plug the keyboard into one of the line level stereo pairs? It's what they are there for surely...

Yes you can... as long as the distance between keyboard and mixer is relatively short and there's not too much electronic mush on the stage. Such links will be using unbalanced lines. Sometimes you'll be OK, but sometimes you won't. Best to be cautious and go the DI route. Of course that can be stereo if you want... Orchid do a nice basic stereo/two-channel DI too.

I'd tend to use line-level inputs for things that are sitting next to the mixer - like CD/MP3 players, laptops etc. But beware of ground-loops with laptops!

[tacitus]

Most keyboard players I've worked with like to use the line out signal from their keyboard or amp into the desk, but if it's at all hot it's a total pain to get right at the mixer. I much prefer to have all the signals from the stage coming down the snake at mike level, so you can swap channels about if you want to and you can be sure of having the right signal level for the gain staging on the mixer. OK, so you lose some quality in theory, but in practice it's MUCH better than trying to cope with a line level signal into a mike input, with the gain turned right down and no adjustment possible.

I keep an ART twin-channel passive DI for this sort of work. It wasn't dear and it's heavy enough to throw at the keyboard player if all else fails ...

[ef37a]

Um? The problem seeme to boil down mostly to the high level presented to the mic input via XLR?

So get a 20dB XLR-XLR pad. No traff so no quality loss and if you are nifty about it, the keyboardist will never know!

Dave.

[Scramble]

Just to echo what's been said here... Don't plug a keyboard into a mic input (which is what an XLR input will usually be).

Either plug into a line-level input using a jack-to-jack cable, or, better still, use a DI box and plug into a mic input. (The latter 'balances' the signal, reducing noise, and changes it to mic level, enabling longer cable runs without signal loss).

Don't use Behringer DIs (at least not the cheap ones).

[Hugh Robjohns]

For a 25 foot long cable carrying a line level instrument (keyboard L/mono output or electric guitar )

Okay, first thing's first: there are a couple of very important differences between working with the output of an electronic keyboard and an electric guitar. Keyboards tend to have an active output driver stage giving a slightly higher overall level (maybe 10dB higher, just to give a ball park indication) than a passive electric guitar. But, more importantly, guitar pickups need to work into very high input impedances to sound as intended.

A keyboard would be happy with an input impedance of a few kilo-Ohms (typically 10K), but a guitar needs something much closer to 1 mega-Ohm -- and you'll often see mixers and interfaces with dedicated 'Hi-Z' inputs specifically intended for that purpose.

There is an inherent problem associated with high-impedance inputs, though, when long cables are being used. All cables have inherent capacitance between the conductors and, when combined with a very high input impedance, this capacitance forms a low-pass RC filter that removes high audio frequencies -- and it doesn't take a very long cable for this effect to become quite audible.

So while there shouldn't be an HF filtering problem when using a keyboard output feeding a conventional line input (of moderate impedance) via a 25 foot cable, there certainly would be if connecting a guitar to a hi-Z input with such a long cable.

The next issue to consider is the fact that keyboard (and guitar) outputs are usually unbalanced. Unbalanced connections are more prone to external interference than balanced connections, and the longer the cable (and lower the signal level), the worse that problem is likely to be. Whether external interference becomes a practical problem, though, depends mainly on the local environment. You might not have a problem at all in a home studio, but could have big issues on a stage with lots of flashing lights!

A secondary issue -- and often a more problematic one -- is that if the keyboard's power supply has a connection to the mains safety ground there is a good chance that a ground loop will be formed with the mixer via the cable screen, resulting in unwanted audible hums and buzzes.

So while it is perfectly possible to connect a keyboard to a mixer's line input successfully via a 25 foot unbalanced cable, there are several potential problems... and I wouldn't recommend doing it at all for an electric guitar.

...what are the advantages and disadvantages of connecting to a mixer via a conventional 1/4 " mono jack cable terminated with an appropriately wired XLR...

The XLR inputs of most mixers are intended for mic level signals. They therefore tend to have a low input impedance of around 1.5k Ohms, a limited ability to handle high signal levels (many mic inputs will clip around 0dBu), and they can provide +48V (phantom power) between the audio terminals and the cable screen.

None of these things are particularly appealing for a keyboard's active output driver circuitry, and are positively nasty for an electric guitar. The most dangerous aspect is that many mixers have globally switched phantom power, and phantom power sent into the outputs of keyboards has been known to affect sound quality quite dramatically, but could even destroy their output stages. I've personally witnessed that happening... saw the curls of rising smoke escaping and everything! The low input impedance of a mic amp also makes for a relatively difficult load for some keyboards to drive, often resulting in distortion or even early clipping.

The mic amp's limited headroom can often be addressed with the mixer channel's input pad, if present, and these sometimes also increase the input impedance to make it more line-source friendly, too. But not all desks have input pads, and they expect line level sources to be connected to the dedicated line inputs instead (usually on TRS sockets). I see Dave has suggested the use of external XLR-barrel type mic pads to address the potential level issue, but most of these are designed to present a 200 or 600 Ohm impedance, which would challenge the keyboard's output stages even more!

...or via a passive DI box (I have the Behringer Ultra 400P and Live Wire SPDI in my inventory currently)?

DI boxes -- either passive or active -- overcome all of the practical issues highlighted above. They present a high impedance to the source, they expect unbalanced instrument or line level inputs, they provide isolation from phantom power, they usually keep the source and destinations grounds separate (avoiding ground loop problems), and they deliver a balanced mic-level signal to the mixer which can be passed down hugely long cables without risk of interference of HF loss.

Active DI boxes generally have higher-impedance active inputs than passive boxes, and so tend to be better suited to electric guitar applications, but they also need power (either from internal battery or phantom power) and are usually more expensive. Passive boxes are fine for keyboard applications. A good DI box should last a lifetime, and it pays to buy a good one from the outset, rather than buy cheap and then have to buy again, and again.

I always assumed that it is preferred to use mic level signals into a mixer but ran into a keyboardist who insisted on using his 1/4" mono jack cable that was terminated with an XLR connector.

It is usually far more practical, convenient, and safer to use on-stage DI boxes to connect the keyboards (and guitars if necessary -- miking amps usually sounds better!) and run only balanced mic-level signals down the multicore to the mixer. If I was working with unknown equipment on stage, I'd ALWAYS use a DI box, no arguments against that approach would be tolerated at all! Not open to debate.

Technically, padding a line-level signal down to mic level via a DI box and then amplifying it again at the mixer will introduce some noise and distortion that might not be present with a direct line level connection. However, the line inputs of most mixers are just padded down and fed into the mic amps anyway, so you're not really losing any quality with the DI box approach (Assuming you're using a decent DI box). More importantly, the DI box solution protects the keyboard from inadvertent phantom power, unwanted ground loops, and local interference. What's not to like, and why would you do it any other way?

There are situations where connecting an unbalanced keyboard to a mixer's mic input via a specially constructed cable can work perfectly well too -- in the studio -- and the SOS shop uniquely offers bespoke 'pseudo-balanced' cables for this purpose. I use some in my own studio... but they are only appropriate in very specific situations. I certainly wouldn't use them in a live sound application with unknown band equipment on stage.

H

[James Perrett]

If this is an older Roland keyboard that we are talking about then it is possible that the keyboard player IS supplying close to mic level signals. They used to have a level selector switch next to the outputs with 3 settings.

[Sam Spoons]

My studio keyboard is an old Roland A90 master keyboard with a VE-RD1 piano expansion board. It has stereo, 600ohm balanced outputs (nominal output -10dBu according to the manual).

Would I be right in thinking the balanced outputs should be unaffected by phantom power (not that I intend to connect it to the mic inputs on my X32, it goes into a pair of the balanced, line level aux inputs)?

[ef37a]

Hi Hugh,
My suggestion of an XLR inline attenuator was more a ref to the OP's comment that the kbd player INSISTS on the XLR feed and the attenuator could be a sort of "slight of hand" kludge!

But you cause me pause? I had assumed such an inline devices would mimic the in and out Zs of microphone and mic amp, i.e. Zin ~1.5k Zout ~200R? (and therefore the loading on the kbd is no different from a mic input)

In order to check this I went to the Canford Audio site and found their XLR slug pt # 20-912 and just to banjax me even more they quote a blanket Z of 200 Ohms! (600R option)

Now, few microphones would be happy with such a load and actives might run into distortion I would think? I have eamiled their technical dept for clarification.

Dave.

[Hugh Robjohns]

Would I be right in thinking the balanced outputs should be unaffected by phantom power

No, that's not an assumption you can safely make. If it has transformer-coupled outputs it would be inherently phantom-safe, but active electronic outputs might be safe... or might not. And if not you're potentially into serious damage, so it's really not worth the gamble.

It might seem logical to assume that since the designers put balanced outputs on the thing they also thought of and guarded against the risks of phantom power on the connections, but in my experience many don't. They assume it will always feed a line input and put low-voltage DC blocking capacitors on the outputs... oops!

It's not a risk I'd take on an assumption. I'd want an explicit statement in writing in the manual before I'd let it anywhere near a phantom supply! If in doubt, use a DI box or -- in your case -- a proper balanced line input or line isolation transformer.

[Hugh Robjohns]

My suggestion of an XLR inline attenuator was more a ref to the OP's comment that the kbd player INSISTS on the XLR feed and the attenuator could be a sort of "slight of hand" kludge!

Yes, I understood that. I was just adding a caveat...

I had assumed such an inline devices would mimic the in and out Zs of microphone and mic amp, i.e. Zin ~1.5k Zout ~200R? (and therefore the loading on the kbd is no different from a mic input)

Ideally, yes. That's certainly the approach you'd take if calculating your own balanced L or O attenuators. But in my experience of commercial in-line pads some have presented surprisingly low input impedances -- I think possibly because they came from the 600-Ohm matched impedance interfacing equipment days! Maybe they were just the ones left in the BBC's stores!

In order to check this I went to the Canford Audio site and found their XLR slug pt # 20-912 and just to banjax me even more they quote a blanket Z of 200 Ohms! (600R option)

Good isn't it! I think what they mean is that they are designed to present a nominal 200 Ohm source impedance to the preamp, that being close to the value of most modern mics. I would hope the input impedance somewhere between 1.5 and 2.5k, replicating that of conventional mic preamps to provide a normal load to the mic. I've always found it odd that the suppliers don't provide the input and output impedance specs along with the attenuation as a matter of routine.

The 600 ohm versions are either deliberately designed to provide a higher source impedance -- which seems an odd idea -- or are more likely intended for use with the now obsolete practice of matched-impedance interfacing. I suspect the latter.

H

[Sam Spoons]

Would I be right in thinking the balanced outputs should be unaffected by phantom power

No, that's not an assumption you can safely make. If it has transformer-coupled outputs it would be inherently phantom-safe, but active electronic outputs might be safe... or might not. And if not you're potentially into serious damage, so it's really not worth the gamble.

It might seem logical to assume that since the designers put balanced outputs on the thing they also thought of and guarded against the risks of phantom power on the connections, but in my experience many don't. They assume it will always feed a line input and put low-voltage DC blocking capacitors on the outputs... oops!

It's not a risk I'd take on an assumption. I'd want an explicit statement in writing in the manual before I'd let it anywhere near a phantom supply! If in doubt, use a DI box or -- in your case -- a proper balanced line input or line isolation transformer.

Interesting Hugh, thanks. As I said I do not connect it to the mic inputs for that very reason, nor would I do so.

On live shows I always use DI boxes of all line/instrument sources.

[Tommy D]

I've just read this entire thread and found some really useful information, but I have some more questions, and wonder whether any of you can recommend something for my own situation. This is a long post so I'm sorry about that... if you want to skip to the end my questions are in bold.

I'm a freelance keyboard player and I do a wide variety of different gigs with bands at functions, festivals, music venues etc. I use one or two keyboards and sometimes a laptop/Mainstage in addition. The monitoring situation can be anything from excellent to non-existent so I always turn up with my own active wedge to be safe.

The number of channels available to me at the main mixer is also hit-and-miss so I tend to assume that I need to sub-mix my keyboards and laptop down to 1 signal to be safe. (It also makes everyone's monitor requests a lot easier as they don't have to try and specify which keyboard they want to hear!) I don't generally bother with stereo unless I'm recording or it's the sort of gig where people are really listening and the keys are very exposed.

Load in/setup/sound check/pack down time varies, but all too often it's very limited (think festivals in particular), and so I've been trying to come up with a setup which is as simple as possible and takes the least time to set up and down.

I'm keen to avoid having to use loads of separate devices and associated cables i.e. a sub-mixer, audio interface, DI box(es), so I spent some time researching audio interfaces which had multiple analogue inputs where I could also connect my keyboards. In the end I bought an Allen & Heath ZEDi-8 (http://www.allen-heath.com/ahproducts/zedi-8/) and in many ways I am happy with it.

The USB interface has a good clean sound (at least to my ears) and has a suitably low latency; it has enough channels for me to plug in my keyboards + laptop, and send a single mixed signal to the desk (I take an XLR from the Main Out R). I can also take a 1/4" jack from the headphone output (giving me L only) to my personal monitor wedge and control the level I get there. Furthermore, if I also want to hear the stage monitor mix to include some vocals etc, I can daisy chain off a stage monitor (at least when there is a link output) and pan it hard left - then I can get it in my monitor without sending it back to the main mixer.

And so until recently I thought this was the bees knees - I had made sure to choose something with balanced outputs to avoid picking up any interference on the way to the main mixer - I thought I had got away with combining an interface, sub mixer, and DI box all into 1 super-convenient unit.

Then I started reading about ground loops and wondered whether I would come unstuck due to my lack of a ground lift button.

Then I started reading about the problems of sending a line level Main Out to a mic preamp on the main mixer. In my ignorance I hadn't realised there was a difference between the XLR out of a DI box and the XLR out of my ZED. At a guess probably 90% of the mixers I plug into don't have a pad button. And I'm struggling to get my head around this impedance business so I'm not sure if this is going to ruin things or not.

To be honest, I haven't actually experienced either of these problems yet, and I've taken it to a bunch of different gigs already. I tend to set my main fader to -5 and haven't had any problems or complaints RE distortion, and I've plenty of 'downward headroom' if it needed to go lower. No ground loop issues yet either - I try and plug my power into the same place as the main mixer if it's nearby, but I've even got away with it so far when this isn't the case. Probably just a fluke and my luck will run out.

THEN I started reading about phantom power being able to cause damage to things which aren't mics which has got me quite worried. Naturally I do not want to fry my ZED on a gig if the main mixer only has global phantom, as that is bound to happen sooner or later.

So my questions are: Is there a device out there that does everything I want in 1 unit? I guess kind of like the ZED with its multiple inputs, but which is phantom-safe, outputs at mic level like a DI box and has a ground lift. I did come across the Radial USB-Pro, which seems to have the right kind of outputs, but doesn't have the additional analogue inputs for my keyboards. Those 2 products combined would be ideal.

I do own a passive DI box... Would it work to plug the ZED Main Out XLR into a DI box using an XLR-to-jack cable?

And just because it's something I want to understand: I've read that there are impedance issues going from the ZED's XLR out to a mic preamp. What would be the symptoms of such a problem? I know that plugging an electric guitar into a line input means you lose some volume and high end but that's about my limit...

Thanks for sticking with me. Surely I can't be the only keyboard player thinking about this?

[Hugh Robjohns]

Then I started reading about ground loops and wondered whether I would come unstuck due to my lack of a ground lift button.

Potentially, yes, but there's an easy fix (see below)

Then I started reading about the problems of sending a line level Main Out to a mic preamp on the main mixer. In my ignorance I hadn't realised there was a difference between the XLR out of a DI box and the XLR out of my ZED.

A decent FOH mixer will have a pad button on its mic preamp to reduce a line level signal to something it can cope with, so it shouldn't be an issue in practice. And if it doesn't, a simple 20 or 30dB in-line XLR attenuator will do the trick.

To be honest, I haven't actually experienced either of these problems yet...

It may just be that you've been lucky, but more likely that you've been working with people who know what they're doing and who have made sure it works!

Naturally I do not want to fry my ZED on a gig if the main mixer only has global phantom, as that is bound to happen sooner or later.

That's certainly the one thing that would concern me. The Zed may well be fine with phantom on its outputs, but I wouldn't want to risk it for the cost of a DI box...

Is there a device out there that does everything I want in 1 unit?

Not that I'm aware of.... But you can provide ground-loop and phantom power protection, and a mic level output with a simple passive DI box, or stick with your balanced line level output and use a passive isolation transformer unit like the ART DTI (or Cleanbox2) or Orchid Electronics' line isolator box to provide ground-loop and phantom protection.

I do own a passive DI box... Would it work to plug the ZED Main Out XLR into a DI box using an XLR-to-jack cable?

Yep.

And just because it's something I want to understand: I've read that there are impedance issues going from the ZED's XLR out to a mic preamp.

I hope you haven't read that here, because it's nonsense! The general requirement is to have a load impedance of 5-10 times more than the source impedance. Most mics have a source impedance of 150-200 ohms, and most mic preamps have input impedances of 1500-5000 ohms. So that ticks the box. The Zed will have an output impedance of less than 100 ohms, so its very unlikely to be bothered about the mic preamp's input impedance. And if you connect via a DI box anyway, it certainly won't be!

What would be the symptoms of such a problem?

Transient distortion on signal peaks and a generally reduced output level.

If I were you, I'd permanently wire in the passive DI box to the output of your Zed. That will provide total protection to the interface from phantom power, and (assuming the DI box has a ground lift switch or is already wired with a ground lift in place), total protection from ground-loop issues between your rig and the FOH mixer. It will also provide a balanced mic-level signal to the FOH, avoiding any potential overload issues with simple FOH mixers. It makes life very easy for the stage sound team, which is a useful bonus.

The other thing I'd recommend is to use a line isolation transformer box between the link out of any stage monitors and your interface, simply to avoid any direct connection of system grounds between your keyboard set up and the on-stage monitor system, and thus avoid potential ground-loops.

[alexis]

I just want to waste the pixels by saying it is threads like this which make SOS forum the bestest on the internets!!

Thank you everyone!

[shufflebeat]

Apologies if this has been mentioned and I've skimmed over it, but, is there any chance that the keyboard player is giving you a stereo line out (a la headphone socket) because that's a whole new bag of worms when fed into a balanced XLR input.

Just checking.

[Hugh Robjohns]

... is there any chance that the keyboard player is giving you a stereo line out (a la headphone socket) because that's a whole new bag of worms when fed into a balanced XLR input.

No, I don't think that situation applies here, but it is an interesting issue worth expanding upon, just for completeness because it does crop up occasionally.

So for anyone wondering about the outcome, the critical aspect is that a balanced (differential) input only passes the difference[/] between the signals received on the hot and cold connections, and it rejects anything that is the same. This is how a balanced input rejects (common mode) interference. However, if fed with an unbalanced stereo signal (eg. from a headphone output), the differential receiver will reject any centrally-panned elements (like bass, vocals, kick and snare in a conventional mix), because they are the same on both sides, and only pass the stereo difference information -- which will be things like reverbs and any widely panned elements.

So, if the source is a typical soung mix the outcome will be a rather thin-sounding and reverberant mix, lacking in vocals bass, kick, snare etc. if the source is a keyboard the result will depend on the specific keyboard sound and applied effects, but typically it will be bass-light, reverby and low in level, or possibly display a significant lack of level through the midrange compass. In the case of organ with Leslie effects, the tremolo volume modulation may appear extreme!

H

[Mike Stranks]

Just a note in favour of always using galvanic isolation either through a DI or isolator like the DTI (or in my case, mic-splitters with an isolated output) when having to connect your gear to other gear of unknown provenance.

When doing radio OBs I learnt the hard way to always have some form of isolation between any of my gear and, typically, a location's sound-system. Wanting to avoid a proliferation of mics, I'd offer to give a feed from individual mics or a sub-mix into the sound-system. Surprising how often I'd encounter Hum-Buzz City before adopting the 'always isolate' approach.

Churches were usually the worst... often a combination of inexperienced, but well-meaning volunteers and an 'interesting' installation. But some other venues could also give you a shock - both metaphorically and literally!

[Hugh Robjohns]

Just a note in favour of always using galvanic isolation either through a DI or isolator like the DTI (or in my case, mic-splitters with an isolated output) when having to connect your gear to other gear of unknown provenance. When doing radio OBs I learnt the hard way....

Re-quoted for emphasis!

H

[Dave B]

Just to throw in my 2c on this one. (Mainly for Tommy D than the OP)

I have in the past used balanced jack leads from a single keyboard to a mixer, but only because I knew that the mixer would take a balanced line input. I then had a return back to my monitor. It worked ok as a setup, but I do things differently now.

Nowadays, I have a stereo DI box which is what sends audio to FOH. Even if I'm just using the one keyboard (with balanced outs), I still use the DI as it gives isolation, a ground lift (if needed) and a pad (again if needed). So I'm pretty covered for any PA gear as long as it can take a standard pair of XLRs. For monitoring, I can use the through connections on the DI to feed a little Mackie SRM150 which sums them nicely for me to mono. If I need an additional return (vocals/rest of band), that can also be added to the SRM. At this point, I am ONLY sending what is needed. Even if I use additional gear, I will mix it with a small mixer and use the line outputs on jacks from the mixer to drive the DI box. The mixer is set to unity throughout and it just a summing unit for the inputs. My theory being that it's the cleanest signal path I can get.

DIs can be expensive, but can also be reasonably priced. I have the Palmer stereo DI and it didn't break the bank and I've had nothing but compliments for my sound. Which is nice.

There is an interesting trend for modern keyboards to give a stereo input to mix in with the main outputs. Some boards, like the new Kurzweil Forte, use a mini jack as they expect it to be for an ipod. Others, like the Korg Kronos, (sensibly) supply a full featured set of inputs on combi connectors - nice. In both cases, another unit can be mixed into the main keyboard so that only a single pair of outputs need be sent to the FOH - again, very nice solution (unless you're a Wakeman). Plus, the massive polyphony of modern synths makes it possible to use a 'dumb' keyboard to drive the main unit over midi (or usb midi) again meaning multiple 'keyboards' give a single output.