DI boxes. How exactly they work??

[Alexxxander]

Hi
So DI boxes accept any music instruments and converts them into MIC level signal that will be used in the mixing desk as u would use any mics?? Or they can output line level signal that can be injected in the line level inputs on the mix desk?

But most keyboards are line level, so DI boxes r not used for them, right?

Or they only accept instrument level signals??
Any guitars with any pups and preamps?

Pls clarify that shit, thnx guys.

 

[rob aylestone]

DI boxes are simply converters - practically all are designed to enable multi level signals to be reduced and fed to a balanced microphone input. So the vast majority will allow from a single coil guitar pickup to a +4dB pro level line input to be converted to a low impedance, low level balanced audio signal - from a higher level, probably unbalanced hugher impedance starting point. At the simplest it's simply a balanced to unbalanced transformer - with probably a passive resitive attenuator able to be switched in. Clever ones will nclude active circuitry to make the design more elegant with perhaps a greater range of connectivity and the best signal to noise ratio - important as the design throws away signal, which has to be made up somewhere in the chain. They offer impedance conversion and most importantly isolation from the in to out. No direct connection between the input and output, and even the ground/earth connection can be broken.

Keyboards are instrument level - so usually a little lower than the traditional line level - and DI boxes will do a great job of converting that level, unbalanced to mic level, balanced. Neat gadgets - ranging in price quite a bit, but most people don't really hear much difference in the sound. I have dirt cheap ones and some quite expensive ones - if I need one, I rarely worry if the first one I find is the cheapest.

A few can even bring a speaker output down to mic level. They usually also dissipate the amp power as heat, as it has to go somewhere.

 

[ashcat_lt]

A few can even bring a speaker output down to mic level. They usually also dissipate the amp power as heat, as it has to go somewhere.

They’re generally not expected to dissip that power. You would wire the DI parallel to an actual load (speaker etc). The impedance of the DI is so high that almost no current flows through its path. This is really the main reason for the -40db pad on most passive DIs.

 

[Alexxxander]

DI boxes are simply converters - practically all are designed to enable multi level signals to be reduced and fed to a balanced microphone input. So the vast majority will allow from a single coil guitar pickup to a +4dB pro level line input to be converted to a low impedance, low level balanced audio signal - from a higher level, probably unbalanced hugher impedance starting point. At the simplest it's simply a balanced to unbalanced transformer - with probably a passive resitive attenuator able to be switched in. Clever ones will nclude active circuitry to make the design more elegant with perhaps a greater range of connectivity and the best signal to noise ratio - important as the design throws away signal, which has to be made up somewhere in the chain. They offer impedance conversion and most importantly isolation from the in to out. No direct connection between the input and output, and even the ground/earth connection can be broken.

Keyboards are instrument level - so usually a little lower than the traditional line level - and DI boxes will do a great job of converting that level, unbalanced to mic level, balanced. Neat gadgets - ranging in price quite a bit, but most people don't really hear much difference in the sound. I have dirt cheap ones and some quite expensive ones - if I need one, I rarely worry if the first one I find is the cheapest.

A few can even bring a speaker output down to mic level. They usually also dissipate the amp power as heat, as it has to go somewhere.

Thnx bro.
So that converted signal is a weak mic level signal and should always be treated as such in a mixing desk?? Like if u were connecting any mic? Right?

 

[rob aylestone]

Yep - your mixer will treat the DI box as if it is a microphone connected. Most DI boxes differ in the switches fitted to make them match better - so you might see switches labelled 0-10-20dB. The hotter the device you connect, you just select the position that gives you the gain setting your mixer/preamp prefers to work at. The only real snag with mic inputs is that they are dealing with microVolts - so need lots of gain. Your typical keyboard and guitar will be working in milliVolts, and a few pro level devices will be Volts. The DI boxes commonly on sale can deal with all these.

The only real problem is with poor pre-amps, that have more hiss than they should have, because they're cheap, or badly designed. Reducing something from millVolt level down to microVolt level and boosting it back up with a rubbish preamp is not good. That said, most preamps nowadays are quite capable of doing this 'invisibly'.

I'm lazy - even if I have perfectly good line inputs on a mixer, if there is already a mic cable connected, I'm more likely to pull off the mic and slap on a DI box than open up the mixer box, and hunt upside down to find a jack or RC input!

 

[Alexxxander]

Yep - your mixer will treat the DI box as if it is a microphone connected. Most DI boxes differ in the switches fitted to make them match better - so you might see switches labelled 0-10-20dB. The hotter the device you connect, you just select the position that gives you the gain setting your mixer/preamp prefers to work at. The only real snag with mic inputs is that they are dealing with microVolts - so need lots of gain. Your typical keyboard and guitar will be working in milliVolts, and a few pro level devices will be Volts. The DI boxes commonly on sale can deal with all these.

The only real problem is with poor pre-amps, that have more hiss than they should have, because they're cheap, or badly designed. Reducing something from millVolt level down to microVolt level and boosting it back up with a rubbish preamp is not good. That said, most preamps nowadays are quite capable of doing this 'invisibly'.

I'm lazy - even if I have perfectly good line inputs on a mixer, if there is already a mic cable connected, I'm more likely to pull off the mic and slap on a DI box than open up the mixer box, and hunt upside down to find a jack or RC input!

Got it!
Also there’s a good info here

What is a DI? - Radial Engineering

A DI box, also called a direct box, converts the unbalanced, high impedance signal output of an instrument to a balanced low impedance mic-level signal.

www.radialeng.com

Gracies amigos

 

[keith.rogers]

Also, I'd say the differences in these devices is generally very small. I've got both a LiveWire one (GuitarCenter/MusiciansFriend brand, IIRC) and the Radial Pro that (now) costs, new, about $100 more. At least in the live situations I've used them, I'd pick either one. Maybe in a studio, or some place where folks might be rolling a grand piano over your gear the Radial is worth the extra money, but I don't know.

Livewire SPDI Passive Direct Box With Attenuation Pad | Guitar Center

Shop for the Livewire SPDI Passive Direct Box With Attenuation Pad in and receive free shipping and guaranteed lowest price.

www.guitarcenter.com

 

[Cosmic]

Can vouch for the Radials as being both very quiet and very well-made.
I use:
One of their passive 2-channels on my digital piano on the way to the pre-amp in's
One of their active 2-channels to boost the sends from my DP24SD recorder to an outboard unit.

The passive unit also excels as a leveler when sending consumer-line outs (from a cassette deck) through a FocusRite 2I2 into the PC.

C.

 

[ecc83]

The term "DI" or Direct Injection box has become a somewhat flexible one. Strictly speaking it means a device that takes the signal from a passive electric guitar and convert it to a mic level signal which is at a much lower impedance than the guitar's output and balanced and usually has the capability to break the ground connection between guitar and mic feed*.

There are however now multiple 'boxes' that have other electrical characteristics notably 1:1 transformers whose main purpose is to ground isolate signals such as between an AI and active monitors. Then, as has been mentioned, DI boxes come it two main types. Passive and active. The former simple use a transformer to effect the balance, isolation and impedance transformation and it is the transformer that limits their performance. High quality audio transformers are difficult to make, large and prone to induced hum and that makes them expensive. Those properties get even harder to achieve well as the turns ratio goes up. The highest ratio I know of in a DI is 12:1 and even that only gives an input resistance of around 200k Ohms (assuming a mic pre Z in of 1.5k) well short of the 1,000,000 Ohms, the 'magic meg' usually desired for guitars. Fortunately people are rarely aware of this shortcoming and actually quite like the tonal change the low Z produces!

The active DI box avoids this compromise with electronics (almost always an op amp) and present 1 meg or even more to the guitar. Many active boxes still use a transformer for isolation but this can be a 1:1 component and, being driven from a low Z op amp, has far better performance than a high ratio traff.

Another variant is the Load Box which does absorb the speaker power from an amplifier and delivers a (usually line level) balanced signal. Often this is combined into a Power Soak device that allows variable power to a speaker.

Another drawback to the actives is that you cannot use a pair 'back to back' for the old 'Control room/studio amp trick whereby the guitar goes into a passive DI in the Ctrl room, feeds a mic tie line and a second passive DI boosts the signal back to guitar level to feed a crazily loud amp in the studio. Many guitarists/studios lay claim to have invented this idea.

DI boxes (of all types) can also be useful to counteract RFinterference, this can rarely 'get across a traff's windings and the better ones with an inter-winding shield are super RFI stoppers.

*It is often said that DI boxes give a safety isolation between say on stage kit and a remote PA desk. Don't you believe it!
First off, ground SIGNAL isolation is only present IF the earth lift switch is used and any spotty ASM yoof could mess with that. More importantly though is the fact that the transformer may NOT have 2kV isolation but even if it did the EL switch and other components are rarely rated at peak mains voltage. RCDs on mains feeds are MUCH more reliable!

Dave.

 

[Alexxxander]

Guys, do Di boxes have the ability to convert signals in reversed fashion- from mic level to instrument or line level signal?? I’ve heard that passives Di can do it ( that the signal can flow in both directions inside of these boxes) right? Big thanks

Dave, is it what u call “ back to back”??

 

[Serendipity Records]

Thank you @Alexxxander for bringing this up, I've been looking for answers to the same questions!

 

[ecc83]

Guys, do Di boxes have the ability to convert signals in reversed fashion- from mic level to instrument or line level signal?? I’ve heard that passives Di can do it ( that the signal can flow in both directions inside of these boxes) right? Big thanks

Dave, is it what u call “ back to back”??

Yes, passive DI boxes (intended for guitar) use a transformer. You can find information about "transformer operation" on the web I am sure, indeed I shall try to find some for you.

The principle is the same as a mains (aka "power") transformer in say an amplifier, it is a laminated iron core wound with many turns of copper wire and the principle is simply that any voltage applied to a winding is "transformed" to another voltage in proportion to the ratio of the two windings. Thus a transformer with 1000 turns on one winding will deliver the same voltage from a second winding of 1000 turns. That is a ratio of 'One to One' written 1:1.

A DI transformer might likely have a ratio of 10:1 and so take the signal from a guitar down from a few tens of millivolts, mV to a mV or so. If you inject a few mV into the XLR, mic input you will get a few tens of mV out of the secondary jack side.
Thus, feed line level into a DI's jack and send the XLR signal off several tens of mtrs to another (ideally identical) DI mic input and your line level is almost restored at the remote end. But, you might say, "why bother?" Well because the mic level signal is transformed down in voltage but up in current and is thus a "low impedance signal" and such signals are very robust and resistant to interference and HF loss. It is also by definition balanced.

Because V is transformed down and I (current) up there is an impedance relationship in any complete transformer circuit. The 'impednace' is transformed by the square of the ratio. Take a mic input on an AI or mixer? It will usually have an input Z of about 1.5 to 2 kOhms. The transformer primary will therefore 'see' 2k x 10x10 = 200k Ohms. Rather a low impedance for a guitar but in practice it often makes little difference.

We can see another use for this impedance transformation in valve amp output transformers. A valve might need a working load of say 5000 Ohms but the speaker is 16 Ohms. Do the math and you find you need a traff with a ratio of 56:1.

Transformers are not of course 100% efficient and so where the voltage needs to be accurate, say we need 12volt pretty closely, the winding ratio is 'kludged' to allow for losses. Clearly this can only work 'one way' so back to back bodges will always lose some signal. Also, the kludge only works accurately when the transformer is loaded close to its specification. But that is a whole other can of worms!

Dave.

 

[Alexxxander]

Yes, passive DI boxes (intended for guitar) use a transformer. You can find information about "transformer operation" on the web I am sure, indeed I shall try to find some for you.

The principle is the same as a mains (aka "power") transformer in say an amplifier, it is a laminated iron core wound with many turns of copper wire and the principle is simply that any voltage applied to a winding is "transformed" to another voltage in proportion to the ratio of the two windings. Thus a transformer with 1000 turns on one winding will deliver the same voltage from a second winding of 1000 turns. That is a ratio of 'One to One' written 1:1.

A DI transformer might likely have a ratio of 10:1 and so take the signal from a guitar down from a few tens of millivolts, mV to a mV or so. If you inject a few mV into the XLR, mic input you will get a few tens of mV out of the secondary jack side.
Thus, feed line level into a DI's jack and send the XLR signal off several tens of mtrs to another (ideally identical) DI mic input and your line level is almost restored at the remote end. But, you might say, "why bother?" Well because the mic level signal is transformed down in voltage but up in current and is thus a "low impedance signal" and such signals are very robust and resistant to interference and HF loss. It is also by definition balanced.

Because V is transformed down and I (current) up there is an impedance relationship in any complete transformer circuit. The 'impednace' is transformed by the square of the ratio. Take a mic input on an AI or mixer? It will usually have an input Z of about 1.5 to 2 kOhms. The transformer primary will therefore 'see' 2k x 10x10 = 200k Ohms. Rather a low impedance for a guitar but in practice it often makes little difference.

We can see another use for this impedance transformation in valve amp output transformers. A valve might need a working load of say 5000 Ohms but the speaker is 16 Ohms. Do the math and you find you need a traff with a ratio of 56:1.

Transformers are not of course 100% efficient and so where the voltage needs to be accurate, say we need 12volt pretty closely, the winding ratio is 'kludged' to allow for losses. Clearly this can only work 'one way' so back to back bodges will always lose some signal. Also, the kludge only works accurately when the transformer is loaded close to its specification. But that is a whole other can of worms!

Dave.

Almost restored? Will always lose some signal? Is it a high frequency content loss? What is the best way to convert instrument level to mic level and back to instrument level?
Thnxx

 

[ecc83]

Almost restored? Will always lose some signal? Is it a high frequency content loss? What is the best way to convert instrument level to mic level and back to instrument level?
Thnxx

Why do you want to do that? If you want to send 'instrument level' from say an AI or pre amp a very long way, say in excess of 10mtrs you just need a 1:1 LINE level balancing transformer. If the other end is not a balanced input you will need another transformer.
If you want to send line level signals over long lines with VERY little loss of fidelity, forget transformers unless you can afford VERY expensive ones* and instead use high quality active balanced driver circuits such as those made by "Spark fun". (I use two to get my telly 3mtrs to a mixer)

If it is a guitar you are sending you could use an active DI one end (no loss) and a transformer DI the other.

*And you STILL need HQ drive circuitry!

Dave.

 

[Alexxxander]

Why do you want to do that? If you want to send 'instrument level' from say an AI or pre amp a very long way, say in excess of 10mtrs you just need a 1:1 LINE level balancing transformer. If the other end is not a balanced input you will need another transformer.
If you want to send line level signals over long lines with VERY little loss of fidelity, forget transformers unless you can afford VERY expensive ones* and instead use high quality active balanced driver circuits such as those made by "Spark fun". (I use two to get my telly 3mtrs to a mixer)

If it is a guitar you are sending you could use an active DI one end (no loss) and a transformer DI the other.

*And you STILL need HQ drive circuitry!

Dave.

Thnx dave, I appreciate