Talk:DI unit

Nice article. I've added a high quality picture of a passive DI box designed by Cyril Jones. In line with neutral tone, I won't grumble about the inclusion of the Behringer unit whatever I think of their designs. --DavidP 21:10, 6 July 2006 (UTC)
 * Like them or no, their low cost and relatively competent design (and/or thievery of competent design, if that's your preferred word for it) has made Behringer a de facto standard, at least for when "I need it quick, cheap & dirty" applies. If you want something better, you generally already know that you want it. -- Jasper Janssen 15:56, 6 November 2007 (UTC)

Reason for using DI boxes
Maybe it would be useful to explain the reason for using a DI box intead of just plugging the instrument into the "line in" input on the mixing console, which is far easier. I do this always and never have had problems with the signals, no hum, no nothing. I've never needed a DI box.--200.14.108.1 19:19, 18 July 2007 (UTC)
 * Something like a keyboard &c will generally make something close enough to a line signal with low impedance to use directly, but a guitar pickup, for example, will have very high impedance and I think fairly high levels. If your mixer isn't designed for it, things can go Horribly Wrong. Typically it seems you'll find the line inputs on the input strip labeled "Inst" or "inst/line" if they can take a direct instrument input. If you want to inject a high-level low-impedance (speaker and/or headphone) signal, you can quite easily send a regular line input into clipping (most mixers operate their opamps at +- 12-18V powersupply or less, which limits the headroom for high-level in considerably) or even damage it in extreme cases, and a clipped output signal will kill your tweeters at any volume above minimal. In addition guitars (and most other instruments) put out unbalanced signals and for long leads back to the mixing board you greatly prefer a balanced run for resistance to RF interference. -- Jasper Janssen 15:56, 6 November 2007 (UTC)


 * I've found using a passive DI box between a guitar and a mixer mic or line in can help in getting the levels in the right kind of range (and I suspect the freq response is flatter/brighter, but that might just be me...), but usually if I'm going direct I'll have the guitar going into some effect or other anyhow, so I'd just plug that output to line in. Where I have found one indispensable is between a Blofeld synth box and the mixer (with ground lifted), without it there are really high noise levels - can only guess why, but it does the trick. —Preceding unsigned comment added by Danja (talk • contribs) 19:33, 8 September 2010 (UTC)


 * I have added a paragraph or two under "Purpose" to explain the audio purpose and introduce the electrical purpose. Russell Cameron Thomas (talk) 21:29, 17 April 2021 (UTC)

Impedance matching
They don't do impedance matching, which would make source and load impedances equal. An active DI box does buffering.

A passive DI box uses a transformer to make the impedances like each other, but doesn't make them equal. I don't know what to call this. I called it "impedance transformation" for now. — Omegatron 22:32, 8 December 2006 (UTC)


 * The term you are probably looking for is "impedance bridging". I've edited the article to reflect this. --Miken2005 00:16, 24 March 2007 (UTC)

Hi. I'm not much of an expert, so hesitate to edit the page, but this article in Sound On Sound seems to contradict the article, re: impedance. "...they function as impedance-matching devices ... The simplest type of DI box comprises little more than an impedance matching transformer with a balanced output." -- Tremolo (talk) 15:58, 16 April 2008 (UTC)

Traditional passive DI boxes very much do provide impedance matching (although rarely will the impedances be absolutely equal). Take a guitar pickup as source with an impedance of 10kΩ and a mixer input designed for a 100Ω mic. Big mismatch, right? But connect the guitar to a suitable DI box and the source you now have is 100Ω, i.e. matched to the load. Active DIs muddy the water a little, they are often more like buffers providing bridging, having a very high impedance and relatively low output impedance and hence working by optimising voltage transfer without consideration for power transfer. I have edited the article to reflect this (without going into detail). Danja (talk) 19:21, 8 September 2010 (UTC)


 * Wrong. The impedance of "a mixer input designed for a 100 ohm mic" is not 100 ohms. A mixer input designed for a 100 ohm mic will have an input impedance of at least 10 times that. Thus, your example DI that converts the guitar to 100 ohms is not performing impedance matching because it's being connected to a 1-2k ohm mic input. --Miken2005 (talk) 06:01, 10 February 2011 (UTC)

Unit Selection section
This reads like a "how to". It should be completely rewritten for encyclopedic tone. I could see two sentences covering the whole section. Binksternet (talk) 14:10, 22 March 2008 (UTC) I noticed this as well and tried to re-write it after reading it for the firs time. Maybe it'll read a little better now, but if not then whatever suggestions are out there, i'll give it a shot. Stang99gtv8 14:25, 24 March 2008 (UTC)

A little on history
The article on Sgt. Pepper's Lonely Hearts Club Band seems to indicate that DI was invented by Paul McCartney while figuring out interesting things to do with his bass. So 2 questions: 1. Any verifiable proof of this? 2. If so, how to work into article? Lockesdonkey (talk)


 * Agreed the history is interesting, but it may not be trivial to source reliably. I've come across Joe Meek as another innovator in the use of DI. --195.137.93.171 (talk) 20:39, 16 June 2013 (UTC)

Were the Beatles notable for being one of the first bands to DI their bass? If so, that's quite interesting and would be worth mentioning. But Adele? Dave Matthews? Doesn't pretty much every record nowadays use this standard practice? — Preceding unsigned comment added by 94.205.244.154 (talk) 18:58, 15 October 2015 (UTC)


 * According to this (which is cites Mark Lewisohn's book of the Beatles recording sessions) McCartney seems to have got one of the engineers to build a DI box and Ken Townshend is quoted as saying "“I think direct injection was probably used on Beatles sessions for the first time anywhere in the world”. However the article on the Wolfbox says it was invented in the 1950s. Richerman    (talk) 10:21, 10 April 2016 (UTC)

Please correct the contradictions
First, in my experience, the term "DI box" is used to refer to several different kinds of devices, including instrument preamps. While I admit that's a misuse of the term, it's so common it should be addressed.

Second, the opening paragraph says a DI box is intended to convert from line outputs to mic inputs, but then gives as an example an electric guitar. Passive electric guitars are not line level devices.

There's already enough confusion about this subject! Please let's not add more. —Preceding unsigned comment added by 69.134.228.199 (talk) 15:54, 6 September 2008 (UTC)

Yes, please correct the following:"is a device typically used in recording studios to connect a high-impedance, ''line level' (…)" — Preceding unsigned comment added by 94.246.22.195 (talk) 19:06, 29 March 2012 (UTC)

Too Technical
I am fairly certain that nobody is going to understand the introduction to this article unless A.) they're an engineer or electrician or B.) they already know what a DI unit is. Please translate from techie to English. It is possible.

Much appreciated, and let me know if I can help.--Atlantictire (talk) 14:56, 8 September 2010 (UTC)


 * I just reviewed the intro as it existed prior to this comment; I'm no EE; I would have been an English major had I matriculated, but, gosh, I don't know that one could put it any simpler. The current version is a bit better but, fortunately, not any less "technical." I believe the meta-comment at the top of the actual article saying it's "too technical" should be removed.

Dogmo1001 (talk) 01:20, 13 February 2011 (UTC)


 * I agree that the intro is just about as simple as it can get for this subject. Removing the tag. Binksternet (talk) 07:23, 13 February 2011 (UTC)

Removed sections
I removed a couple of sections here from the end of the article:

Vibrational recorder
A high-voltage turntable cartridge can be used with a direct box to directly record the vibrations from the cartridge's needle to a suitable recording device. This is an inexpensive way to create 'field' recordings of vibrations, such as the substrate-borne vibrational communication of insects. This can be accomplished with a cartridge hooked up to a tripod for stability and a DI box for amplification, which can then be hooked up to a recording device or headphones. By setting the needle on leaves or small branches or grass blades, any insect motion or vibrational communication can be heard through the amplified transduced signal (although this may require trying multiple different blades of grass, branches, etc.). It's an obscure use of a DI box (for amplification?) that really has no relevance here, although on its own it sounds interesting.

Valve (tube) amplifiers
Valve (tube) amplifiers need to 'see' a loudspeaker connected to their outputs to prevent damage, and should not be used with DI boxes unless a pass-through link to a loudspeaker is in place, or the DI box is designed for use with these amplifiers.

Completely unclear. First of all, I think it's a pretty strange claim about preventing damage and there are no references to support it; second and more important, what does a DI box have to do with this? Is this a warning against connecting the speakers through the DI box? Why would anyone want to do that?

Sergivs-en (talk) 12:49, 30 December 2011 (UTC)


 * The phono needle application is too unusual to be mentioned in the article.
 * The danger of running a tube amplifier without a load is real—the amplifier components can be damaged or destroyed. Thus, replacing the loudspeaker with a DI unit is not recommended. Instead, the DI should be connected in parallel to the loudspeaker load so that both receive signal. DI units that are designed for this purpose generally have a 'pad' (resistor) that can be switched in to attenuate hot audio such as loudspeaker-level signal. Binksternet (talk) 15:56, 30 December 2011 (UTC)
 * Do you have any references or an explanation of this risk specifically for tube amps? I'm an electrical engineer by education and I have a bit of experience with circuit design. While I haven't dealt much with tubes, I find this questionable. My guess is that it may be a problem for a specific circuit, if it's not an urban legend. On the other hand, you'd probably want to have the load connected, anyway, at least to get the amp to operate under normal conditions and have the same sound. BTW, a proper attenuator consists of at least two resistors. You want either that or a high-ratio transformer. But is this a typical usage of a DI box, anyway? You mention that there are units designed for this purpose, can you give a specific example, by any chance?Sergivs-en (talk) 09:02, 3 January 2012 (UTC)


 * The need arises when the sound of the tube amplifier is desired to be mixed live or recorded. The most common instance is electric guitar amps on stage and in the studio. Often, the artist will wish to capture the sound of the tube amplifier working hard into a load, to get the characteristic sound of the amplifier circuitry. The voltages involved are significantly higher than microphone or line level audio signal... up to 100 volts or so. The DI unit is connected to the tube amplifier in parallel with the load (the speaker or a dummy load or both). DI units designed for this duty come with a resistor pad that can be switched into the input of the DI to lower the voltage it is exposed to. Examples are: Radial ProD2 (15dB pad), Whirlwind Director (30dB pad), Radial JDI (15dB pad), Countryman Type 10 (15 and 30dB pads), Klark Teknik DI100 (30dB pad). Hope that helps! Binksternet (talk) 09:47, 3 January 2012 (UTC)
 * In my (limited) experience, a guitar amp is usually miked, because the sound depends a lot on the speaker and the cabinet. An attenuated input doesn't necessarily mean it's designed for an amp output, but at least one of your examples (Whirlwind Director, with a 30-dB attenuator) does mention an amp output. I'd say it's rather unusual. A couple of others don't. By the way, your 100 V example is pretty extreme. And the attenuation factor needed to bring 100 V down to the instrument level (is this the standard for the non-attenuated DI box input?) would be about 60 dB, so even the Whirlwind won't help you there. In any case, I don't think that the danger of damage to tube amps should be mentioned in the article, especially without a good reference, as this is not a typical use of a DI box to begin with. Sergivs-en (talk) 11:02, 4 January 2012 (UTC)

A few Diagrams
This article would benefit from a few simple block diagrams showing how DI boxes would be connected in common applications. — Preceding unsigned comment added by 74.95.43.249 (talk) 22:59, 2 June 2015 (UTC)

Direct-injection tracking??
The section Usage Examples mentions "direct-injection tracking", but doesn't define what that is, nor how it is evident that the songs listed use it. The article doesn't use that "tracking" terminology at all. — Preceding unsigned comment added by 98.67.190.147 (talk) 06:03, 5 July 2015 (UTC)

Calculation in History seems completely wrong?
If i look at the formula for transformers (for instance http://www.vias.org/matsch_capmag/matsch_caps_magnetics_chap6_04_04.html ) then the calculation in the History section of this article doesn't make any sense to mr. Not even close. Maybe if someone with proper knowledge could have a look? With my calculations the input impedance, assuming 1500 ohm load, would be 666.66 ohms. If i'm correct it would mean most of the History paragraph is fiction... 82.72.139.164 (talk) 09:32, 14 May 2018 (UTC)