The dBu re-examined...

[Externet]

Hi fellows.
Think of this post as me walking in front of a firing squad, willing to do it.

The dBu is the industry standard for measuring audio voltages. Do not confuse dBu with the deciBel dB, nor the dBm, nor acoustical sound pressure levels.

The 0dBu reference is 0.774 Vrms, measured with no load, an odd figure carried from the past, and not even the 1 round Volt world standard.

Please be gentle, in a sane discussion, use slow rubber bullets , and convince me why :
-->the dBu should not be discarded (or life would not be easier without it) and
-->that it is not just a habit born from the days of early telephony, being carried in a new form to the 21st century with no need.
-->or give examples of audio levels and measurement where the Volt, well known and world standard would be akward or incorrect to use instead of the dBu.

If someone else wants to join me in dodging bullets, welcome.

Miguel

[dyarker]

The no load bit is wierd, so dBu was probably invented by marketing types, or engineers under pressure by marketing.

Whether above is true or not, 0.775VRMS abouit equals 1Vpeak. In order to have calculations of power at speaker terminals (the usual end point of an audio setup) work out with logrithmic dB units, the base number needs to be apparently awkward.

With voltages in dBu and gains in dB the calculations are addition/subrtaction. With voltages in volts and gain as "A", the calculations are multiplication/division. Addition/subtraction is considered easier by most folks.

(I tried to shoot with marshmellows )

Cheers,

[Chris Smith]

As long as Analog recording is still around, it will be needed as a reference point to where distortion occurs.

[Robert Reed]

Odd that DBu voltage is almost exactly DBm (1 MW/600 ohm=0.775 V) voltage. I had never heard of DBu's until a while back their was another post on this subject. I can't see any use for a standard that has no load and no mentioned source impedance.Who actually uses these things and why? The whole idea of DB ratings is to quickly calculate system (or circuit) gains or losses while at the same time reducing ridiculously large numbers to simple "in your head" arithmetic. Does the 'u' stand for unloaded?

[dyarker]

I think just coincidence that the voltage is the same as 0dBm at 600 Ohms, because the voltage of 0dBm is different at other impedances is different.
0.775 times square root of 2 equals 1 (the 0.775 is rounded up)

[Externet]

Thanks, Dale.

Let me explain to the extent of what I know the weird unloaded part: When invented, the dBm was referred as 1milliwatt into 600Ω, and that yields 0.775V as V= sqrt(1mW x 600Ω)
The 600Ω is not in use in audio equipment for at least 25 years; but the voltage reference was kept (by a reason I do not know, or because was already a familiar figure)

As audio equipment has much, much larger input impedances in modern equipment, the power delivered is irrelevant, the decision was to make the dBu an amplitude measure unit (not which amplitude yields 1 mW, but amplitude alone with no power dissipated at load -or pure Volts instead of milliwatts created by those Volts-)

You mention: "In order to have calculations of power at speaker terminals"... A (sine) audio signal of say 16Vrms into 8Ω yields 32 Watts, with no need to use the semi-obscure dBu. [I say obscure because even you in your expertise mention dBunloaded as "weird"]

You mention: "With voltages in dBu and gains in dB"... With a signal amplitude in plain Volts instead, as the gain of an amplifier is not given in decibels, is given in a factor, as Rf÷Rin having a gain of say 10. Then, if the input to such op-amp is 1.25V, the output is plainly 1.25x10= 12.5 V, with no need to convert the gain to decibels. Your multiplication against addition observation is valid; probably 60 years ago without today's calculators but I think that any electronics worker has to be able to multiply, easier than conversion to logarithms.

Anyway, within the conversion to logarithms there is multiplication and division (20 log V2/V1) plus the need for a logarithm table.

Hope I am clear expressing myself properly.


Hello Robert.

..."Who actually uses these things and why?"...
Actually the whole professional audio industry. Been into dBu daily for 14 years, and cannot understand other reason than "habit"

..." reducing ridiculously large numbers to simple "in your head" arithmetic.
The ridiculously large and smaller numbers ARE reduced with the prefixes micro, mili, Kilo, Mega, etc. Cannot see that as a reason in my opinion.
..."Does the 'u' stand for unloaded?".. Yes, unloaded or unterminated.

Thanks for shooting gently,
Miguel

[rshayes]

The "u" probably refers to "volume unit" or "VU". The volume unit was defined about 1939 as the response of a meter with specified nonlinear and transient characteristics. The nominal impedance of the telephone system was considered to be 600 ohms. For a steady sine wave signal, 1 volume unit represents 1 milliwatt in a 600 ohm system. For other levels, the meter reading will not indicate the power of the signal, due to the nonlinearity of the rectifier. Thus it does not accurately indicate dBm. For a changing signal, the meter reading will also not accurately indicate power, due to its transient characteristics, which are between peak reading and average.

Normally, a VU meter will be operated at a level close to 1 VU in a system where the impedance is 600 ohms. Other levels and impedances are accomodated by using resistive pads with known attenuation characteristics, usually expressed in dB.

[dyarker]

Like I said, forget about 0dBm at 600 Ohms, coincidence.

0.774VRMS * 1.414 = 1.094436Vpeak or 1Vpeak rounded.

Like Robert Reed, the only place I've seen dBu mentioned is here. You want to get rid it, fine, I think it's a marketing gimik anyway (confirm that, you said professional audio). Never seen dBu applied to op amps or circuit internals. You're right, that doesn't make any sense. Using dB (any letter) does make sense applied to a "box". The log conversion only needs to be done once by the guy that made the "box". It should make it easier for the person patching together mics, preamps, mixer boards, equalizers and power amps.

While you say you want to get rid of dBu, it sounds like you want to get rid of all dB types. 600 Ohms isn't the only impedance where 0dBm is valid. And audio wasn't the only place 600 Ohms was/is used.

[Externet]

Hello Dale.
..."Like Robert Reed, the only place I've seen dBu mentioned is here"...

Examples of manufacturers using dBu, choosing only sites with no .pdf just to make access to specifications easier, and none of them I worked for :

http://www.mackie.com/products/onyxseri ... specs.html

http://www.alesis.com/products/studio32/

http://www.yamaha.com/yamahavgn/CDA/Con ... TE,00.html

http://www.shure.com/mixers/models/auxpander.asp

http://www.innovason.com/pages/prod_spec2.html

http://www.audiotoys.com/pmonitor.htm

http://www.d-r.nl/dnrsite/products/octagon/octagon.html.

It truly is the standard, am not making that up. Sorry if my way to write is not refined, this is my third language, am not good at expressing myself.

Miguel

[dyarker]

Didn't say you were making it up.

[Chris Smith]

The LM 3914 was a much used led VU meter back in the 80.

Again its still used for the distortion level in audio even to this day.

Im not aware of any high end audio that still doesnt incorporate a VU meter on its front panel.

Any search of the web will show that the VU meter is far from obsolete.

Volume Units is simply a cut off point or value, where you can guarantee either reproducable quality, or distortion.

NO LOAD is not a mystery, it’s a standard.

When you have a industry like the sound business that didnt have uniform standards, [the early days] then you need a set point that all others must meet if you want uniformity with all equipment.

Microphones vary in impedance, and any sensing load on the microphone will cause distortions.

You need the “Equivalentâ€

[Robert Reed]

If DBu represents a Voltage with no reference to load or source impedance, then why not just use the term voltage? Also 600 ohm terminology is very much in use today -Just look at any microwave base band specs or telephone equipment. Remember audio covers a lot more ground than just home entertainment equipment. From what I have gathered the "600" standard was derived from the twisted pair impedance at the approximate audio frequency band. And when they say nominal-it's really NOMINAL.

Why is this post so wide?

[JohnDay]

Well, the dBu, like the dBm, dBV and so on are all defined in ANSI T1.523-2001. The dBm is a power ratio referred to 1mW in 600 ohms, for audio, or in 50 ohms usually for RF, dBu and dBV are voltage ratios. The dBu is grandfathered in on the basis that 0.775V RMS is the standard reference voltage for audio, but more importnaly telecommunications.

Because it is a voltage ratio it is not u = unterminated but u = unknown or dont care! We quite often see things specified in dBu but with a terminating impedance which is not 600 ohms.

This conversion tool may be useful:

http://www.analog.com/Analog_Root/stati ... nvert.html

John