Jump to content

Talk:Decibel

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

dBd vs dBi for antennas

[edit]

In the page it is stated that 1 dBd equals 2.15 dBi. Unfortunately, this is only true in free space. At ground levels, depending on the antenna height, this can be 6-7 dBi. 5.168.129.245 (talk) 16:49, 25 March 2022 (UTC)[reply]

Clarify in lede how decibel is strictly a relative value but may in practice may be used in context of fixed reference

[edit]

Frustrating thing with decibel is how it strictly represents a relative value but in common usage or in specific practices is simultaneously used to represent an "absolute" value relative to a reference that may or may not be explicitly stated. I'm sorry I got carried away with making edits that got reverted. I'll just put my attempt here in talk...hopefully someone can figure out something better than the self-contradictory way it is written there currently:

"The decibel (symbol: dB) is a relative unit of measurement equal to one tenth of a bel (B) that expresses the ratio of two values of a power or root-power quantity on a logarithmic scale. In practice, the decibel may be expressed in the context of a fixed reference value, in which case its symbol should be suffixed with letter codes to indicate the reference value (for examples see § Suffixes and reference values)." Em3rgent0rdr (talk) 15:55, 27 July 2022 (UTC)[reply]

No need to apologise. Your edits were perfectly legitimate, and in the spirit of BRD. My concern about your text proposal is the use of the word "should", when (IMO) "should not" is more appropriate. How about this
"The decibel (symbol: dB) is a relative unit of measurement equal to one tenth of a bel (B) that is used to express the ratio of two values of a power or root-power quantity on a logarithmic scale. In practice, the decibel may be expressed in the context of a fixed reference value, in which case its symbol is sometimes suffixed with letter codes to indicate the reference value (for examples see § Suffixes and reference values), though the practice of using such a suffix is deprecated by standards bodies."
Dondervogel 2 (talk) 20:53, 27 July 2022 (UTC)[reply]
ah yes good point that I should not use "should" when talking about something deprecated by standards bodies. Yes that variant looks good. Em3rgent0rdr (talk) 21:06, 27 July 2022 (UTC)[reply]
Great! Thank you for listening. Dondervogel 2 (talk) 21:47, 27 July 2022 (UTC)[reply]
There is a strong analogy with informal usage where people say "kilo" when actually meaning "kilogram". The decibel (or bel) as a prefix can be seen as an exponential variant of the scaling function kilo (or deci etc.) in the following sense. If is the preceding number, k (kilo) transforms to whereas (bel) transforms to and (hence) transforms to . In fact, the simplest definition of the decibel is as a function defined for any real by . Boute (talk) 13:27, 16 July 2023 (UTC)[reply]

Article issues and classification

[edit]
The article has not qualified for B-class since tagged "citation needed" in 2008 (also February 2021). There is also a "clarification needed" tag dated March 2018. This is a technical article with a lot of formulas that need sourcing. Reassess to C-class. -- Otr500 (talk) 00:53, 1 March 2023 (UTC)[reply]

Typo?

[edit]

Is that a typo? "1 dB = 0.115 13… Np = 0.115 13…." I didn't want to just delete things for no reasons but it just read a bit weird. 24.49.238.103 (talk) 16:55, 7 February 2024 (UTC)[reply]

  • The statement looks correct to me. By "correct" I mean it is the inevitable consequence of combining the relationship 20 dB = ln(10) Np, with the definition 1 Np = 1.
  • One can still question these two equalities though. It seems undesirable (to me) to link the decibel with the neper in this way, but that's what the standard says. Are there other definitions (of the decibel) we could cite that would make more sense to you?
Dondervogel 2 (talk) 18:32, 7 February 2024 (UTC)[reply]
Hi, Dondervogel, nice to see you're still active on this topic. Once a year I check for progress on a few Wikipedia pages. You are right in suggesting that the current standard inhibits useful generalizations. Recall the idea of simply defining the decibel as a function defined for any real by ? This then just expresses a number. Adding the unit (see the microvolt issue raised under the next header) completes the picture. This still leaves the freedom of using the factor x/20 (depending on the unit) for backward compatibility with the power vs. root power dichotomy. Strictly speaking, it would be a good idea to gradually deprecate that convention, because it is a known source of mistakes (even in NIST documents) and in view of the choice versus leaves resistance in limbo. For more information, see [1]. I'll sign off now for about a year, but remain reachable by email. Boute (talk) 15:23, 2 March 2024 (UTC)[reply]

dBμ

[edit]

In radio receivers, dBμ is commonly used dB relative to 1μV at the specified impedance. (75 or 300 ohms for TV and FM, 50 ohms for some other radios.) I don't even know if μ is allowed in article titles. This page could discuss it, and other voltage relative units (with a known impedance). Gah4 (talk) 21:38, 20 February 2024 (UTC)[reply]

Good observation. It raises the following issue: if dBm implicitly refers to milliwatt and dBμ to microvolt, would it not be more reasonable to adopt the convention of explicitly writing the unit, as in dBmW and dBμV? Then one can also write dBmV and dBμW without ambiguity. All these suffixes to dB are nonstandard anyway, so one might as well do it cleany. Boute (talk) 14:41, 2 March 2024 (UTC)[reply]