discuss-gnuradio
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Discuss-gnuradio] to prevent mental damages, avoid dB's.


From: John Ackermann N8UR
Subject: Re: [Discuss-gnuradio] to prevent mental damages, avoid dB's.
Date: Thu, 28 Sep 2006 19:51:57 -0400
User-agent: Thunderbird 1.5.0.7 (X11/20060922)

I have to say that I like dBs, too.  All you have to do is remember two
things: 3dB doubles the power, and 10dB is 10 times the power, and from
that you can SWAG just about anything.

John
----

Berndt Josef Wulf said the following on 09/28/2006 07:26 PM:
> It don't see how this makes the calculation of RF power any easier, to the 
> contrary it confuses the issue. 
> 
> cheerio Berndt
> 
> On Thursday 28 September 2006 17:50, John Gilmore wrote:
>>> transmit power converted to dBm (1 dBm == 1 mW) minus the attenuator
>>> loss = output power in dBm.
>>>
>>> E.g.
>>>   100 mW -> 20dBm
>>>   20dBm - 15 db att = 5 dBm
>>>   5 dBm -> 3.2 mW
>> Actually, I think 0 dBm = 1 mW.
>>
>> dB's are a royal pain in the butt.  They eluded me for years because
>> they required a lot of rote memorization and made no sense.  For those
>> of us not pickled in radio-speak from an early age, but who know basic
>> algebra, there's a simple way to deal.  Ignore deciBels.  Use Bels.
>>
>> Bels are easy and obvious.  They're a straight logarithmic scale in Base
>> 10. 100 mW is 2 Bm.  10 mW is 1 Bm.  1 mW is 0 Bm.  0.1 mW is -1 Bm.
>>
>> DeciBels are just tenths of a bel.  So if you shift the decimal point
>> one place, you're suddenly calculating in an easy to use notation.
>>
>> Here's the above calculation in Bels:
>>>   100 mW -> 2 Bm
>>>   2 Bm - 1.5 B att = 0.5 Bm
>>>   0.5 Bm -> 10 to the 0.5 power -> the square root of 10 -> about 3.2 mW
>> See, now you not only know the answer, but you know WHY "5dBm" is 3.2 mW.
>>
>> Why the EE universe settled on doing everything in tenths of a
>> logarithmic unit is way beyond me.  It's as if every carpenter figured
>> every length in deciInches or decimeters, even if inches, kilometers
>> or meters would be the more straightforward unit.  How often do you
>> calculate in decivolts, deciwatts, or decimeters per second per
>> second?
>>
>> The rumor is that decibels were invented because somebody at Bell Labs
>> couldn't cope with decimal points or negative numbers, in the days when
>> equipment wasn't capable of dealing with large orders of magnitude
>> (e.g. the painful-to-someone 0.3 Bel became the friendly-to-someone 3
>> deciBel).  Of course, now that people regularly see 5 to 10 orders of
>> magnitude (5 to 10 Bels) (50 to 100 deciBels) (factors of 10000 to 10
>> billion) in ratios, such as in radar, digital signal processing, or
>> fiber optics, the "deci" has just become a hindrance.
>>
>> You can do your part to clear up this idiocy by using Bels in most
>> places where the lemmings use deciBels.  You may actually get them to
>> think (briefly).
>>
>>      John
>>
>> PS: Don't even get me started about why dBm's aren't referenced to
>> watts rather than milliwatts!  Since a "milli" is 1/1000th and that's
>> just 3 orders of magnitude, referencing to ordinary watts would merely
>> involve subtracting 3 or 30 from the number, e.g. 40 dBm = 4 Bm = 1 BW
>> = 10 dBW.  It reminds me of how we're still calculating speeds in
>> 5280-foot units per 3600-second units rather than in some sane system
>> using basic decimal units.  Actually using BW notation in your
>> thinking and writing may overload lemming brains, though.
>>
>>
>> _______________________________________________
>> Discuss-gnuradio mailing list
>> address@hidden
>> http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
> 
> 
> _______________________________________________
> Discuss-gnuradio mailing list
> address@hidden
> http://lists.gnu.org/mailman/listinfo/discuss-gnuradio
> 
> 





reply via email to

[Prev in Thread] Current Thread [Next in Thread]