Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

dB

Status
Not open for further replies.
Obviously you haven't worked in the real world of RF and microwave. All spectrum and signal analyzers are calibrated to read directly in dB. For example, displaying a signal and its spurious harmonic which might be 60dB down is impossible to see on a linear scale but is readily displayed on a dB scale.

Hi Carl,
It would also be interesting to see how these 'engineers' would present a Bode plot or FFT plot using a linear Yaxis scaling instead of the log dB scaling.
 
As a fan for dB, Please can you respond to my question of how do you treat with dB when hear of a dB value?
What if somebody tells you about a 40dB high pass filter? And what is your idea when there is no clue if the a dB is based on power or based on voltage? I think you should know it so that you could treat good enough.
Typically, unless you are dealing with RF, dB references voltage.

A 40dB filter is not a proper filter definition. Typically filters are rated for the steepness of their rolloff (such as 40dB/decade of frequency, which is typical of a 2-pole filter) and their -3dB rolloff frequency.

It helps in using dB if you remember a few simple relationships to give you a feeling of what a particular dB value means. For voltage ratios 3dB is approximately a factor of √2, 10dB is a factor of 3.16, 20dB is a factor of 10, 40dB is a factor of 100, 60dB is a factor of 1000, etc. The difference between positive and negative dBs is that positive dBs are gains, and negative dBs are attenuation. Thus +20dB is a voltage gain of 10 and -20dB is a voltage attenuation of 1/10.
 
Hi Carl,
It would also be interesting to see how these 'engineers' would present a Bode plot or FFT plot using a linear Yaxis scaling instead of the log dB scaling.

That's an easy one Eric - they use a fresh roll of newsprint (like the presses use) for each plot. :p
 
...
Obviously you haven't worked in the real world of RF and microwave. All spectrum and signal analyzers are calibrated to read directly in dB. For example, displaying a signal and its spurious harmonic which might be 60dB down is impossible to see on a linear scale but is readily displayed on a dB scale.

You are correct, I don't use microwave spectrum analysers.

But I wasn't expressing annoyance at dB used in microwave spectrum analysers. I was experessing annoyance in all those other "low-ish gain" scenarios where there is a simple amp or filter and the gain is something quite usable as a linear gain in the range 0.1 to 10, and they have chosen to express that gain in dB which is "idiotic" as it is counterproductive for the majority of tasks needed like viewing/measuing Vout:Vin, or calculating resistors that set the gain R1:R2 etc.

...
So people who use dBs are idiots, eh? I resemble that.

Of course I'm aware that you're not an idiot, the word was used in context re the specific "what ticks me off" scenario I outlined above! And I'm sure you know that was in no way directed at you (or any other forum member).

If you were describing a simple op-amp circuit with R1:R2 resistors that set its gain to 2.7 I'm sure you would intelligently describe its gain as "2.7", rather than "idiotically" express its gain in dB. Unfortunately some people do chose the worng option, probably as a bad habit.

...
But I have more company than you and Externet.

These days the number of new engineers reaching for a microcontroller or ADC system to work with gain compared to the number of new engineers reaching for a logarithm to work with gain might surprise you!

ericgibbs said:
...
It would also be interesting to see how these 'engineers' would present a Bode plot or FFT plot using a linear Yaxis scaling instead of the log dB scaling.

Did anyone say to stop all charting on log scales? :eek: To me the issue is about misusing dB, ie using a dB value when it would have been much superior to use a linear value.
 
Last edited:
Did anyone say to stop all charting on log scales? :eek: To me the issue is about misusing dB, ie using a dB value when it would have been much superior to use a linear value.
hi Roman,
My argument would be that someone who did not understand 'dB's would most likely not understand a simple Bode or FFT plot.

Most practising engineers understand dB usage as its derived from the ratio's and so can easily mentally transpose the two meanings when required.
 
RB, here's another dB usage you can rail at. All audio amplifier and speaker response accuracies are given in dB.

And it isn't just microwave or RF analyzers that display dB. Audio and other low frequency response analyzers also display in dB. There's more out there than just oscilloscopes and voltmeters.
 
Last edited:
There's more out there than just oscilloscopes and voltmeters.
And some voltmeters, particularly ones with an analogue meter, have db scales.

JimB
 
Status
Not open for further replies.

Latest threads

New Articles From Microcontroller Tips

Back
Top