Mic frequency response graphs ?

All response graphs I’ve seen are based on a 0dB centre line upon which is drawn the relative gain as the frequency increases - e.g.  


That’s pretty self-explanatory.

I’m now looking at the graph in the data sheet for another mic (actually Thomann’s SM58 clone), and I don’t understand it.

What is the y-axis scale and why are there two lines in the graph?




Can anyone help explain the second graph to me please ?

Cheers.
Lack of planning on your part does not constitute an emergency on mine.
Also chips are "Plant-based" no matter how you cook them.
0reaction image LOL 0reaction image Wow! 0reaction image Wisdom

Comments

  • octatonicoctatonic Frets: 33783
    From the look of the curves it will be in response to two different input sources at different distances.
    The blue line shows a bass lift, which is usually a result of proximity effect, ie a source closer to the mic.
    0reaction image LOL 0reaction image Wow! 0reaction image Wisdom
  • Emp_FabEmp_Fab Frets: 24263
    I guessed that might have been the case from looking at other mic graphs.  However, that y-axis is still beyond me...  it goes from 50 to 100 and is apparently in dB.  The axis is labelled “FR Magnitude dB re 20uPa/2.83V (smoothed 1/5 Oct)”  I’m guessing that it’s plotting the output of the mic with a “20uPa/2.83V” signal input - whatever the hell that is.

    I’m trying to create a graphic EQ shape to counter the peaks and troughs, i.e. to end up with a flat a response as possible, so I suppose I could just take an arbitrary horizontal line through the graph as a centre point, as, ultimately, the absolute values will change with overall gain levels but the relative gain levels across the spectrum will remain fixed.

    Have I got that right ?
    Lack of planning on your part does not constitute an emergency on mine.
    Also chips are "Plant-based" no matter how you cook them.
    0reaction image LOL 0reaction image Wow! 0reaction image Wisdom
  • octatonicoctatonic Frets: 33783
    edited May 2018
    Could be- usually the only numbers I setup that high is SPL but that graph makes no sense if it is.
    I'll just have to say I don't know, but it seems to me to be a pretty atypical way of expressing a frequency response.
    You could email Thomann and ask them.

    I can't be bothered to do that maths but you could use the formula dB = 10*log10(Amp) to convert from magnitude to db and repoint the data points to see if it gets close to what you would normally expect.
    Or just buy an SM58 and save yourself all that tedious calculation.

    Better yet buy a Beta 58.
    0reaction image LOL 0reaction image Wow! 0reaction image Wisdom
  • Emp_FabEmp_Fab Frets: 24263
    I emailed Thomann four days ago.  No response yet.

    I’ve done some digging and found out that uPa is micropascals - the Pascal being the S.I. unit of sound pressure.  As you say, the scale still doesn’t make sense as the dB values are way too high, unless I’m missing something.  According to that graph, the mic outputs between 75dB and 85dB - which seems nuts.  Obviously it isn’t and they’re not “dB” in the sense I’m thinking of, but therein lies the problem.

    The graph itself doesn’t look crazy if it were on a regular +/- dB plot, so I’m inclined to think the y-axis scale is effectively the same if used for relative values.
    Lack of planning on your part does not constitute an emergency on mine.
    Also chips are "Plant-based" no matter how you cook them.
    0reaction image LOL 0reaction image Wow! 0reaction image Wisdom
  • octatonicoctatonic Frets: 33783
    Unhelpfully Thomann's data sheet for some of their mics is even less useful.

    https://images.static-thomann.de/pics/atg/atgdata/document/specs/c_datasheet_141194_v2_en_online.pdf

    You don't have anything marked on the X or Y axes.
    I would just assume it is meant to be a rough descriptor and relative.
    0reaction image LOL 0reaction image Wow! 0reaction image Wisdom
  • Danny1969Danny1969 Frets: 10398
    The thing is dB on it's own is meaningless .... it's a ratio but you gotta know from what. Even voltage level or gain expressed in dB while assumed to be referenced to 1V is often (and better) expressed as dBm so you know what the gain or voltage level will be when driving a load ..... like a mic into the low Z input of a mixer ...dBm being 775mV into 600 Ohms

    For vocals I wouldn't want a flat response, that little peak in the upper mid is quite useful and you don't want anything below 100Hz or above 14K or so for live vocals

    A good cheap 58 clone is the Behringer XM8500 ...... when I was running rehearsals at 2020 people were breaking the 58's so we packed em away and kitted the rooms out with XM8500's and they did the job fine at £20 a pop from CPC


    www.2020studios.co.uk 
    0reaction image LOL 0reaction image Wow! 0reaction image Wisdom
  • Emp_FabEmp_Fab Frets: 24263
    I’m not wanting the flat response for vocals, I’m just playing with a graphic that has an auto feature that plays white/pink noise through the PA and uses a reference mic to detect, and correct any extreme discolouration in the spectrum.  I could buy a proper reference condenser mic but I’m not willing to splash out coin on a bit of toy-fiddling.  You can program the Graphic with the response curve of whatever you’re using as a reference mic, so the system can compensate.  I’m just trying to use my Thomann ‘58 beta’ as a reference mic, hence the need for an understandable graph.
    Lack of planning on your part does not constitute an emergency on mine.
    Also chips are "Plant-based" no matter how you cook them.
    0reaction image LOL 0reaction image Wow! 0reaction image Wisdom
  • Danny1969Danny1969 Frets: 10398
    edited May 2018
    Ah yeah I know what you mean, I used to use a DB Driverack and reference mic, Used to get some strange looks when we blasted white noise in the venue for 6 minutes ...... only used it for 2 gigs

    A dynamic mic can't move fast enough or linearly enough to be used as a ref mic regardless of EQ  .... most problems with live sound are to do with bass peaks and troughs due buildup in corner and the adding \ subtraction of waves  ..... dynamic mic moving coil is to0 heavy to be moved at low frequencies unless the source is right on top of it 
    www.2020studios.co.uk 
    0reaction image LOL 0reaction image Wow! 0reaction image Wisdom
  • Emp_FabEmp_Fab Frets: 24263
    Well to be honest, I tried it at last week's rehearsals without compensating for the mic and the sound from the PA after I'd applied the resulting EQ curve was amazing!

    Now that either tells you how good my auto-EQ gear is or how truly shit the sound was before! (I'll give you a clue... it's the latter!).

    When it comes to the skillful operation of a 31 band graphic, I'm the equivalent of Joey Essex trying to operate an Enigma machine.  If I can get an automatic system to get me somewhere close... ANYWHERE near, then I'll take it!

    I've got a little condenser mic I could use instead... Or one of the drummer's AKG overheads.
    Lack of planning on your part does not constitute an emergency on mine.
    Also chips are "Plant-based" no matter how you cook them.
    0reaction image LOL 0reaction image Wow! 0reaction image Wisdom
  • FunkfingersFunkfingers Frets: 14412
    Emp_Fab said:
    why are there two lines in the second graph?
    The black portion in the second graph might be the response curve when the proximity effect is occurring? 

    The other difference is that, in the Shure graph, the y-axis readings are Relative whereas, in the Thomann graph, they are Absolute - and not expressed in units of much help to the typical end user.

    The legend on the Thomann graph appears to read 20uPa/2.83V. If this is a reference or datum point, it would have been a good idea to represent it on the graph as a dotted line or something.

    You say, atom bomb. I say, tin of corned beef.
    0reaction image LOL 0reaction image Wow! 0reaction image Wisdom
Sign In or Register to comment.