Current capsule options (Aug. 2020): Edge and center-terminated

[Icantthinkofaname]

Takstar (p420) one is closer to Neumann's. The one found in B2 is not the right shape, it is closer to C800 response. The one found in Townsend Labs Sphere is closer in shape to the original, but still bright.

Using brighter capsule but of the same FR shape could be a good idea. Use more of deemphasis filter, you get to the same target frequency response but you get less noise than the original

Ah, so was the right 797 capsule a different center terminated one? Or am I misremembering there being a correct shape 797 one? I haven't seen an actual C800G measurement, just the chart from the manual, so for cardioid, is the difference that there's less of a mid boost (it looks like 5K on the C800G drops back down to 0 dB) and the high boost starts isn't as broad?

 

[kingkorg]

@Icantthinkofaname
Blue - CY002 the one found in TL Sphere
Green - the one found in B2
Purple - Takstar

 

[Icantthinkofaname]

@Icantthinkofaname
Blue - CY002 the one found in TL Sphere
Green - the one found in B2
Purple - Takstar

Oh wow that's actually really, really close to the published frequency response of the C800G. Doesn't look like it's much brighter at all.

 

[tomek4226]

@Icantthinkofaname
Blue - CY002 the one found in TL Sphere
Green - the one found in B2
Purple - Takstar

@kingkorg can you describe how you get so pretty "absolute responses" ?
I found your post where you instructed to divide the microphone response by the flat microphone response.
I have a sonarworks microphone with a calibration file. I took measurements of several microphones, the same distance of the microphone diaphragm from the speaker. after dividing I get bad looking graphs, what am I doing wrong? (all 1/24 smoothing)

BLUE - sonarworks
RED - b2 pro
GREEN - china m5500
ORANGE - sct800



measurements after dividing by sonarworks microphone:
BROWN - b2 pro /sonarworks
PURPLE - china m5500 /sonarworks
BLUE - sct800 /sonarworks

 

[kingkorg]

That should be good enough to get the idea. Can't go into details on my measurement setup, sorry.

 

[tomek4226]

I spent some time and today my graphs look better, still far from perfect but more and more real information can be read from them.