It's not as though, for the purposes of DIY, one is shooting for creditation by the National Physics Laboratory or whatever............
Displaying dBu accurately is a doddle. The calibration of microphones so as to to measure SPL isn't, but for DIY purposes all that matters is one isn't wildly out. If one has a scientific or commercial purpose, for example making claims for products, that is another matter.
Displaying dBu accurately is a doddle. The calibration of microphones so as to to measure SPL isn't, but for DIY purposes all that matters is one isn't wildly out. If one has a scientific or commercial purpose, for example making claims for products, that is another matter.
depends what you mean by wildly!
Other than the lucky people with speakers that can do 120dB peaks most of us (assuming things don't start sounding nasty) would potentially be pushing it to get to 105dB peaks. At this level 3dB error in the SPL measurement could be the difference between all ok and damaged drivers.
This all of course assuming you are correct and that realistic SPLs are a key part of fidelity
Other than the lucky people with speakers that can do 120dB peaks most of us (assuming things don't start sounding nasty) would potentially be pushing it to get to 105dB peaks. At this level 3dB error in the SPL measurement could be the difference between all ok and damaged drivers.
This all of course assuming you are correct and that realistic SPLs are a key part of fidelity
It's all because of Ed
Long Term Stability of Condenser Microphones
http://www.bksv.com/doc/TechnicalReview1969-2.pdf
See the artificial ageing procedure (exposure to increased temperature)
Condenser calibration using a pistonphone
http://www.bksv.com/doc/TechnicalReview1964-4.pdf
This is how Ed dismantles his pistonphone
http://exodus.poly.edu/~kurt/manuals/manuals/Other/BRUEL%20&%20KJAER%204220%20Instruction.pdf
George
Long Term Stability of Condenser Microphones
http://www.bksv.com/doc/TechnicalReview1969-2.pdf
See the artificial ageing procedure (exposure to increased temperature)
Condenser calibration using a pistonphone
http://www.bksv.com/doc/TechnicalReview1964-4.pdf
This is how Ed dismantles his pistonphone
http://exodus.poly.edu/~kurt/manuals/manuals/Other/BRUEL%20&%20KJAER%204220%20Instruction.pdf
George
Last edited:
Great stuff, thanks as always George.It's all because of Ed
Long Term Stability of Condenser Microphones
http://www.bksv.com/doc/TechnicalReview1969-2.pdf
See the artificial ageing procedure (exposure to increased temperature)
Condenser calibration using a pistonphone
http://www.bksv.com/doc/TechnicalReview1964-4.pdf
This is how Ed dismantles his pistonphone
http://exodus.poly.edu/~kurt/manuals/manuals/Other/BRUEL%20&%20KJAER%204220%20Instruction.pdf
George
John Eargle once told me that the ultimate limitation of a condenser microphone was the thermomechanical noise of the diaphragm. I wonder if B&K (or somebody) ever determined its magnitude? Most studies I've seen (for example in Van der Ziel, Noise) just consider the Brownian motion of the atmosphere in the vicinity.
I wasn't terribly pleased to hear Eargle's assessment, as I thought there might be room at the bottom (as Feynman used to say) for a still-quieter preamp. Of course in real recording situations the room noise will dominate.
Yes, thermomechanical noise is a fundamental limitation of condenser microphones, but it is usually minimum with simple, omni models. It comes from the damping added which comes out resistive as noise.
However, some condenser mike preamps are extra noisy because they do not use 1G or higher resistors on the input.
However, some condenser mike preamps are extra noisy because they do not use 1G or higher resistors on the input.
Brad you are welcome.
Re ‘thermomechanical’ noise.
Does a capacitor possess an inherent noise generator?
If the pressure equalisation vent is inside the sound field, I can’t see how the noise measured at the mic terminals is affected by anything else than air molecule dynamics.
I can only see an environmental noise amplification (not self noise generation) due to mic diaphragm tension and mic equivalent volume.
I have some reservations of possible noise generated by leakage resistance of the capsule.
Sections 6.6, 6.9 and 6.12
http://www.bksv.com/doc/be0089.pdf
Then see measured inherent noise diagrams on Figures 2.18, 3.16, 4.16, 5.16, 6.16, 7.16
http://www.bksv.com/doc/be1373.pdf
The mic noise dominates at mid and high frequencies.
The noise peaks at frequencies quite lower than the diaphragm resonance frequency. Strange
George
Actually, all dielectrics are lossy and have a sort of thermal noise, although it is usually so drastically smaller than other circuit components that it is rarely considered.
Where it does become significant is in cooled charge preamplifiers, most frequently used with cooled particle detectors. Radeka has discussed this at length in nuclear science journals.
I have told the story of the group, Landis et al. at Lawrence Berkeley, who decided that the feedback resistor noise in a charge preamp could be eliminated by shining light on the input JFET to reset the feedback capacitor, and to facilitate this they removed the transistor from the case, which was sealed with borosilicate glass, a decent match for the thermal coefficient of expansion of the kovar.
They got a dramatic reduction of noise, and attributed this to the absence of the feedback resistor.
However as was so often the case with Bell Labs, a pair there doubted that the reduction was due just to this. Kern and Mackenzie put some JFETs into lower-loss packages (I think they were a couple of materials, BeO and BN, and maybe alumina), and made conventional preamps with feedback resistors (around a couple Teraohms iirc). They got similar noise performance.
Having said that, I don't think Eargle was talking about dielectric losses, but other thermal noise, perhaps the noise associated with the damping mentioned by John C.
There is also DA loss from the sheet resistance of the diaphragm. One needs a measurement in vacuum and air to separate the effects. B&K also has a paper on their -4dBA 1" capsule with a discussion specific to mechanical sources.
Last edited:
This may help, see p. 2-35 and 3-11. http://www.bksv.com/doc/be1447.pdf
If not, it's bound to be in here somewhere. Primers & Handbooks - Brüel & Kjær
Thank you Brad, Scott, Mr. Curl,
A big Thanks to Ray for the link. Yes, pages 2-25 to 2-41 (zoom into 2-35 to 2-37)
Scott, is this the article (P16-22) ?
http://www.bksv.com/doc/bv0015.pdf
See also this (P3-23)
http://www.bksv.com/doc/TechnicalReview1977-3.pdf
George
A big Thanks to Ray for the link. Yes, pages 2-25 to 2-41 (zoom into 2-35 to 2-37)
Scott, is this the article (P16-22) ?
http://www.bksv.com/doc/bv0015.pdf
See also this (P3-23)
http://www.bksv.com/doc/TechnicalReview1977-3.pdf
George
Attachments
George,
You made me have to look up the word psophometric, now I know a new word. I wondered why the graphs were diverging so greatly. You can learn something new every day it seems. Thanks for the reference. I also now see that the larger diaphragms actually have higher inherent S/N ratios, I would have thought the opposite.
You made me have to look up the word psophometric, now I know a new word. I wondered why the graphs were diverging so greatly. You can learn something new every day it seems. Thanks for the reference. I also now see that the larger diaphragms actually have higher inherent S/N ratios, I would have thought the opposite.
Hiten - the reciprocity you imagine does exist and excellent (dynamic) speakers can make excellent microphones! I have forgotten the paper that shows this.
However, also there is a difference because large diaphragms are not needed to detect LF signals, you will find that very small capsule microphones (e.g. DPA4060) can have excellent LF response to sub sonic frequencies. It is only omni (pressure) mics that have this response. Directional (pressure gradient) capsules have a falling Lf response. There is a great difference between detecting the LF pressure and regenerating it. The previously linked to Neumann and B&K docs show the physics of this.
However, also there is a difference because large diaphragms are not needed to detect LF signals, you will find that very small capsule microphones (e.g. DPA4060) can have excellent LF response to sub sonic frequencies. It is only omni (pressure) mics that have this response. Directional (pressure gradient) capsules have a falling Lf response. There is a great difference between detecting the LF pressure and regenerating it. The previously linked to Neumann and B&K docs show the physics of this.
Last edited:
The diameter of the mic diaphragm basically only alters the polar response and lobing of the polar response at higher frequencies, and smaller diameter diaphragms push the lobing to a higher frequency.
Smaller diaphragm mics also have lower SNR... There is no free lunch!
I find the Neumann doc by Bore and Peus the clearest on this.
Smaller diaphragm mics also have lower SNR... There is no free lunch!
I find the Neumann doc by Bore and Peus the clearest on this.
I had to look that up myself (*)
The root is the verb ‘psopheo’ meaning “to produce noise”, the noun ‘psophos’ meaning “noise”, so “psophometric” means a process (here a measurement) that uses noise as the stimulus
Greek Word Study Tool
Greek Word Study Tool
George
(*) In fact, the verb ‘psopheo’ was the opposite of the verb ‘phoneo’ that was used to mean “to produce a structured sound (human voice)”.
Thus the verb ‘psopheo’ was (is) used to mean the forced death of animals. It is used too for to describe a human’s uneasy death (slug).
You can learn something new every day, indeed!
Yes, B&K has by far the best stuff if you want all the equations to match the measurements.
John, you can still have noise in a vacuum from purely electrical sources, even in a mica cap it is easy to measure the noise from DA and I have measured microphone capsules with DA in the electret.
Last edited:
- Status
- Not open for further replies.