J Sigmo
Known around here
- Feb 5, 2018
- 996
- 1,336
This will be too long, but then you're all prepared for that with my posts. 
I don't have any Dahua cams with microphones or the capability to support an external mic. But my lowly Reolinks produce very good audio with their built-in microphones with the following caveats, which I believe will likely be true for just about any camera:
1. Wind rushing over the small opening for the microphone will create very annoying and loud noise, and we have a LOT of wind here. So this makes the audio so annoying as to force you to mute it when playing back or watching live on cameras with microphones on my house, if the wind is blowing (which is is a lot of the time). I installed some cut-up sweat socks so that they fit over the mic openings, and this helps somewhat, but not as much as a proper "dead cat" wind cover would.
2. Loud sounds (such as wind or kiddos and their muffler-less cars, or adult children riding their Harleys) causes clipping (hard limiting) either in the mic preamps, the A-D conversion, or elsewhere in the signal chain. That, too, makes listening to the audio unbearable.
I've designed and manufactured several audio systems that include compression to prevent this sort of issue. While not something I generally like for high-fidelity music listening, compression can be fantastic for something like our security camera systems, police/fire/emergency communications, etc. In commercial and civil radio comm systems, there is often NO compression, which is stupid beyond all belief, especially in a $10,000 dollar (no exaggeration!) radio. You always have three types of radio users: Those who hold the mic at a proper distance from their lips and speak clearly and at a reasonable volume. Those who apparently sit on the microphone and mumble softly as if they had a mouth full of marbles. Those who think they need to yell really loud because the person on the other end is really far away!
As a result, the fire/police/whatever folks have to turn up their receivers very loud to be able to hear the under-their-ass-microphone-placement mumblers. And that means that the yellers and even the folks who understand how it works all end up making everyone's ears bleed. Surprisingly, as a fireman, it wasn't the gas-operated chop saws, or jaws of life and the like that wrecked my son's ears, it was the radios in firetrucks and fire stations being cranked up super loud so they could hear the soft-talkers.
Anyhow, the same lack of compression is responsible for a lot of the complaints we read about on here with poor audio quality from security cameras or even external microphone setups.
My suggestion is that the manufacturers of these cameras use one of the commercially available microphone preamp ICs that has compression built in. Analog devices manufactured the ICs I've used, but I'm sure there are other IC manufacturers who make similar chips:
These are some that I used in the past:
https://www.analog.com/media/en/technical-documentation/data-sheets/ssm2166.pdf
https://www.analog.com/media/en/technical-documentation/data-sheets/ssm2167.pdf
In the ideal setup, you choose the components in the compression amplifier design to work with your chosen microphone (or mic element) so that soft sounds are still clearly audible and above the system noise floor, but loud sounds are compressed downward so that at no point, does the mic/preamp output saturate anything further down the processing chain, and thus, you don't have to put up with that extremely annoying, and unintelligible clipped (hard limited) audio. Keep in mind that digital audio systems are far less tolerant of overdrive than were analog recording systems of the past. When you overdrive magnetic tape, you get higher distortion, but not a "brick wall" of limiting like you get with a digital recording or transmission system. So it's extremely important to prevent feeding a "too hot" audio signal into a digital audio system.
Even if you don't use a compression amp, you still need to match the audio level coming out of your mic/preamp system to the input range of your camera or whatever input you're feeding. Clipped audio sounds like crap! If the microphone/preamp maker doesn't give you a sensitivity rating for their setup, and the camera (or whatever) maker tell you what the maximum input voltage to their device is, you must make some tests and measurements to determine how to set things up.
You can feed the microphone input to your camera (or whatever) with a signal generator and determine the clipping point. Then you can measure the microphone/preamp output when exposing it to known sound pressure levels and determine that system's sensitivity. Then you can set the gain of an intermediate amplifier or build a "pad" to adjust the level going into your camera (or other thing) to an appropriate range. This will NOT JUST MAGICALLY HAPPEN. Nobody is that lucky!
Signal levels must be matched.
Understanding Microphone Sensitivity | Analog Devices
I don't have any Dahua cams with microphones or the capability to support an external mic. But my lowly Reolinks produce very good audio with their built-in microphones with the following caveats, which I believe will likely be true for just about any camera:
1. Wind rushing over the small opening for the microphone will create very annoying and loud noise, and we have a LOT of wind here. So this makes the audio so annoying as to force you to mute it when playing back or watching live on cameras with microphones on my house, if the wind is blowing (which is is a lot of the time). I installed some cut-up sweat socks so that they fit over the mic openings, and this helps somewhat, but not as much as a proper "dead cat" wind cover would.
2. Loud sounds (such as wind or kiddos and their muffler-less cars, or adult children riding their Harleys) causes clipping (hard limiting) either in the mic preamps, the A-D conversion, or elsewhere in the signal chain. That, too, makes listening to the audio unbearable.
I've designed and manufactured several audio systems that include compression to prevent this sort of issue. While not something I generally like for high-fidelity music listening, compression can be fantastic for something like our security camera systems, police/fire/emergency communications, etc. In commercial and civil radio comm systems, there is often NO compression, which is stupid beyond all belief, especially in a $10,000 dollar (no exaggeration!) radio. You always have three types of radio users: Those who hold the mic at a proper distance from their lips and speak clearly and at a reasonable volume. Those who apparently sit on the microphone and mumble softly as if they had a mouth full of marbles. Those who think they need to yell really loud because the person on the other end is really far away!
As a result, the fire/police/whatever folks have to turn up their receivers very loud to be able to hear the under-their-ass-microphone-placement mumblers. And that means that the yellers and even the folks who understand how it works all end up making everyone's ears bleed. Surprisingly, as a fireman, it wasn't the gas-operated chop saws, or jaws of life and the like that wrecked my son's ears, it was the radios in firetrucks and fire stations being cranked up super loud so they could hear the soft-talkers.
Anyhow, the same lack of compression is responsible for a lot of the complaints we read about on here with poor audio quality from security cameras or even external microphone setups.
My suggestion is that the manufacturers of these cameras use one of the commercially available microphone preamp ICs that has compression built in. Analog devices manufactured the ICs I've used, but I'm sure there are other IC manufacturers who make similar chips:
These are some that I used in the past:
https://www.analog.com/media/en/technical-documentation/data-sheets/ssm2166.pdf
https://www.analog.com/media/en/technical-documentation/data-sheets/ssm2167.pdf
In the ideal setup, you choose the components in the compression amplifier design to work with your chosen microphone (or mic element) so that soft sounds are still clearly audible and above the system noise floor, but loud sounds are compressed downward so that at no point, does the mic/preamp output saturate anything further down the processing chain, and thus, you don't have to put up with that extremely annoying, and unintelligible clipped (hard limited) audio. Keep in mind that digital audio systems are far less tolerant of overdrive than were analog recording systems of the past. When you overdrive magnetic tape, you get higher distortion, but not a "brick wall" of limiting like you get with a digital recording or transmission system. So it's extremely important to prevent feeding a "too hot" audio signal into a digital audio system.
Even if you don't use a compression amp, you still need to match the audio level coming out of your mic/preamp system to the input range of your camera or whatever input you're feeding. Clipped audio sounds like crap! If the microphone/preamp maker doesn't give you a sensitivity rating for their setup, and the camera (or whatever) maker tell you what the maximum input voltage to their device is, you must make some tests and measurements to determine how to set things up.
You can feed the microphone input to your camera (or whatever) with a signal generator and determine the clipping point. Then you can measure the microphone/preamp output when exposing it to known sound pressure levels and determine that system's sensitivity. Then you can set the gain of an intermediate amplifier or build a "pad" to adjust the level going into your camera (or other thing) to an appropriate range. This will NOT JUST MAGICALLY HAPPEN. Nobody is that lucky!
Signal levels must be matched.
Understanding Microphone Sensitivity | Analog Devices
