Matt,
Thst is an interesting challenge. It is one of the key things needed to measure DDR, but low level in the presence of a more normal signature.
One can ask lots of questions, one woul be: the “big” response measures flat, but what is the “little” FR? What is the “little” FR when measured with the “big” response.
An anechoic chamber would be required, and possibly something like Microflown.
dave
Thst is an interesting challenge. It is one of the key things needed to measure DDR, but low level in the presence of a more normal signature.
One can ask lots of questions, one woul be: the “big” response measures flat, but what is the “little” FR? What is the “little” FR when measured with the “big” response.
An anechoic chamber would be required, and possibly something like Microflown.
dave
I've had a quick look in the mic locker. Candidates include:
- SE4400a - large diaphragm condenser, 25mV/Pa, 16dBA noise
- Beyer MC930 - Small diaphragm condenser, 30mV/Pa, 16dBA noise
- AKG C3000 - Large diaphragm condenser, 20mV/Pa, 18dBA noise.
- Beyer MM1 - Omni measurement mic, 15mV/Pa, 26dBA noise
- A few dynamic mics, but with 1/10th the voltage output.
Looks like the MC930 wins for low level measurements - joint lowest noise figure, but more electrical output means the mic preamp won't need as much gain, so lower electrical noise there.
The input device will be a QSC TM30Pro, which is a digital mixing desk which can feed audio to/from the laptop over USB.
I quite like the concept of DDR. I've been trying to work out how to measure it, and it's difficult to come up with anything that you can put a solid number to.
Some ideas:
- Play pink noise, run a frequency sweep at a lower level. Record with/without the sweep with the same section of pink noise, use software to cancel the two and you're left with a frequency sweep extracted from the noise. Analyse in REW.
- Same idea, but low-level music. Extract the music for subjective evaluation.
Instinctively, a driver's masking of low-level details ought to show up in frequency response sweeps, perhaps as higher harmonic distortion. While harmonic distortion and inter-modulation aren't directly linked, it's difficult to imagine a driver producing one but not the other.
I have the equipment and time to make a reasonable go of this, so if anyone has thoughts on the signal processing and methodology, I'd be interested to hear it.
Chris
- SE4400a - large diaphragm condenser, 25mV/Pa, 16dBA noise
- Beyer MC930 - Small diaphragm condenser, 30mV/Pa, 16dBA noise
- AKG C3000 - Large diaphragm condenser, 20mV/Pa, 18dBA noise.
- Beyer MM1 - Omni measurement mic, 15mV/Pa, 26dBA noise
- A few dynamic mics, but with 1/10th the voltage output.
Looks like the MC930 wins for low level measurements - joint lowest noise figure, but more electrical output means the mic preamp won't need as much gain, so lower electrical noise there.
The input device will be a QSC TM30Pro, which is a digital mixing desk which can feed audio to/from the laptop over USB.
I quite like the concept of DDR. I've been trying to work out how to measure it, and it's difficult to come up with anything that you can put a solid number to.
Some ideas:
- Play pink noise, run a frequency sweep at a lower level. Record with/without the sweep with the same section of pink noise, use software to cancel the two and you're left with a frequency sweep extracted from the noise. Analyse in REW.
- Same idea, but low-level music. Extract the music for subjective evaluation.
Instinctively, a driver's masking of low-level details ought to show up in frequency response sweeps, perhaps as higher harmonic distortion. While harmonic distortion and inter-modulation aren't directly linked, it's difficult to imagine a driver producing one but not the other.
I have the equipment and time to make a reasonable go of this, so if anyone has thoughts on the signal processing and methodology, I'd be interested to hear it.
Chris
Aren't they both due to the same non-linearities and as one increases so will the other?
I'm hoping the right equipment will help with that.
What're your thoughts on this, as a test method:
- Play pink noise out of the DUT
- Record that
- Play test signal + same section of pink noise
- Record that
- Subtract the two recordings, which should result in the pink noise signals being cancelled out and the test signal should remain.
It seems to me that a perfect DUT would recover the test signal perfectly, while a device which innately masks low-level signals would also mask the test signal.
I just can't decide on a good test signal.
Something like a sawtooth wave might be useful - lots of harmonics, and a cyclic signal. I suppose you could then directly calculate the distortion level from there.
Thoughts?
Chris
That would be me.
I was trying to pull very low level signals out of higher level signals played by a speaker.- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.