0.5 meters is too close for proper summation of the drivers. You can skim through the ARTA Application Notes to find a formula that lets you calculate the far field conditions for a given speaker, or stick to 1 m. I guess far field for this speaker begins somewhere between 80-90 cm.
But the result looks proper. High frequencies would be a little lower and bass gain show yet a bit stronger modal issues with a moving mic in-room measurement. A gated quasi-anechoic will most probably reveal a balanced frequency response.
But the result looks proper. High frequencies would be a little lower and bass gain show yet a bit stronger modal issues with a moving mic in-room measurement. A gated quasi-anechoic will most probably reveal a balanced frequency response.
Thank you for your answer.
To be more clear,i mesasured 3 positions as below.
First, microphone in a high between woffer and horn.
Second,mic.in a high between w&h but 30°off axis to the left.
Third,mic in a high between w&h but 30° off axis to the right.
Then i repeat the same 3 positions measurements but with the mic in a high of horn center.
Please take into account that the room is acoustically unacceptable. In the middle of the ceiling there is a plasterboard partition that goes down about 1 meter.
Unfortunately i can't carry the speakers in another more suitable place now.😞
To be more clear,i mesasured 3 positions as below.
First, microphone in a high between woffer and horn.
Second,mic.in a high between w&h but 30°off axis to the left.
Third,mic in a high between w&h but 30° off axis to the right.
Then i repeat the same 3 positions measurements but with the mic in a high of horn center.
Please take into account that the room is acoustically unacceptable. In the middle of the ceiling there is a plasterboard partition that goes down about 1 meter.
Unfortunately i can't carry the speakers in another more suitable place now.😞
Okay, maybe show us an overview shot of the room so that we can help with placement for measurement. If you consider a relocation of the speaker/listening situation, maybe get yourself some help!
But for now, let me say that your measurement approach is an unnecessary compromise. Here are key the points:
Gated vs. ungated measurements:
a) If we go for an ungated measurement, this is because we usually want to measure the interplay between speaker and room at the listening position. This is often about 2.5-3 meters from the speakers. Due to the long distance the room dependency gets clearly visible in the modal region (<400 Hz), and the room gain in bass is documented, as well as frequency dependent cancellations due to the distance of the speaker from close by boundaries (SBIR, Speaker Boundary Interference Response). So, ungated = measuring from listening position + (both) speakers at their dedicated position. The measurements are either acquired via the Moving Mic Technique over the area of the listening position for each speaker, or a set of fixed measurements at different positions around the listening is created and subsequently summed. The latter preserves phase information, the first is more easily done.
b) If we want to measure the speaker performance sans the room, we need to find a position for the speaker where the measurement axis (i. e. halfway between woofer and tweeter) is as far as possible from the four walls, the ceiling and floor. A gate of 4 ms seconds for example requires 1.37 meters distance of the drivers from the next boundary. If you like to get some help, tell us the dimensions and we can find the optimal placement together. The gated measurement is created in the far field, meaning the distance from the speaker where the individual drivers play together and create a single wavefront. This condition can be calculated from the size of the drivers and their crossover, but it suffices to simply measure from 1 m distance. If you measure too close and the far field condition is not yet established, the results for the crossover region are flawed.
Your approach is a mix of everything and though it might sometimes help to do an ungated measurement a bit closer, for example if the gate is too short, if possible in any way, you should try to find a better placement of the speaker and measure with a sufficiently long gate.
But for now, let me say that your measurement approach is an unnecessary compromise. Here are key the points:
Gated vs. ungated measurements:
a) If we go for an ungated measurement, this is because we usually want to measure the interplay between speaker and room at the listening position. This is often about 2.5-3 meters from the speakers. Due to the long distance the room dependency gets clearly visible in the modal region (<400 Hz), and the room gain in bass is documented, as well as frequency dependent cancellations due to the distance of the speaker from close by boundaries (SBIR, Speaker Boundary Interference Response). So, ungated = measuring from listening position + (both) speakers at their dedicated position. The measurements are either acquired via the Moving Mic Technique over the area of the listening position for each speaker, or a set of fixed measurements at different positions around the listening is created and subsequently summed. The latter preserves phase information, the first is more easily done.
b) If we want to measure the speaker performance sans the room, we need to find a position for the speaker where the measurement axis (i. e. halfway between woofer and tweeter) is as far as possible from the four walls, the ceiling and floor. A gate of 4 ms seconds for example requires 1.37 meters distance of the drivers from the next boundary. If you like to get some help, tell us the dimensions and we can find the optimal placement together. The gated measurement is created in the far field, meaning the distance from the speaker where the individual drivers play together and create a single wavefront. This condition can be calculated from the size of the drivers and their crossover, but it suffices to simply measure from 1 m distance. If you measure too close and the far field condition is not yet established, the results for the crossover region are flawed.
Your approach is a mix of everything and though it might sometimes help to do an ungated measurement a bit closer, for example if the gate is too short, if possible in any way, you should try to find a better placement of the speaker and measure with a sufficiently long gate.
Please re-read this manual: https://www.audiosciencereview.com/...ents-spinoramas-with-rew-and-vituixcad.21860/
To answer your question whether you want to bring the speaker to the terrace, the question you need to answer yourself is what gate is achievable in your room. The instructions how to determine this can be found in the above link.
To answer your question whether you want to bring the speaker to the terrace, the question you need to answer yourself is what gate is achievable in your room. The instructions how to determine this can be found in the above link.
The gate is still a bit too big with 5.4ms, in practice it is rather between 3 and 4ms, which can be achieved cleanly in a normal living room.
That might also be the reason why your measurement is clearly louder in the low frequency range. There is already a part of the room influence in the measurement and the level therefore does not fit.
My measurements were also made at a distance of 1m, which makes a difference with such large distances between the drivers.
If you change your scaling so that the diagram starts at 20Hz, it looks very similar to my measurement and the crossover should be correct.
A slight dip in the range around 2kHz is intentional, by the way. In this frequency range the ear is very sensitive. Especially distorted electric guitars are very quickly annoying if too much energy is emitted into the room in this range. Therefore, I always try to take out a little bit of sound pressure level in this frequency range to make listening more pleasant.
That might also be the reason why your measurement is clearly louder in the low frequency range. There is already a part of the room influence in the measurement and the level therefore does not fit.
My measurements were also made at a distance of 1m, which makes a difference with such large distances between the drivers.
If you change your scaling so that the diagram starts at 20Hz, it looks very similar to my measurement and the crossover should be correct.
A slight dip in the range around 2kHz is intentional, by the way. In this frequency range the ear is very sensitive. Especially distorted electric guitars are very quickly annoying if too much energy is emitted into the room in this range. Therefore, I always try to take out a little bit of sound pressure level in this frequency range to make listening more pleasant.
The interference starts much earlier in your measurement. At -1ms they are already clearly visible, which distinguishes them from the example and must not be.
The measurement chain is everything that belongs to the measurement. So PC, sound card, microphone, amplifier etc. It looks to me as if one of these parts is not working quite cleanly.
The measurements themselves look ok though, I would close the window at 3ms as I said.
The measurement chain is everything that belongs to the measurement. So PC, sound card, microphone, amplifier etc. It looks to me as if one of these parts is not working quite cleanly.
The measurements themselves look ok though, I would close the window at 3ms as I said.