understand, but what confuses me is the across the board drop of about 5 db when gate is down at .5ms
I can get some pulse responses this weekend
Last edited:
The driver is a 75mm long by 16mm wide ribbon under development at the moment
The measurment equipment and sine sweep are all done with OMNI MIC. Electronics are just the lab rat Yamaha reciever.
These FR measures are very close to whats measured in the normal measurment room so dont think the shop environment is causing too much issue
This measure was taken on the fly in the shop ( not the normal measurement room), and there are structures close by wich may be the ringing?? BTW note the dB scale. I was actually just working on cabinet / baffle shape size when I noticed this 5 db drop when going to a very low .5 ms gate.
Last edited:
For 0.5 ms gate length, I would discard information below 8 kHz (4 periods within gate). The fact that the frequency responses from 8 - 20 kHz differ between these plots, implies that the impulse has not sufficiently decayed within the gate time. Note that the software calculates the total energy within the gate, as a function of frequency. As the 0.5 ms gate is not sufficiently long to capture the full impulse, it misses the impulse response energy that happens after 0.5 ms. Apparently the 1 ms gate captures nearly twice as much energy (+6 dB) as compared to the 0.5 ms gate.
There should be no difference in frequency response as calculated from a sufficiently long gate and an even longer gate, as both capture the full impulse response. The latter only contains more silence after the impulse has decayed.
Here is some information about gating: B&O Tech: The Cube – earfluff and eyecandy
An interesting note he makes (at the bottom of the page) is that loudspeakers with low frequency output require longer gates, even to see what is happening at high frequencies.
There should be no difference in frequency response as calculated from a sufficiently long gate and an even longer gate, as both capture the full impulse response. The latter only contains more silence after the impulse has decayed.
Here is some information about gating: B&O Tech: The Cube – earfluff and eyecandy
An interesting note he makes (at the bottom of the page) is that loudspeakers with low frequency output require longer gates, even to see what is happening at high frequencies.
What do you mean by 'ringing'?
Last edited:
I did a sweep with my Fountek NeoCD3.5H with a cap in series. No smoothing is applied. dB Scale is at 1dB.
Blue trace is at 1ms wndow, Red is at 0.5ms.
There is indeed a 5dB drop in response.
However, anything longer than 1ms results in the same amplitude.
My bad. I thought it was at 5dB resolution.
Blue trace is at 1ms wndow, Red is at 0.5ms.
There is indeed a 5dB drop in response.
However, anything longer than 1ms results in the same amplitude.
My bad. I thought it was at 5dB resolution.
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.