With a driver this size, my guess would be a total weight of 5 quarters in a circle.
Measurement of distortion is a complicated matter because you have will have different values in far field vs near field. Then depending on the type of sweep you use in far field.
Trying a bit more distortion stuff here. Moving the mic farther away doesn't seem to change anything except for making the frequency response a bit messier, and adjusting the preamp levels all the way from buried in noise to just below clipping gives the same results too. So, I forged ahead and tried 2VAC, which raised the percentage from what's above by, I dunno, maybe 0.03%. Next I tried 0.5VAC, which brought D2 down to 0.3%.. a bit more of a change than I expected.
Well I don't have a scale anyway, but I rolled out a ball about 2cm in diameter and stuck it in the middle ring of the spider. The change was either nothing or 0.2dB off a few spikes, depending on how much you want to trust my measurements.
I don't think it is hard to say. If the issues associated with this driver were being controlled via the addition of the blue tac into the spider then it would show up in the FR and impulse response. Or let me put this another way, if the blue tac was having any kind of effect that was worth keeping then it would be showing up.
I've just finished writing a guide btw on making loudspeaker measurements that you may find useful.
measurement guide.pdf
Yes the anomaly shows up everywhere, as wiggles in the impedance plot, as wiggles in the frequency response and as bumps in the impulse response. You'd figure that if the blue-tac had solved the issue, or helped the issue, that all three would show signs of improvement.
I am not knocking the blue-tac, clearly there are resonances going on here that mar the performance quite significantly in the drivers upper range.
Dumptruck. I would still be interested in seeing a distortion measurement, done at the same drive level as before, but with the mic a bit further away and with the gain turned down slightly on the mic pre. Sometimes my mic pre doesn't show signs of clipping, but still the distortion level has risen slightly before that occurs.
measurement guide.pdf
Yes the anomaly shows up everywhere, as wiggles in the impedance plot, as wiggles in the frequency response and as bumps in the impulse response. You'd figure that if the blue-tac had solved the issue, or helped the issue, that all three would show signs of improvement.
I am not knocking the blue-tac, clearly there are resonances going on here that mar the performance quite significantly in the drivers upper range.
Dumptruck. I would still be interested in seeing a distortion measurement, done at the same drive level as before, but with the mic a bit further away and with the gain turned down slightly on the mic pre. Sometimes my mic pre doesn't show signs of clipping, but still the distortion level has risen slightly before that occurs.
This is not exactly the same level, but I tried to include as much as possible in one shot. This is all 2nd harmonic percentage at 1V. It doesn't look like I was very wrong about the preamp not making any difference, but clearly I was mistaken about distance. The "mic@dB" numbers are referring to the digital level that results from the mic preamp level control, because that's the only easy reference point I have. -40dB is as low as it goes (unless I turn down the digital output of the software - I could only try so much at once).
Also, I did check for consistency by repeating three of these, though it isn't shown here. The magnitude of peaks and valleys changes a bit from measurement to measurement, but the overall curve/magnitude of the distortion percentage and the location of peaks and valleys do not change one bit.
I will definitely check out your guide, but I haven't yet.
Attachments
Last edited:
Yeah thanks for this, this is what I wanted to see. The microphone preamp clearly isn't anywhere near clipping, but having the microphone at 30cm with the driver driven at 1V causes the microphones distortion to increase. This is probably why the second harmonic also dropped off considerably in your other distortion plot beyond 4khz, because the output level of the driver significantly drops at that level and in turn so does the SPL the microphone is exposed to.
When measuring this drivers distortion with your mic, it would make sense to measure it at about 50cm-80cm, but turn the drive level down so that the microphone is happy. This is always something I have to play around with when making measurements myself, especially when done at quite high SPLs. Normally I measure drivers with 2.83Vrms, but this is simply too much sometimes when a high sensitivity driver is used.
It would be nice to see a full range sweep, say from 50-10kHz with the mic set at an appropriate distance and the drive level turned down so that the mic is happier. What mic is it that you are using?
When measuring this drivers distortion with your mic, it would make sense to measure it at about 50cm-80cm, but turn the drive level down so that the microphone is happy. This is always something I have to play around with when making measurements myself, especially when done at quite high SPLs. Normally I measure drivers with 2.83Vrms, but this is simply too much sometimes when a high sensitivity driver is used.
It would be nice to see a full range sweep, say from 50-10kHz with the mic set at an appropriate distance and the drive level turned down so that the mic is happier. What mic is it that you are using?
It's just the Dayton EMM-6, with PE's calibration. How's this?
Attachments
Distortion measurements can be a bit of a pain to do sometimes and this shows somewhat in your diagram.
I use the Behringer ECM8000 and haven't had any huge issues with exaggerated 2nd order, but then again I know there can be consistency issues with that mic and the EMM-6 is supposed to be a close relative.
I suppose what I'm saying here is that the 2nd order distortion seems too high for a driver of that calibre and indeed your measurements, with a fixed drive level, but different measuring distances, showed that the mic is having a considerable effect on this.
Usually though, when I clip the mic it's extremely obvious as the other distortion products shoot up too. You're not clipping the mic as per say though, it's just becoming less linear as SPL increases, which is to be expected, but not to this degree I'd say.
I mean here's a measurement I did, with the mic around 50cm away from the loudspeaker with the loudspeaker producing around 105dB.
As you can see, at this kind of level the 2nd harmonic is still around -45dB on average which = ~ 0.5% distortion. And as expected, at lower drive levels the distortion simply decreases by an expected amount. Maybe it is just your mic and there's not a lot you can do about it.
One thing to mention though is how ragged your response is. This improves the closer the mic is to the loudspeaker as it helps to remove the room from the equation, but as explained you do need the mic a certain distance away from the loudspeaker to get accurate high frequency data.
From a visual point of view though, it is easier to see what's going on if you squash the vertical axis, that is give it the full range from 0.00001% up to 100%. You can also tend to improve things by increasing the duration of the tone bursts, this can be controlled by altering the min integration time and the transient time. It is worth playing with all of the 'timing' settings a little to see if you can improve things. Using the 'thick pen' setting can also help.
It all depends on how much effort you want to put into trying to get the best measurements you can. I find that I want to make very good distortion measurements so I put quite a lot of effort into getting decent results. Some would say it's not so critical though
You might find it helpful to post two graphs too, one that displays 2nd and 3rd harmonic and one that displays the 4th and 5th.
I mean here's an image with just the 2nd and 3rd on display and it makes them both very easy to look at.
I use the Behringer ECM8000 and haven't had any huge issues with exaggerated 2nd order, but then again I know there can be consistency issues with that mic and the EMM-6 is supposed to be a close relative.
I suppose what I'm saying here is that the 2nd order distortion seems too high for a driver of that calibre and indeed your measurements, with a fixed drive level, but different measuring distances, showed that the mic is having a considerable effect on this.
Usually though, when I clip the mic it's extremely obvious as the other distortion products shoot up too. You're not clipping the mic as per say though, it's just becoming less linear as SPL increases, which is to be expected, but not to this degree I'd say.
I mean here's a measurement I did, with the mic around 50cm away from the loudspeaker with the loudspeaker producing around 105dB.
As you can see, at this kind of level the 2nd harmonic is still around -45dB on average which = ~ 0.5% distortion. And as expected, at lower drive levels the distortion simply decreases by an expected amount. Maybe it is just your mic and there's not a lot you can do about it.
One thing to mention though is how ragged your response is. This improves the closer the mic is to the loudspeaker as it helps to remove the room from the equation, but as explained you do need the mic a certain distance away from the loudspeaker to get accurate high frequency data.
From a visual point of view though, it is easier to see what's going on if you squash the vertical axis, that is give it the full range from 0.00001% up to 100%. You can also tend to improve things by increasing the duration of the tone bursts, this can be controlled by altering the min integration time and the transient time. It is worth playing with all of the 'timing' settings a little to see if you can improve things. Using the 'thick pen' setting can also help.
It all depends on how much effort you want to put into trying to get the best measurements you can. I find that I want to make very good distortion measurements so I put quite a lot of effort into getting decent results. Some would say it's not so critical though
You might find it helpful to post two graphs too, one that displays 2nd and 3rd harmonic and one that displays the 4th and 5th.I mean here's an image with just the 2nd and 3rd on display and it makes them both very easy to look at.
Attachments
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.