I don't care much about measuring accurate levels of distortion for my purposes/priorities, aside from curiosity and learning how to do it. It is helpful if I can do it well enough to see the real trend in distortion, and for this driver if distortion is not rising with frequency then I don't need to know what it is. Maybe next I'll try it on an HF driver, where the trend should be more obvious and important.
It's somewhat possible that the high 2nd harmonic is partly from the amp, which I will check in time here.
It's somewhat possible that the high 2nd harmonic is partly from the amp, which I will check in time here.
Well considering that the 2nd order was grossly affected by moving the microphone further away it kind of points towards the issue being contained to the mic.
I did wonder if it could be the amplifier myself, maybe you're using a SET tube amp with very high 2nd order distortion, but this wouldn't explain why the problem goes away when the mic is moved further away.
Obviously turning down the drive level reduces the SPL that the microphone sees, but it also reduces the distortion that the driver will produce. Given the situation however the driver is very sensitive so you're probably doing these tests at levels greater then you would normally listen at.
If I was in your position I would want to continue doing measurements at around 50cm distance, but keep turning down the drive level until the loudspeaker distortion looks like it's dominating. Once you've done this you will have an accurate representation for how the drivers distortion really varies with frequency. It might look the same, but it might not. Currently we've got the mic distortion playing a large roll in the overall measurements and as a result we don't know for sure what's going on.
I did wonder if it could be the amplifier myself, maybe you're using a SET tube amp with very high 2nd order distortion, but this wouldn't explain why the problem goes away when the mic is moved further away.
Obviously turning down the drive level reduces the SPL that the microphone sees, but it also reduces the distortion that the driver will produce. Given the situation however the driver is very sensitive so you're probably doing these tests at levels greater then you would normally listen at.
If I was in your position I would want to continue doing measurements at around 50cm distance, but keep turning down the drive level until the loudspeaker distortion looks like it's dominating. Once you've done this you will have an accurate representation for how the drivers distortion really varies with frequency. It might look the same, but it might not. Currently we've got the mic distortion playing a large roll in the overall measurements and as a result we don't know for sure what's going on.
Finally have a chance to mess around some more. I tried Steps on a couple other drivers (a DE250 and a random woofer) and found that there is always a point in SPL where the 2nd harmonic starts to shoot up disproportionately to the other harmonics. So... I went back to the TD15 and tried to find a point where it seemed settled down, 200mV. This should be around 90dB at 2kHz, I guess.
Attachments
Interesting. That certainly is quite a low SPL, it makes me wonder if there's something amiss here.
It makes me wonder whether or not there's a connection issue, like maybe only one of the balanced signals from the microphone is not getting through, like there's a cable issue or a connection issue somewhere. A balanced connection will inherently cause even order distortion products to cancel out so if only the 'hot' were connected, depending on the circuitry inside the mic, this could be causing a problem.
Sometime tomorrow I will see if I can get around to performing a measurement with one of the signal wires of my ECM8000 lifted to see if it makes much of a difference and if it does what the difference is. I cannot believe that your mic, which looks visually identical to the ECM8000, would be troubled this much by such comparatively low SPLs.
Certainly the driver is capable, distortion wise, up to 2khz, so this does open up some freedom perhaps for your crossover point. Still, the driver shows clear energy storage issues over around 1.2kHz and I would really be looking to avoid them at all costs.
It makes me wonder whether or not there's a connection issue, like maybe only one of the balanced signals from the microphone is not getting through, like there's a cable issue or a connection issue somewhere. A balanced connection will inherently cause even order distortion products to cancel out so if only the 'hot' were connected, depending on the circuitry inside the mic, this could be causing a problem.
Sometime tomorrow I will see if I can get around to performing a measurement with one of the signal wires of my ECM8000 lifted to see if it makes much of a difference and if it does what the difference is. I cannot believe that your mic, which looks visually identical to the ECM8000, would be troubled this much by such comparatively low SPLs.
Certainly the driver is capable, distortion wise, up to 2khz, so this does open up some freedom perhaps for your crossover point. Still, the driver shows clear energy storage issues over around 1.2kHz and I would really be looking to avoid them at all costs.
Hmm, well thanks for checking. On the plus side, you got me thinking about balanced connections and I realized that the problem I was having with the impedance measuring cable was because I had messed up wiring it to the balanced TRS outputs (that is to say, I didn't even realize they were balanced).
Anyway, I'm back to trying to get the measurements I need for crossover design, accepting the woofer anomaly as real. I just had an accidental discovery that is sort of interesting: a 300Hz first order high-pass on the woofer reduces the problem to almost nothing in terms of frequency response, even though it is still obvious looking at the impulse.
Anyway, I'm back to trying to get the measurements I need for crossover design, accepting the woofer anomaly as real. I just had an accidental discovery that is sort of interesting: a 300Hz first order high-pass on the woofer reduces the problem to almost nothing in terms of frequency response, even though it is still obvious looking at the impulse.
Interesting.
It would seem that the apex of the cone and former joint is probably quite resonant, this isn't perhaps a surprise.
We've seen that the edges of the cone don't resonate at high frequencies so this isn't sound waves travelling down the cone and being reflected off of the surround and then back to the former, so it must be entirely local.
Now the coins themselves might be acting a little like bracing in a loudspeaker cabinet and are acting to help prevent resonances occurring by reinforcing the 'weak-spot' and breaking/splitting up the resonating area.
The blue tac would more be trying to dampen the resonances, which are already pretty strong, so it's no surprise that it had little effect.
I know that a lot of larger pro drivers use ribs in the cone to reinforce it, perhaps this is en example of where some re-enforcement at the neck of the cone might be useful, or a very rigid dust-cap. Of course the dust-cap could add in trouble all of its own.
It would seem that the apex of the cone and former joint is probably quite resonant, this isn't perhaps a surprise.
We've seen that the edges of the cone don't resonate at high frequencies so this isn't sound waves travelling down the cone and being reflected off of the surround and then back to the former, so it must be entirely local.
Now the coins themselves might be acting a little like bracing in a loudspeaker cabinet and are acting to help prevent resonances occurring by reinforcing the 'weak-spot' and breaking/splitting up the resonating area.
The blue tac would more be trying to dampen the resonances, which are already pretty strong, so it's no surprise that it had little effect.
I know that a lot of larger pro drivers use ribs in the cone to reinforce it, perhaps this is en example of where some re-enforcement at the neck of the cone might be useful, or a very rigid dust-cap. Of course the dust-cap could add in trouble all of its own.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.