No problem can be resolved with THD measurements. But it makes sense to investigate what improvements can be achieved by different means. Human perspective is an important factor just as evaluation of aircraft handling qualities using the Cooper Harper rating system.
Some quick and dirty measurements show quite an amount of inductance difference with shorting ring thickness, so I as wondering what the effects would be if you did both in the former and the pole piece. No document showing anyone whom has done this.
Some quick and dirty measurements show quite an amount of inductance difference with shorting ring thickness, so I as wondering what the effects would be if you did both in the former and the pole piece. No document showing anyone whom has done this.
Last edited:
But here is my point. We must know what needs to be fixed - "needs improvement" - and what doesn't. It makes no sense improving things that don't matter just because we can. From where I sit all the problems with speakers are at the systems level, where it all comes together. The drivers are good enough, or at least "good enough" drivers exist. It just seems to me to be a waste of time obsessing about the drivers if they don't matter.
Could be, but I have worked through all the analog section of the system and find the dominant issue is with the driver, both in terms of measurement and perception.
Absolutely true. No sense obsessing about trivial issues.
But for me, it leads to one inescapable conclusion: there is no way to make a better system without a DSP. Waste of effort otherwise*.
That conclusion may not sit well with a manufacturer who sells boxes, however good, with crossovers in them**. But it makes sense for DIYers who will sit their speakers in their rooms and run REW curves to see what they got.
B.
*wasn't somebody going to post their derogatory measurements of the Behringer? Waiting, waiting....
**of course, even with turn-key boxes, you can still have a DSP in the chain... just does make nearly as much sense at a multi-amped system
Last edited:
The dominate problem is the "speaker system", but I would disagree that it is the individual drivers. It's how you put those drivers together that matters most. The driver matters, sure, but not nearly as much as the system design. The system design follows from how it interfaces with the room and from that follows the drivers, but I have never had a problem finding drivers that fit the requirements.
Getting the system smaller would be, IMO the #1 biggest issue. In a large size near perfection can be achieved. Smaller, and problems get big - fast!!
Getting the system smaller would be, IMO the #1 biggest issue. In a large size near perfection can be achieved. Smaller, and problems get big - fast!!
That DSP is a major advantage is clear. But, and I have made many speakers both ways, I do not find DSP is a "game changer". It's easier, for sure, and it can address ever smaller issues, but the big issues can be handled equally well with passive as with active. When active was very expensive, it wasn't worth it, but now ... why not!?
Actually when one has a 7.3 system, passive looks pretty good. I had passive and went active for my last set. Was the improvement worth it? Its so hard to tell that I would have to say ... Not!!
SO WHAT!!!? The numbers don't matter in the least. Why are you obsessing about them!?
Amps sound worse with very low THD than speakers with very high THD. And you are going use that data to make judgments?
I don’t think I have seen phase response of your system.
I just used FIR because it was convenient enough to try. It does not cause too much work to an end user using a normal computer either.
I thought we’re were focusing on objective data. But maybe you want to voice your opinion about low distortion amplifiers in the Mod86 amplifier thread.
Normally I try to correlate numbers with subjective review. If they do not match up, it just means I need to work on a different issue.
You are trying hard not to address the "elephant in the room", namely, "the room".
Whatever qualities an off-the-shelf box might have, any normal room at home monumentally alters the sound and there is just no feasible way to make a speaker or make a room that faces that challenge.
If you want to address the room challenge, you must have some levers of control. Granted, that starts with making the speakers and the room good. But that takes you only so far.
B.
The problem for me with active crossovers and DSP is the cost of DACs. I can use a passive crossover speaker system with a single Benchmark DAC-3 and hear very detailed audio. To maintain that level of detail with active crossover DSP, among other things I would need more DAC channels. If cost were no object, then fine, but that's not the case right now.
Last edited:
If you mean the VC former, that would be a very bad idea. It would brake the VC movement through induced eddy currents. That is why all VC's made out of conductive material have a vertical slit: to prevent such currents from developing.
If you mean the top plate of the magnet assembly, that wouldn't make any sense, since it lays outside the magnetic field produced by the VC.
Yes, it is correct that I have never shown it, not convinced that it is relevant, like THD, which I don't show either.
But I did get curious once and found that the phase stays with 30° over the entire usable bandwidth of the system. Not an FIR 0° variation, but not bad at all for a passive system.
Seems unlikely in a two way? It's because the tweeter is mounted so much further back than the woofer and there is usually a phase reversal. This brings the phase into a much more linear result than one would have with two direct radiating speakers unless the tweeter were moved back substantially.
I was wondering whether the conductive material had a slot due to manufacturing cost or not. But it cannot have eddy currents to create a brake because the speed is not enough, and I wonder whether the magnetic field is strong enough as well since the conductivity and the strength and speed determine the braking force.
But the thickness of the former wall will probably not reduce the coil inductance so much due to thickness limitation as well.
I have tried FIR filters based on different measurement locations, with and without mic calibration file application to the measurement data. The best result was near field measurement with no mic calibration data applied. My assessment is that near field measurement reduces diffraction in the measurement, while not using mic calibration file sort of allows the FIR filter to also compensate for the recording mic somewhat.
Using digital crossovers has the benefit of implementing time alignment digitally.