Design considerations for active speaker systems

amirm

Banned
Apr 2, 2010
15,813
37
0
Seattle, WA
In another thread, Jørn post an excellent and exhaustive list of considerations when one looks at designing active speakers. What problems it solves. What difficulties it brings.

Here is his list from the other thread:

Well, since this thread is quite far off its initial question, I'll try to bring it back... for a while :D


I think many issues in active vs passive has to be investigated separately:
1: In active speakers, one can match the amplifier perfectly to the drivers. This can lead to less expensive amplification. One also has perfect control over clipping limits, impedance in driver vs load margin of the amplifier and other basic parameters.
2: In active speakers, nonlinear parameters like Kms and BxL nonlinearity can be corrected by using dynamic digital filters in a feed forward design.
3: Active speakers can be servo controlled.
4: Active speakers can be used with current feedback amplifiers. Such amplifiers will ignore voice coil heating, Le nonlinearities and so on.
5: Current feedback amplifiers will potentially have problems with change in resonance frequency over time and resonance jumping due to Kms nonlinearity.
6: When dealing with high Q phenomena like metal cone breakup phenomena, a passive crossover can be used as an electrical load that extracts energy from that resonance. In theory, this could be done with an amplifier directly on the driver, but in a real world, this does not work.
7: In theory, a measured impulse response can be corrected digitally for each driver in an active speaker. However, there are multiple phenomena that are not currently possible to solve correctly.
8: Different directivity effects can be achieved much more effectively with implementing individual DSP filters for each driver.
9: In some crossovers (passive) a lot of energy is being lost. (I always use thermographic measurements to map power loss when building crossovers).
10: Precision filtering close to a drivers resonance frequency is impossible with passive crossover.
11: Total energy from an amplifier will not change when splitting it into several limited frequency bands. The sum of power at a given moment in a multi amplifier setup is the same as for a single amplifier setup, given that the efficiency is the same.
12: In a passive system, the crossover is a contributor to lower efficiency. Especially where matching sensitivity leads to extensive use of resistors. Also shunt filters are very ineffective.
13: In an active system, the idle power of the multiple power amplifiers are in theory higher than the idle power of the single amplifier for a passive speaker.
14: The use of very steep filters do come with a price. If the filters are IIR type, they will have a lot of phase distortion.
15: The use of steep and phase linear FIR filters will cause a significant delay and pre ring effect. Some claim to have solved this but skeptics are not convinced yet.
16: A loudspeaker drivers nature of being a time variant load makes high resistance series components typically cause a significant raise in distortion.
17: In a passive speaker, a well designed SD system (copper sleeves covering critical parts in the magnet system) is key. In an active system, this is less important (and copper parts adds cost).
18: In an active speaker, one can compensate for theoretically improper use of drivers. For example, a woofer used in an enclosure smaller than it is designed for, can be EQ-ed to perform well. In a passive speaker, this EQ has to be added somewhere in the signal chain.
19: In an active speaker, one do not have to care that much about driver sensitivity. A 84dB bass section can be added to a 92dB midrange section and a 90dB tweeter section. This gives a wider selection of drivers to choose from as well as the possibility of even more powerful bass performance from a small enclosure.
20: Active speakers can more easily be equipped with dynamic compressor, temperature monitoring and so on to prevent the speaker from malfunction.
21: DSP based speakers makes upgrade very easy and cost effective.
22: Active speakers can more effectively be combined with room adapting tools (compensation for listening distance, room size, distance to near boundaries at low frequencies etc).
23: DSP based speakers can easily be equipped with EQ that enables the user to reduce the impact of room resonances.

I believe there are a lot of more issues, that I can not think of right now. An active speaker is not a completely obvious choice over a passive one and vice versa. It is often a question of budget.

I thought I put up the list and then ask people to ask specific questions about the items above.
 

amirm

Banned
Apr 2, 2010
15,813
37
0
Seattle, WA
OK, I will go first :).

If I live in a world where I am already performing room correction and so a DSP and set of filters are in the loop, what drawbacks are still left?
 

DonH50

Member Sponsor & WBF Technical Expert
Jun 22, 2010
3,947
306
1,670
Monument, CO
Handling dynamic signal-dependent distortion, including the effects of self-heating and dynamic phase/group delay variation, would be an interesting issue to tackle. Can DSP counteract ringing in the cones?
 

Midgard Audio

Well-Known Member
Nov 29, 2010
23
0
871
Handling dynamic signal-dependent distortion, including the effects of self-heating and dynamic phase/group delay variation, would be an interesting issue to tackle. Can DSP counteract ringing in the cones?

I believe that will be the main issue, a DSP can not successfully remove all kinds of resonances. If you look at resonance points where sound is phased out, all types of DSP tends to try to increase the SPL. However, a phase out is still a resonance. An increase in SPL is not necessarily a great idea :D

Since a resonance is defined by its Q, and since Q can only be changed by changing one of the actual defining parameters for the resonance, you can not successfully remove a resonance by altering the input signal. In room acoustics, the measured result is the sum of the room resonance and the direct sound. In a nearfield measurement, the difficult resonances are those that are not directly in the path of the input signal, but still is trigged by the input signal. An example of this is the cone ringing phenomena. Another is standing waves in an enclosure. An example that is in the path of the input signal is the fundamental resonance of a driver. The problem is, as Earl Geddes also mentioned in the initial thread, that these artifacts do happen in more dimensions than what you have access to through the DSP correction.
 

DonH50

Member Sponsor & WBF Technical Expert
Jun 22, 2010
3,947
306
1,670
Monument, CO
Every now and then I get lucky and guess right... :)

There are compensation schemes in the signal-processing world that could (I think) compensate cone ringing (if re-applied to that problem instead of say radar/lidar signals) but in some ways the speaker problem is even more complex due to higher interaction with the environment.

If you hit cone breakup, I think all bets are off (turn it down!) - Don

p.s. Interesting topic that I hope to find more time to follow!
 

Midgard Audio

Well-Known Member
Nov 29, 2010
23
0
871
Ringing is just a defined break up resonance. But the ringing phenomena is also followed by some additional artifacts. Linkwitz has a very interesting article on energy storage issues.

http://linkwitzlab.com/frontiers_2.htm#M
http://linkwitzlab.com/frontiers-2-M2.htm

The really interesting bit is what happens at half the break up frequency. Since the breakup is at 4,4kHz, a dip should be found at half that frequency, 2,2kHz. But here you actually find a resonance, but not the dip that is associated with it. The dip is at 2,1kHz. This is because the effect is caused by the edge and center moving in oposite direction, and at the resonance, the two parts of the cone have very different effective cone area. At the point where they have the same area, the frequency is a bit lower. This is more complex than the kind of filtering you are talking about. Normally, DSP units will raise the dip, adjust the phase and

I am not sure if it is possible to do it with DSP in theory, but it is indeed a few challenges in generating a filter that does this correctly every time. These things chage so much depending on the input signal.
 

About us

  • What’s Best Forum is THE forum for high end audio, product reviews, advice and sharing experiences on the best of everything else. This is THE place where audiophiles and audio companies discuss vintage, contemporary and new audio products, music servers, music streamers, computer audio, digital-to-analog converters, turntables, phono stages, cartridges, reel-to-reel tape machines, speakers, headphones and tube and solid-state amplification. Founded in 2010 What’s Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals, we enjoy learning about new things and meeting new people, and participating in spirited debates.

Quick Navigation

User Menu

Steve Williams
Site Founder | Site Owner | Administrator
Ron Resnick
Site Co-Owner | Administrator
Julian (The Fixer)
Website Build | Marketing Managersing