Tech Talk

Tech Talk

Active Speaker Systems and Their Advantages

By Armando Martinez & Larry P. Brown


Active speaker systems have been on the market for many years. At the same time we are seeing how streaming services are becoming ubiquitous, so it makes perfect sense to integrate such functionality into our products, while expanding the line of conventional speaker systems to support them.

Independently of their form factor, there is no doubt that active speaker systems bring an extra level of convenience to our daily lives. They also close the gap between streaming/digital services and local sources - TV, DVD players, etc. - becoming an all in one solution.

Some of us would think that such levels of convenience come with a price, but not necessarily. Technology is opening the door to new opportunities in many different areas, allowing System Engineers to develop new systems without compromising performance.
From the development perspective, active systems bring some additional benefits which at the end translate as a direct benefit to the end user.

We will agree that the main goal behind every system is not to add any artifacts in any of their forms to the reproduced content. At the same time, from a distance, speakers can be seen as simple devices. There are not many components in their construction, however, as simple as they appear they are complex in nature. Any variables, if not well considered, can have a detrimental effect on their performance.

“When you want to know how things really work, study them when they are coming apart.” William Gibson

One of the biggest benefits of an active system is that the System Engineer has control over all the components in the reproduction system, other than the source.

In a passive speaker design, the System Engineer specifies the voltage range that the speaker is suitable for (Amplification) and that’s it. He has no control about what is connected, and/or the capabilities derived from it, relying on the proper selection of the components in the signal chain.

Let’s evaluate how active systems can help during the development process. Of course, comparing both systems, active are far more complicated, but at the end the advantages derived from such complexity, if well implemented and or utilized, can have a positive balance. Join us in this short trip across some of the components in the system.

As in any system, passive or active, System Engineers design transducers based on the acoustic output that the system is designed for. System response across the audible bandwidth is a given, but in this case the designer is ahead of the game because driver capabilities and the amplifier selection are perfectly matched. No less and no more than is needed to provide the required dynamics and maximum output the transducers at system level are designed for. Why put a V8 engine into a compact size car? We’re not saying that it is not possible. (INFAMOUS below is a good example) It will work, but if not well balanced, eventually something is going to give up.

In the case of the high frequency components used in JBL Studio Speakers (compression drivers), their natural high sensitivity requires much less power compared to their low frequency counterparts. In this case and on purpose, the amplifier is sized accordingly. Also, any additional attenuation is implemented in the digital domain (DSP), eliminating the need of passive/lossy components to balance the acoustic response across the audible spectrum. This is not possible in passive systems even in bi-amp mode.

Active Crossovers vs Passive
DSP is one of the key elements. Digital processing is everywhere, it is part of our daily life in many different forms. In fact, it works so seamlessly that we are not even aware of its existence.

In the case of an active audio system, DSP provides far more flexibility. Do not get us wrong, a very well designed passive crossover works in an excellent way, but the System Designer is limited in how much can be done at a practical level. There are always limitations in how all these components fit together and also not much anechoic correction can be applied - not because it is not possible, but the System Engineer needs to be practical.

Lots of respect goes to them because even with those constraints they can produce amazing sounding systems. This is where experience plays a big role during development process through to the proper selection of the components in the system.

Audio DSPs are designed/structured to perform computations in a very efficient way. For instance, selected DSP can process over 3,000 double precision biquads filters per sample. The biquad term is derived from the mathematical representation of the Digital IIR filters in use and defined by a set of quadratic equations. The same quadratic equations taught in middle school.

Just to give you an idea, in a system sampling at 192KHz, DSP needs to process all the equalization (comprised of several filters per channel and every one of them represented by a biquad). What is fascinating is the fact that those little integrated machines are crunching numbers at a very fast pace. In some ways it is difficult to picture all the arithmetic taking place per sample (every 5.2uS!).

From the design perspective, the System Engineer can implement not only basic crossover functions, but additional improvements derived from their anechoic measurements - time alignment, phase correction, limiting, compression, etc.

Another big plus from Digital Processing is that it does not suffer from variations/tolerances derived from passive components. In the digital domain, one plus one always equals two. The response derived from an active crossover will be consistent across all speakers, which is not the same in their passive counterparts unless very tight tolerances are in place.

A valid concern is how transparent digital processing is. DSP features full 32-bit processing, capable of performing more than 1.2 billion MAC operations per second. Reviewing performance numbers, its impact is minimum if compared with other components in the signal chain. Just the dynamic range derived from a 24 bit system goes beyond our very basic human needs.

Logical selection of amplifier design is based on Class D, simply because of the power density and efficiency. Their performance is close to high performance analog systems because of the latest improvements, and more and more systems are moving to this topology, lately extending even to audiophile level systems.

Good Class D solutions use global feedback, offering ultra-low noise and distortion levels with a very decent operating bandwidth.

From our perspective, one of the biggest differentiators in Class D systems in the market is the control loop. This is where the secret sauce is. Our latest JBL active speaker systems uses the AXIGN solution, which is a customizable controller that offers digital input and global digital feedback.

The stability/control loop is one of the critical areas not only of an amplifier, but of any other control system. It determines how the system will react to the stimulus signals and/or loading. Imagine a system that is correcting in real time with frequencies up to 20KHz or higher. In this case control loop is tailored to provide an adequate/stable system response over the specific bandwidth.

Important note: since the amplification stage has digital input, acting as a final DAC, if using a digital source no other conversion stages are in between. Digital inputs will offer the best performance.

Final comments
As we move forward, new processing and amplification technologies are raising the bar in product performance and opening the door to implement new ideas/concepts which at the end translate as a direct benefit to listener. HARMAN Engineering is constantly working on how to bring those benefits closer to listener, but also it comes with a responsibility. This is where years of experience play a big role in applying technologies without crossing the very fine line where the source material is altered in a way that changes the very nature of the reproduced material itself.