Amplifier Specifications and Testing

What makes a good amplifier?

This is an interesting question, and quite often it can arrive at having more to do with what can be measured than what matters.  In the case of the DickinsAudio application and design, we targeted an amplifier that was ‘fit for purpose’.  A key goal was to achieve a cost target that made high channel count spatial sound experimentation more accessible.  Rather than racks of equipment costing tens of thousands, we wanted to achieve 10x more scalable and cost-effective.  But what are the requirements for this?  In addition to being the complete system from network to speaker cable, the following early targets are what led to the design and achieved specifications:

  • Low-Cost targeting much lower cost than separate Dante, Conversion and Amplification Gear

  • High channel count convenient for 16 – 512 channel systems

  • Compact around 1 litre or 100mm x 100mm x 100mm in size

  • Easy Cabling custom cabling with less effort and error in connecting to array

  • Output Power perceptually loud without distortion in room-sized arrays

  • Sound Quality clean sound with low distortion

  • Quiet Operation ideally convection only with no fan or other noise

  • Safe no high voltage or excessive temperatures

  • Fault-Tolerant protected from and clear indication of output shorts, over temp and under power

  • Flexible Power operation from PoE, DC wall plug, laptop brick or dedicated power – 9-24V, 20 – 600W

  • Scalable able to daisy chain for more channels

  • Robust suitable for portable use, quick setup and long service life

After 4 design iterations, the amplifiers are now here and full specifications are provided on in this BROCHURE.  However, it is not enough to work from design specifications.  We seriously tested and stressed these amplifiers.  Faulty units or restricted functionality is just not good business, and there is also engineering pride at stake.  In addition to soak testing with actual arrays, here is a photo of one test bench, and some of the interesting (and fun) tests we carried out.

  • Channel Load individual channels sustain 30W into 4 ohm (30V peak to peak output)

  • Bulk Load amplifier stacks with all channels running sustain 90W (laptop brick) fanless

  • Extended Load stable operation at 400W total into the stack – requires an additional fan

  • Thermal Shutdown with limited ventilation, amplifiers shut down as required

  • Current Overload short circuit or low impedance loads

  • Extended Cable operation remains stable with longer cable runs

  • Insufficient Power when using PoE or inadequate power supply, system remains stable

  • Meltdown Tests with restricted ventilation and pushed to power limits, retain structure and safety

 

Previous
Previous

Spatial Microphone Calibration – ICASSP 2020