At the power supply input there is a large "cap farm" with 24V to 35V 1,200F caps for an input capacitance of 28,800F. The power supply switching is handled by no less than 20 large TO-247 case-size Mosfets, switching the inputs of two very large toroids at about 25kHz. The energy output of the toroids is stabilized and stored in another huge capacitor farm, this time using 20V to 200V 1,200F caps, for a total secondary capacitance of 24,000F.
On the signal side of things, I was pleased to see high-quality, low-tolerance parts used throughout the input stages, and even though this is a bandwidth-limited amplifier, the caps in the circuitry were very good and easily could have been used in a good full-range amp. A lot of manufacturers cut some corners here, but not in the Goliath. Nice.
The stereo input signal is summed to a mono signal and passed through the gain section, an adjustable 15Hz to 36Hz -24dB subsonic filter and the -24dB crossover, which can be set anywhere between 15Hz to 250Hz. Then it goes through an adjustable phase control, which has the ability to provide 144 degrees of adjustment at 100Hz. The last stage of signal control before it's passed on to the amplification stage is a 0 to +18dB EQ circuit which can provide as much as 18.5dB of boost at 45Hz. The amp comes with a remote mounted box that provides control of the EQ circuit from the driver's seat.
Amplification is handled by 24 large Mosfets with a modulation frequency of about 118kHz. These TO-247 case devices leave no doubt as to the current capability of the Goliath. The output signal is extensively filtered with four large toroids and another set of capacitors and sent to the output terminals by a pair of relays, each capable of switching 30 amps of current. The signal travels from about mid-point on the board through a pair of 8-gauge cables. In an amp that has evidently had a lot of care and attention paid to the PCB layout and parts selection, these cables were strangely routed adjacent to the toroid and switching devices, allowing noise to be radiated back into them. Now I realize that these devices are switching at ultrasonic frequencies and the amp will be hooked to a woofer, but the noise does show up on the test bench, even with extensive pre-filtering applied. Can you hear it? Probably not, but it is there. I'll let you decide if it matters to you.
PerformanceWhen I began to measure power levels, I found the power at 1.0% THD+N was a little lower than quoted in the specifications. After double checking all my measurements, I realized that Hifonics does not qualify their power figures with a distortion limit. By letting the amp run a little into clipping, I could get the claimed power from it, but at distortion levels around 2% to 5%. At no point during the power testing did the amp complain or show any sign of stress - it just kept cranking out gobs of power.
I ran frequency response plots and found the response of the Goliath to be within 1.0dB from 18Hz to 195Hz. These limits are caused by the amps filter features: The subsonic limits the low end, and the crossover controls the roll-off at higher frequencies.
As you'd expect with a Class D amp, the slew rate, damping factor and S/N ratio are not exceptional, although still decent for this topology. I measured a slew rate of just over 2.5V/S and damping factor was high for a Class D amp at 157.5. Idle current was pretty high at 8 amps, but considering the amount of devices in this giant, not out of line at all. When you put a 5,000-watt amp in your trunk, you're probably not too concerned with overall system efficiency. With the Goliath, it seems you can have your cake and eat at least some of it, too. At full power into 1 ohm, it was an average 70%, but at full power into 2 ohms, it jumped to 82%. That's a really good number from such a large amplifier.
There is a lesson here for all you bass heads: Buy an amp large enough to give you the power you need without having to push it to its design limits and you'll be rewarded with a cooler running, better sounding, more efficient system.
I did notice that it took about two seconds for the amp to stop playing after I removed the remote turn-on voltage, not a problem in and of itself, but it could cause a turn-off noise in your system if you have a source upstream that shuts off before the amp. On the bench, the amp exhibited no turn-on or turn-off noise.