AMPed


Still noteworthy, this is an audio amplifier I built from scratch back around 1988.

Based on the Hafler DH-220 but modified, the unit uses additional paralleled MOSFETs in its output stage and a higher rail voltage. The project began after I came across two 220 to 120 VAC step down isolation transformers which when connected to 120VAC made a fairly decent split rail power supply at +/- 85 VDC !! The original Hafler supply rail's were +/- 60 VDC.

I hand drew photographic negatives and etched the copper PCBs myself; collected components from various parts suppliers like Digikey and Newark; acquired plate and extruded aluminium for the base and sides. Overall weight is about 25-30 lbs.

AMPed 2.0 (~2010)
Semi-recently, I gave the amplifier a much more capable power supply. While the isolation transformers served well for almost 20 years, they were a little short on output power. I purchased an off-the-shelf 800 VA toroidal transformer on-line and removed 12 windings from the secondaries in order to match the output voltage of the isolation transformers that I had been using. It was the only size-able transformer that would fit into the original case cleanly, unfortunately the output voltage was too high. The modification to the new transformer went as expected and the insulated wrapping was re-wound to look and function as good as new.

Like Hafler, the amplifier still retains a slightly warm tube-amplifier-like sound quality.

AMPed 3.0 (2013)
More recently, I thought I'd give the amp a few updates: new Lateral MOSFETS and over temperature switches. Sounds simple enough but with the slow collection of lab equipment I was able to load and sample the audio output after the update. To my surprise, the output at certain levels would carry high frequency noise sporadically. A few selectively placed decoupling capacitors now seem to keep the amplifier stable. There aren't many choices when it comes to selecting the required Lateral MOSFETs but there are still some out there. Newark Electronics stocks a number of them.

So, it's always been a bit of a mystery as to the actual output power rating. I knew the output was clearly adequate and had a rough idea but now with the passing of time was now able to measure it. I had an old adjustable power resistor that seemed suitable to load one channel of the amplifier and was able to sample the output while feeding in a 1 KHz sine wave. Not bad... in excess of 225 watts RMS. Yup, over 120Vp-p. Since only very little of the ceramic power resistor was used to maintain the eight ohm load, after only a few seconds of full load testing smoke would start to rise from the load and after testing, that portion of the load resistor had a white oxidized coating.