I have built and used this for a while. But my voltage ripple is not what it should be (I think it should be like a couple mV). Without a load the ripple it's about 70mV at a few hundred kHz. With a 12V and 1.65Amps light bulb the ripple is about 200-250mV at 50Hz.
As a modification from the original schematic, recently I have included 2 x 2N3055 in parallel with 0.33 ohms in their emitors and the filter capacitor is a 6800uF Nichicon.
Also the short circuit protection might not work correctly as after a few shorts the 2N2219A got burnt. But it could also be due to the fact that initially I had only 1 x 2N3055 on a very puny heat sink that melted my case with only a load of a few watts and apparently "provided" a short circuit between the pins of the 2N3055. After that I have mounted the pair of 2N3055 on a serious heat sink with proper isolation and now it holds easily the 3Amps with the 3055 tranzistors barely getting a few degrees hotter.
I probably should add that besides the power supply itself, the same transformer secondary powers in parallel (I had no room for a second transformer) a very simple LM317 auxiliary power supply for the panel voltmeter and a 12V fan.
Can anyone tell me what am I doing wrong regarding the ripple?