This review page is supported in part by the sponsors whose ad banners are displayed below

For the full explanation on the F6, read this PDF. Or as Nelson put it on his forum at DIYAudio, "the IRFP240 will sound better in the 'new' topology because it is largely the same. The clear indication here is that for this specific circuit, high drain-resistance Fets are generally preferred. But I would not assume that this would apply to every listener and his system. The differences in topology are mostly different voltage regulators. Other differences are simply part values. Other transistors that would give comparable gain, distortion and bandwidth are the UJN1208K and IXTH20n50d. Although the latter is a depletion mode part, it will self bias to the correct current with a little ballast resistance. The UJN part will require substantial negative bias voltage." In toto, the production F6 is still a 25/50wpc into 8/4Ω push/pull stereo amp with Jensen input transformers for phase splitting and local feedback*. For formal production the SemiSouth enhancement-mode power JFets had simply gone International Rectifier FP240 Mosfets.

* On how these transformers differ from what's in the M2, Nelson had this: "The input transformer on the F6 provides no voltage gain which instead comes from the output devices operated in Common Source Mode where you get both voltage and current gain. The transformer simply provides a mechanism for driving the outputs with isolated signal with the proper phase; and processes the feedback from the output of the amplifier. This was the same technique as with the Sony SIT amplifier at the Burning Amp Festival - except that there was no feedback in the SIT amp. In the M2, the transformer (actually auto-former) provided all the voltage gain, drove output devices as voltage followers and it too operated without feedback. Interestingly, the two transformers (M2 and F6) required different core materials to get their sounds. The M2 has an Edcor M6 steel core and the Jensen in the F6 is 50% nickel. The primary difference is that the steel has a high tolerance for saturation, the nickel alloy gives higher permeability and lower distortion at the lower levels required by the F6. It is a case of horses for courses..."

This happened 1.5 years after the DIY version first published. Now the circuit had some time under its belt. It had grown legs. As such the F6 struck me as the long-awaited fully matured replacement for the discontinued F5. The latter's equivalent power rating had made it one the best choices for conventional speakers in the special-ops stable of FirstWatt. Using an input transformer, the F6 of course also borrowed from the earlier M2. That made for a bit of friendly incestuousness between siblings.

A DIY-version F6 from Nelson's own hands.

About that incestuousness, the glue which binds the various FW amplifier circuits together is naturally the vision of their designer. After his very many years in the field, there can no longer be much leeway for his ideas on what good sound is. Hence thinking folks would expect that despite their different circuits as discrete routes of ascent to the peak, all amplifiers in this catalogue would have to arrive more or less in the same place when used as intended. On that note the F6 embeds a very fortuitous aspect. Within the FirstWatt lineage, it was preceded by the $10'000/pr SIT mono and equivalent $5'000 SIT stereo models. Those came first. At least to these ears they set a new standard for the brand.

The DIY F6 had already emulated them to a high degree quite as though it were directly imprinted by the SIT precedents. Wasn't the final F6 predestined to do that even more? If so, it'd pull that stunt for less coin than the exotic static-induction transistors yet produce significantly more useful power (50 Mosfets watts into 4 ohms versus 8 SIT watts). Could the F6 be the poor but power-hungry* man's SIT? I had the right hardware on hand to find out.

* Obviously that's relative. For fancy face plates, hearnia-inducing bragging rights and high power, FirstWatt has always been and continues to be entirely the wrong brand. That's what Pass Labs is there for. But speakers of average 90dB sensitivity played in a standard-size room to normal not silly levels should do fine off the F6's power rating.