I have learned the hard way that speaker protection is necessary. Very much so!Speaker protection really isn’t necessary, and is not used in the commercial version of this amp as sold by First Watt.
I was away for the weekend, and forgot to turn off my First Watt F5. The system somehow developed an issue where the amp would output full positive DC rail voltage (around 24 V) to the speakers. Needless to say the woofers were toast. Worse thing was the terrible smell of burnt electronics in the house, and I am still glad there was no fire.
I added a speaker protection to this (and every other) amp since then. Burning down the house is just not an option.
@codyt,- two output transistors (one IRFP-240 for CCS, and one Semisouth for output)
- optional ground lift as SMD on bottom side of board
- 100R gate resistors on jfets are SMD on bottom side of board
- The 1.2K resistor that traditionally determines the amount of AC gain/loopback the Aleph Current Source sees is replaced by a 2K pot.
- All pots have provisions for being replaced by THT resistors once final values have been established
I just ordered the gerber files you shared. ^^
BTW, I have a few questions.
In this work, you used IRFP240, which is different from before. Is there a special reason why you changed it from 150 to 240?
And is the optimal value of the 2K trim pot, called 'Aleph Gain', just the point that sounds the most pleasant to the ear? Or is there another standard?
This is the first time I've seen a trimpot that can be turned by hand. Could you please share the part name?
Thank you
I believe I used the IRFP240 because NP had recently (at that time) used that part in the top section of his F8, so I felt comfortable using it too. That said, I've used both parts and don't have a preference. The 'Aleph Gain' pot value of 2K is meant to cover the original Aleph J value of 1K2. You can run into stability issues if you run that value too low (say below 800R). If I were doing this project over again I'd probably put 750R in series and make the pot 750R. You can probably figure out how to modify the board to make that happen, or just be attentive when adjusting that pot. I would set it at 1K2 initially, then experiment with a turn or two in each direction once everything else is dialed in. I always made that adjustment realtime while measuring the distortion, and I always preferred it very close to the stock 1K2. Of course, that's the other option – just use a resistor instead of the pot.
P1 is most likely Bourns 3362.
P1 is most likely Bourns 3362.