Hi,
With the growing popularity of Class D amplifiers -> use of MOSFETs, I’m wondering about the future availability of BJT transistors. Could their production eventually decline, making them harder to find , purchase? Aside from their use in high-voltage applications and certain SMPS designs, will BJTs remain readily available?
Mainly I'm talking about Power transistors , complementary NPN-PNP parts.
Additionally, what are currently considered " the best " power transistors for output stages in audio amplifiers?
I often come across designs that feature older power transistors, so I’m curious—are there any newer options on the market?
When it comes to output transistors in audio power amplifiers, how do epitaxial planar transistors compare to triple diffused ones? Is one generally a better choice, or do the other specifications ultimately dictate the decision, regardless of the transistor type?
For instance, if two transistors share the same specifications, but one is epitaxial planar and the other is triple diffused, which would you choose? Would there be any noticeable difference in sound quality or performance? From what I’ve read, epitaxial planar transistors tend to be faster and more linear, while triple diffused transistors are more rugged with higher current and voltage ratings.
- Bruno.
With the growing popularity of Class D amplifiers -> use of MOSFETs, I’m wondering about the future availability of BJT transistors. Could their production eventually decline, making them harder to find , purchase? Aside from their use in high-voltage applications and certain SMPS designs, will BJTs remain readily available?
Mainly I'm talking about Power transistors , complementary NPN-PNP parts.
Additionally, what are currently considered " the best " power transistors for output stages in audio amplifiers?
I often come across designs that feature older power transistors, so I’m curious—are there any newer options on the market?
When it comes to output transistors in audio power amplifiers, how do epitaxial planar transistors compare to triple diffused ones? Is one generally a better choice, or do the other specifications ultimately dictate the decision, regardless of the transistor type?
For instance, if two transistors share the same specifications, but one is epitaxial planar and the other is triple diffused, which would you choose? Would there be any noticeable difference in sound quality or performance? From what I’ve read, epitaxial planar transistors tend to be faster and more linear, while triple diffused transistors are more rugged with higher current and voltage ratings.
- Bruno.
Instead of asking diyAudio "tell me what to think?" you have the option to study and reflect instead. As Francis Bacon said 399 years ago, "Read not to contradict and confute; nor to believe and take for granted; nor to find talk and discourse; but to weigh and consider."
One useful book you could read, would be Audio Power Amplifier Design (6th edition) by Douglas Self (link). He vigorously advocates for the epitaxial bipolar transistor pair MJL1302A(PNP) / MJL3281A(NPN) . Study his reasons, see WHY he loves these transistors so much, and then look into your heart. Do his reasons resonate with you? Or, are you looking to optimize different performance characteristics than Douglas Self?
Become a thinker and a student, not a beggar, and understand why you've chosen the devices you've picked.
One useful book you could read, would be Audio Power Amplifier Design (6th edition) by Douglas Self (link). He vigorously advocates for the epitaxial bipolar transistor pair MJL1302A(PNP) / MJL3281A(NPN) . Study his reasons, see WHY he loves these transistors so much, and then look into your heart. Do his reasons resonate with you? Or, are you looking to optimize different performance characteristics than Douglas Self?
Become a thinker and a student, not a beggar, and understand why you've chosen the devices you've picked.
@xXBrunoXx : Most of what you have read seems out of date. Back in the day of the phase linear (at least the S2) triple diffused was far and away superior for SOA among “faster” types. And if you wanted real ruggedness, you needed Motorola’s PowerBase 3 MHz epitaxial types or a single diffused type (RCA’s “hometaxial” was one such variety). The Japanese went to war over this, and the result was each manufacturer perfecting their own recipe for “audio output” application. The result was fast types, 20 MHz and up, that would hold up as well as the old RCA 800 kHz types. And a lot less process variation. They kind of left poor old Motorola in the dust. They had their ruggedized epi-base type too, the MJ15024, but it’s now considered a dinosaur because it hit the speed limit of the process. I would not worry over what type/process it is, other than falling into the slow group or the fast group. Go by spec alone - but along with that, how far beyond spec you can usually push them. If you take the sane liberties with SOA on a Sanken LAPT that you do with Motorola/ON or Toshiba (say 200% over rated Ic at Vce for 1 second) one usually ends up blowing them out. Toshiba usually sandbags their SOA spec, Sanken is pretty spot on actual capability, and ON is somewhere in between. If you are looking for the fastest thing available on the market, right now that is the ON MJL4281. But the hFE at 10A suffers compared to 3281 (And so does the fT). Its only rated 35 MHz but really hits 60 like the fastest Sankens. And nobody can touch the Vceo. And Toshiba could come up with a triple diffused part that blows it away tomorrow. Assuming of course that there is still a market for class AB linear amps. When that happens manufacturers will keep two types each: the cheapest and the best, and everything else goes EOL. Remember how many 2N or 2S types there used to be? It’s already consolidated down to a handful.
Last edited: