I updated my version of the PCB with a combined footprint for TO3 and plastic transistors:
Attachments
Hello!
I'm very interested in what happened to the BUX20 transistors that were bought here?
Did you manage to determine the authenticity of these transistors?
I wrote a message to the BUX20 seller on eBay asking about the authenticity of the transistors.
The seller replied:
"Hello, they are genuine new old stock - ex-military stock, the tag is from NATO."
Hi!
Thank You very much for Your reply!
About BDY58, I read your messages in a parallel topic.
BDY58 is in the purchase plans.
In total, for testing in the clone NAP250/2 there will be the following transistors:
1. MJ802 Motorola
2. KD503 Tesla
3. BDY58 Comset
There is a BUX20 transistor in the Comset catalog, but Halfin does not have it.
I've asked Comset about production plans for the BUX20, but haven't received a response yet.
I would also like to know where to find such brackets. Thanks in advance for the answers.
Hi!
Thank You very much!
15003 - not my choice.
15003 really didn't like in my NAP140 prototype.
On paper, the MJL4281AG looks great!
But, unfortunately, delivery to Mouser 29-Jan-24
Have You heard about the online store buerklin.com?
Can he be trusted?
BUX20 in stock!
But!!!
What is this manufacturer - DSI?
According to the description - everything is fine, Germany. Production, according to Buerklin - Taiwan.
Looking at the schematics of Naim's discrete amplifiers, you'll notice they are all Class AB and in early models, that required using fast switching power transistors to get the best possible performance with least "shoot-through" or cross-conduction effects. Since the 1980s and Naim's "chrome bumper" series models, new high linearity and high frequency audio types became available from Sanken and current versions of these are still in use today. These types are mandatory for the best in linear class AB and Onsemi group their versions as what they call "audio transistors". Don't try to substitute old general purpose, epitaxial types like 2N3055, MJ15003 etc that are limited to relatively slow, likely only class A operation, where there is no hard switching involved. Those types are still OK for sound reinforcement, PA and instrument amps but I assume none us are building Naim clones for those purposes. That's why switching transistors like the TO3 style BUV and BUX series were used by Naim in early models. Similar T0220 power switching types were used in the early NAP90/Nait series amps too.
Naim still use mostly Sanken audio transistors. Otherwise they were very expensive custom products from sources like Semelab. Some are very fast and expensive though and if availability is also a problem, you might also consider the very common 2SC5200 (genuine Toshiba) or Onsemi's MJL3281 cheaper versions. So if you are looking for good sound quality, really, slugs like 2N3055 or MJ15003 are among the worst types of power transistor to use in class AB.
Naim still use mostly Sanken audio transistors. Otherwise they were very expensive custom products from sources like Semelab. Some are very fast and expensive though and if availability is also a problem, you might also consider the very common 2SC5200 (genuine Toshiba) or Onsemi's MJL3281 cheaper versions. So if you are looking for good sound quality, really, slugs like 2N3055 or MJ15003 are among the worst types of power transistor to use in class AB.
Last edited:
Assuming yours is a general question about observing class B amplifier switching behaviour, I think for starters, you'd need a decent 50-100 MHz oscilloscope, a suitably high current, clean squarewave drive signal for the output stage under test and a suitable load (i.e. capable of dissipating at least 50% of the amplifier's maximum rated output. In other words, this type of testing won't be simple or cheap and you'll need to do a little study of the topic first.
Hard switching just means fast and furious, relative to the technology, like when you try to generate HF square or triangular waveforms at high power levels with BJT semis. This is tough not only on BJTs but also the boards and passive components, usually because the peak currents are substantially higher than average and physical damage can occur and accumulate from the rapidly changing thermal stresses unless suitable protections are in place. When it comes down to clean switching behaviour though, very fast devices like mosfets are much to be preferred for their toughness and efficiency.
Almost any worthwhile textbook on audio amplifiers or general electronics that covers BJT switching devices will also be useful in understanding the fundamental problems with crossover behaviour in class B (that includes class AB). There is a lot of scholarship and explanation out there already on the net and you can find links to respected texts out there too: For example, you can buy the book or just take a peek at chapter 4 here://nvhrbiblio.nl/biblio/boek/Self%20-%20Audio%20Power%20Amplifier%20Design.pdf
Hard switching just means fast and furious, relative to the technology, like when you try to generate HF square or triangular waveforms at high power levels with BJT semis. This is tough not only on BJTs but also the boards and passive components, usually because the peak currents are substantially higher than average and physical damage can occur and accumulate from the rapidly changing thermal stresses unless suitable protections are in place. When it comes down to clean switching behaviour though, very fast devices like mosfets are much to be preferred for their toughness and efficiency.
Almost any worthwhile textbook on audio amplifiers or general electronics that covers BJT switching devices will also be useful in understanding the fundamental problems with crossover behaviour in class B (that includes class AB). There is a lot of scholarship and explanation out there already on the net and you can find links to respected texts out there too: For example, you can buy the book or just take a peek at chapter 4 here://nvhrbiblio.nl/biblio/boek/Self%20-%20Audio%20Power%20Amplifier%20Design.pdf
Last edited:
You might need a differential scope probe to measure the base current vs emitter current under dynamic conditions.
In a simplified BJT model, zero base current should always result in zero emitter current. In reality, this is not true. The transistor has a sort of inertia and it takes time for it to reach a steady state after a change in base current. What's worse, this time varies with emitter current. Many BJTs get slower as the current decreases and I imagine the 1970s ones were generally poor.
In a simplified BJT model, zero base current should always result in zero emitter current. In reality, this is not true. The transistor has a sort of inertia and it takes time for it to reach a steady state after a change in base current. What's worse, this time varies with emitter current. Many BJTs get slower as the current decreases and I imagine the 1970s ones were generally poor.
Hello, I would like to ask where the aluminum U-profiles shown in the picture can be purchased.
Basically, the sizes are 31.75mm/31.75mm/31.75mm/4.76mm.If it cannot be bought in these dimensions, then at least indicate a manufacturer with close to these dimensions. I mean the profile for the power transistors. Thanks in advance.
Basically, the sizes are 31.75mm/31.75mm/31.75mm/4.76mm.If it cannot be bought in these dimensions, then at least indicate a manufacturer with close to these dimensions. I mean the profile for the power transistors. Thanks in advance.
Attachments
I've been looking at this thread for a while and wondering if a NAP250 would scratch my itch to replace my 110 (I've had it since the 1980's) I really like the sound and I've had a long standing urge to try a NAP250 thinking it would be the same but MORE.
I'm assuming that by now all the spare boards are gone, and I'll need to get some made. From reading, the Kicad project of Jeff's boards are at post #211 and there's nothing later. Is that correct? Also, JPK73 has made different layouts, but I don't see published gerbers for those. Correct?
I'm assuming that by now all the spare boards are gone, and I'll need to get some made. From reading, the Kicad project of Jeff's boards are at post #211 and there's nothing later. Is that correct? Also, JPK73 has made different layouts, but I don't see published gerbers for those. Correct?