Sub MJ21194 for NEC 2SD555 or not?

[SQLGuy]

Hi folks,

I've got a PS Audio 200c that I bought completely blown up. I've since repaired it - all but one of the output transistors were blown, as were a couple of drivers and a couple of bias control transistors.

Anyway, since the original 2SD555's and 2SB600's are near-unobtanium, I checked with the original manufacturer and the original assembler, and both suggested going to MJ21193/MJ21194. The original devices are 15MHz fT, the new ones are 4MHz. So, with the MJ's in there, the amp has been running fine for about six months now, and sounds fine, but I'm wondering what I'm compromising with the slower devices. This amp got rave reviews with its original outputs and also used very high speed devices elsewhere in the chain. Should I maybe be looking at the newer 2ST5949 and 2ST2121 in order to preserve more of the original character of the output section?

Here's a similar schematic to mine. Mine does not have base resistors on the output, and it has less resistance (52 Ohms?) between the emitters of the drivers.

Thanks,
Paul

[wahab]

MJ15022/15023 or MJ15024/15025 are the closest replacement parts from
onsemi..they have better Ft than the MJL211193/94 , and identical to the original
parts from NEC which are 4/7 mega for the npn/pnp respectively..

[SQLGuy]

Hi Wahab,

Where did you find these details? I looked at On's web site and data sheets. They show 5Mhz fT for the 21193's and 15022's in their selection guide (Audio Transistors) and 4Mhz for 15024's. In the data sheets they show 4Mhz for all.


Cheers,
Paul

[wahab]

oups..according to onsemi s datasheet, they are exactly the same devices..
only the casing is different..i have 4mhz typical for both..other caracteristics are the same..

regards,

[AndrewT]

Reading the NEC datasheet tells me the MJ devices will perform better than the 2s devices.
Improved power handling, improved current handling, improved temperature rating and combining these shows a very significant improvement in elevated temperature SOAR, i.e. a much more robust output stage.

BTW,
NEC are kidding us on that a pair of 10A, 200W, 150degC devices can be used to make a reliable 200W to 300W audio power amplifier.

Does your version have a 3pair output stage with Re=0r1?

Check the voltage drop (Vre) across each and every Re to see how well matched the transistors are. If the voltages vary by more than 10%, you should consider rematching them to get them to run at a more even temperature.

What is the voltage across the two base feeding resistors?

[SQLGuy]

Hi Andrew,

Yes, I have 3 pair output stage, but, as I mentioned earlier, no base resistors. The outputs are within about 15% of each other. I don't plan to pull them out and match them further, especially since PS Audio didn't match outputs in the first place. If I were doing the work again, I proabably would have bought matched triples from the Company Store, but I don't think it's worth pulling and redoing them.

I agree that the MJ21193/94's are more rugged devices than the ST's, but I was more concerned with the tradoffs of speed versus the original outputs. The ST's are, after all, more rugged than the NEC's as well.

Also, I don't think NEC was meaning that a single pair of their outputs was suitable for 200 - 300 Watts... just that these transistors would be useful in high power amplifiers. Rowland and Levinson used fairly large numbers of them in parallel in their output stages.

Cheers,
Paul