Can anyone advise on replacement for a 2SK105H transistor please?

Hey all - I'm wondering if anyone can help me find a replacement to this transistor please. If this isn't the right subforum to post this, please let me know. Thanks!

I'm trying to develop my repair skills, and am doing some repair work on an 80s integrated amp; this transistor has tested faulty, and the service manual says it's a 2SK105H. This is apparently obsolete, can anyone tell me what a modern equivalent would be please? Any help appreciated!!:D

WOW..what a beautiful beast...I like it, a lot!
I was sure I had a service manual for it, but's also impossible to find it online for free. Can you upload it somewhere... even temporarily, I'd really like to take a peek inside...
Out of curiosity, do the J-fets (marked in red) in the shunt regulators have a problem?


  • NEC A-10_shunt regs.png
    NEC A-10_shunt regs.png
    366.6 KB · Views: 45
Of course, here's a link to my copy! I processed it for basic OCR.

And you're quite right, it is TR161 on the +voltage side that (hopefully) is the culprit. The -voltage works fine, but the + side draws so much current that it blows resistor R501 on the power board (essentially the last component before power goes to the phono board).

This is basically a learning project for me to improve my diagnostic skills.

Thanks for the help, much appreciated! Everything on the amp works except the phono stage, and it sounds wonderful. It's a keeper!! :D
Thanks for the service manual...and you're right, it's a keeper!
If you are absolutely sure that only TR161 is broken, you can put an ordinary resistor instead of it as a temporary fix... measure the voltage across TR162, divide by approx. 4mA and you will get the required resistance value so you can see if the positive regulator is working as it should.
Thank you for the tip; I'm not absolutely sure though..... it'sjust a theory (plus my transistor tester doesn't consistently test it as good). What I thought I'd do today is grab TR162 from the - side (and remove the fuse resistor from this side to cut power) and fit it to the faulty side to see if it cures the problem.......
I don't think that's a very good idea because if there's a problem somewhere else, the J-fet will act as a fuse so you could blow too. Measure the reference voltages across all components on the +/- side of the regulator, write it down on the schematic and post it here. I have to go to sleep now.😴
  • Like
Reactions: 1 user
Thank you, I won't try that! I took these voltages just now (pic attached), please let me know if I need to grab something else. The problem is that on the bad side, I only have a few seconds before the resistor on the power board overheats and blows. However I have a pile of resistors, so if you need me to get through a few I can!!

If I disconnect the R151 fuse resistor, the voltage at the + side input jumps up to 50.1v.

If it helps, I've already tested these components:

C153: Ok (in spec)
C155: 200mf (should be 220mf)
C159: Ok (in spec)
D153: Ok
R151: Ok (in spec)
TR151: tests as PNP
TR155: tests as 2 diodes
TR161: unknown (can't get consistent reading)
TR162: tests as 2 resistors

I also tested the + and - input pics to ground. I get 17.14 KΩ on the + side, and 6.3 MΩ on the - side. I'm assuming that this means there's no dead hsort to ground??

Thanks again for the help, and apologies for my lack of knowledge! Learning all the time......:)

Ok, let's try something...instead of TR161, put a 5k9/0.6W resistor and a multimeter in series with the resistor in such a way that you solder only one end of the resistor (place it vertically where the gate-source connection is), wrap the other end around the probe, place the other probe where the hole for the drain J-fet is. Set the multimeter to the mA range, turn on the PSU, carefully monitor the current that will flow through the resistor in the first few seconds. If the current through the 5k9 resistor is up to 6mA, nothing smokes or burns, and the voltage at the output of the positive regulator is approximately the same as the voltage at the negative regulator, you have found where the problem is. If the current through the 5k9 resistor exceeds 6mA, turn off the power immediately, the problem exists elsewhere.
Last edited:
  • Like
Reactions: 1 user