Yes I've done plenty of checks and nothing is cracked - I think I'm starting to see that PCB when I shut my eyes now haha! 
15 2SC2023s on their way to me. I'll have to measure the hFE of all of them when the arrive and select the best candidates in each case:
I think Sony had special 'grades' selected for them - 'R' and 'O' grades. According to the service manual, the four switching transistors could be either all 'R', or all 'O' varieties, but Q601/613 had to be 'R' grades (those two are listed as 'R' grade page 43 of the manual).
Only one of the transistors that I removed from my SMPS was still any good. Fortunately that was one of the 'R' marked C2023s - putting it in the hFE tester on my multimeter gives me a reading of 009. So at least I know for Q601/613 if I select two from my batch of 15 with the same I'm good to go for those two positions.
Not sure about the other four - I guess just any four that have the same hFE readings...? Above or below a 009 hFE reading I'm not sure however...
- John
15 2SC2023s on their way to me. I'll have to measure the hFE of all of them when the arrive and select the best candidates in each case:
An externally hosted image should be here but it was not working when we last tested it.
I think Sony had special 'grades' selected for them - 'R' and 'O' grades. According to the service manual, the four switching transistors could be either all 'R', or all 'O' varieties, but Q601/613 had to be 'R' grades (those two are listed as 'R' grade page 43 of the manual).
Only one of the transistors that I removed from my SMPS was still any good. Fortunately that was one of the 'R' marked C2023s - putting it in the hFE tester on my multimeter gives me a reading of 009. So at least I know for Q601/613 if I select two from my batch of 15 with the same I'm good to go for those two positions.
Not sure about the other four - I guess just any four that have the same hFE readings...? Above or below a 009 hFE reading I'm not sure however...
- John
I should hopefully receive them Tuesday or Wednesday next week - it's going to be a LONG wait...
Really hope they solve the problem - if not I cannot see what else could be wrong other than that pot. All the bad components were replaced. Anyway I'm jumping the gun here - I'll see what happens when those NOS transistors arrive.
If the worst comes to the worst (and I really hope it doesn't after all this effort, and everyone's help) there's just enough space in there for me to replace the SMPS with a linear PSU.
- John
Last edited:
Those 15 Sanken 2SC2023s arrived today. Not quite in the condition I would hope hoped for however. They don't look used, but fairy grotty and all the legs are bent and haven't been supplied in their original packing. So I'm going to test them all now, then put them into batches of hFE. I hope I get some decent matched groups.
With a bit of luck, I'll then be able to finally get this PSU sorted
With a bit of luck, I'll then be able to finally get this PSU sorted
Right measured all the hFEs of the Sankens. Unfortunately none of them is close to the 010 reading of the 'R' grade Sanken (the only original switching transistor that wasn't blown) that survived in my PSU. For Q601 & Q613 the Sony service manual clearly specifies 'R' grade only...
For Q609-612 the manual specifies either 'R' grade hFE or 'O' grade. I have NO idea what hFE 'O' grade measures unfortunately...
These are the grades I've ended up with:
45, 48, 50, 50, 56, 58, 61, 63, 71, 72, 72, 80, 92, 120 and one I cannot measure as for some reason the legs have been cut shorter than the rest!
At this point I haven't a clue if any of these are suitable for the Sony PSU - none of them are close to the 10 hFE reading for Q601 & Q613 though that much is certain.
The manual does state (scan at the top of this page, post #361) that for Q609-612 use transistors within the same hFE ranks, so I guess I could take four from the 50-58 batch I listed above. Perhaps use the 45 and the 48 for Q601 & 603?
I do have a bit of a sinking feeling as all the distributors (worldwide!) I've gotten replies back from about 2SC2023s have all said none of them end in a 'R' or a 'O'. The fifteen transistors I have end in a 'Y'. It's looking like Sony got Sanken to supply special grade C2023s for this PSU possibly? What effect will my not using these grades have on the PSU?
I'm out of my depth basically and haven't a clue what hFE actually means and how it is relevant to this circuit
- John
For Q609-612 the manual specifies either 'R' grade hFE or 'O' grade. I have NO idea what hFE 'O' grade measures unfortunately...
These are the grades I've ended up with:
45, 48, 50, 50, 56, 58, 61, 63, 71, 72, 72, 80, 92, 120 and one I cannot measure as for some reason the legs have been cut shorter than the rest!
At this point I haven't a clue if any of these are suitable for the Sony PSU - none of them are close to the 10 hFE reading for Q601 & Q613 though that much is certain.
The manual does state (scan at the top of this page, post #361) that for Q609-612 use transistors within the same hFE ranks, so I guess I could take four from the 50-58 batch I listed above. Perhaps use the 45 and the 48 for Q601 & 603?
I do have a bit of a sinking feeling as all the distributors (worldwide!) I've gotten replies back from about 2SC2023s have all said none of them end in a 'R' or a 'O'. The fifteen transistors I have end in a 'Y'. It's looking like Sony got Sanken to supply special grade C2023s for this PSU possibly? What effect will my not using these grades have on the PSU?
I'm out of my depth basically and haven't a clue what hFE actually means and how it is relevant to this circuit
- John
Last edited:
Hi John... just for interest could you perhaps post a picture of one of them. Might not achieve much but I'm curious.
hfe is one of the h parameters of transistors and refers to current gain. Put simply a high hfe device needs less current to be flowing in the B-E juntion to achieve a given flow in the C-E path. It's normal practice to design so that a circuit works with any device within the manufacturers spec. The audio part of your amp would be just the same with transistors of 20 or 200 for hfe.
The transistors in your PSU (the four switchers) operate either fully on or fully off... and as this is an early discrete type PSU the only reason I can see for matching transistors is to ensure there is no "overlap" in conduction which would generate heat.
A modern design would overcome that limitation with a much more complex drive arrangement.
I would try and keep each "pair" as close as possible. Q609/611 and Q610/12
hfe is one of the h parameters of transistors and refers to current gain. Put simply a high hfe device needs less current to be flowing in the B-E juntion to achieve a given flow in the C-E path. It's normal practice to design so that a circuit works with any device within the manufacturers spec. The audio part of your amp would be just the same with transistors of 20 or 200 for hfe.
The transistors in your PSU (the four switchers) operate either fully on or fully off... and as this is an early discrete type PSU the only reason I can see for matching transistors is to ensure there is no "overlap" in conduction which would generate heat.
A modern design would overcome that limitation with a much more complex drive arrangement.
I would try and keep each "pair" as close as possible. Q609/611 and Q610/12
Hi Mooly!
Thanks for the explantion re. the hFE. Makes more sense now
Here's a pic of the transistors:
Interesting to note that the case of the "Sony" C2023 is a very dark grey, whereas the supplied ones I got are all black. I didn't show a photo of it but if you turn over the transistor there is a very tiny number embossed within a circle into the plastic. This is there on both the original C2023 and the ones I received, so I'm pretty sure they're all from Sanken.
Thanks for the explantion re. the hFE. Makes more sense now
Here's a pic of the transistors:
Interesting to note that the case of the "Sony" C2023 is a very dark grey, whereas the supplied ones I got are all black. I didn't show a photo of it but if you turn over the transistor there is a very tiny number embossed within a circle into the plastic. This is there on both the original C2023 and the ones I received, so I'm pretty sure they're all from Sanken.
Last edited:
That's certainly a possibility. It might also have been something of a custom job too though I guess.
I guess all I can do is fit 6 of them and hope for the best! So you think matching them in pairs would be best then? How does this look:
Q601 & Q613 - 045 and 048 hFE
Q609 & Q611 - 050 and 050 hFE
Q610 & Q612 - 056 and 058 hFE
- John
They look pretty closely matched.
I 've been looking to see if I could turn anything up on Sonys selected transistors but have not found anything.
Looking at what you have I might be inclined to use as high a gain as possible while keeping them matched.
Q601 & Q613 - 072 and 072 hFE
Q609 & Q611 - 061 and 063 hFE
Q610 & Q612 - 056 and 058 hFE
Q601/613 being in parallel means that they should be as closely matched as possible as there are no resistors to "equalise out" any variations (as you would in say a power amplifier with parallel devices).
I 've been looking to see if I could turn anything up on Sonys selected transistors but have not found anything.
Looking at what you have I might be inclined to use as high a gain as possible while keeping them matched.
Q601 & Q613 - 072 and 072 hFE
Q609 & Q611 - 061 and 063 hFE
Q610 & Q612 - 056 and 058 hFE
Q601/613 being in parallel means that they should be as closely matched as possible as there are no resistors to "equalise out" any variations (as you would in say a power amplifier with parallel devices).
Oh for crying out LOUD!!!!!!!!!
I powered up and.....
...managed to accurately set it to 99v !!!
Woohoo-ey! No weird noises from the SMPS either this time absolutely silent, and the VU meters light up fine too.
The manual specifies 98-99V so I'll set it to 98V now just to be on the safe side (our mains can hit 250V here often so I'd rather it was running slightly lower). All I need to do now is put the SMPS into that shielding can, put it back into the amp... and I should be good to go then!
VERY very happy!
- J
I powered up and.....
...managed to accurately set it to 99v !!!
Woohoo-ey! No weird noises from the SMPS either this time absolutely silent, and the VU meters light up fine too.
The manual specifies 98-99V so I'll set it to 98V now just to be on the safe side (our mains can hit 250V here often so I'd rather it was running slightly lower). All I need to do now is put the SMPS into that shielding can, put it back into the amp... and I should be good to go then!
VERY very happy!
- J