I am working on an amplifier which has a bad channel. The amplifier has been repaired in the past (not by me) - about 11 years ago. The good channel has two pairs of A1186/C2837 (original). The bad channel has 2 original A1186, 1 ECG36, and one significantly smaller transistor labelled 2sc2837, but most of the ink has wiped away.
The owner said he accidentaly shorted the outputs for a second and the sound has been distorted since.
It appears that the replacement transistors have failed, I have not gotten to the drivers yet. I have found these transistors listed on Sankens web site, but have not been able to figure out if they are still in production or where to get real ones.
The amplifier is a 22 year old Onkyo Integra A8087. I would not mind replacing them with on-semi replacements if there is something suitable in a TO-3P package.
are there any proven replacement transistor pairs?
Thanks
The owner said he accidentaly shorted the outputs for a second and the sound has been distorted since.
It appears that the replacement transistors have failed, I have not gotten to the drivers yet. I have found these transistors listed on Sankens web site, but have not been able to figure out if they are still in production or where to get real ones.
The amplifier is a 22 year old Onkyo Integra A8087. I would not mind replacing them with on-semi replacements if there is something suitable in a TO-3P package.
are there any proven replacement transistor pairs?
Thanks
MCM is reliable - they are part of Farnell. The 1186/2837 is still in production according to Sanken.
OnSemi have NJW0281/NJW0302 in TO-3P packages but they are not as fast or low Cob as the Sankens. They might work acceptably, but they might need compensation adjustments to the rest of the circuit.
OnSemi have NJW0281/NJW0302 in TO-3P packages but they are not as fast or low Cob as the Sankens. They might work acceptably, but they might need compensation adjustments to the rest of the circuit.
These Sanken power transistors have high a Ft and good 100W/60pF Cob ratio.
The 100W NPN Sanken 2SC2837 Y Rank has hfe= 90-180, Ft=70Mhz, and a low COB=60pF
The 100W PNP Sanken 2SA1186 Y Rank has hfe= 90-180, Ft=60Mhz, and a low COB=100pF
I use 1-2 pairs for low wattage Class A amps.
I use 8-16 pairs for high current Class A amps.
Optimal Class-A transistion to Class-AB biasing requires ~23mv across each emitter resistor. A high current amp with optimal bias requires many pairs. The low 60-100pF Cob, vs. 400-500pF Cob for most 200 Watt transistors, allows paralleling many of these Sankens with a normal VAS stage.
The 100W NPN Sanken 2SC2837 Y Rank has hfe= 90-180, Ft=70Mhz, and a low COB=60pF
The 100W PNP Sanken 2SA1186 Y Rank has hfe= 90-180, Ft=60Mhz, and a low COB=100pF
I use 1-2 pairs for low wattage Class A amps.
I use 8-16 pairs for high current Class A amps.
Optimal Class-A transistion to Class-AB biasing requires ~23mv across each emitter resistor. A high current amp with optimal bias requires many pairs. The low 60-100pF Cob, vs. 400-500pF Cob for most 200 Watt transistors, allows paralleling many of these Sankens with a normal VAS stage.
there are some new bjt's available, e. g. "ST" and "ON"
I have listed the most populare types - go to
http://www.diyaudio.com/forums/showthread.php?s=&threadid=147656
I have listed the most populare types - go to
http://www.diyaudio.com/forums/showthread.php?s=&threadid=147656
Yep - they're going in. I bought a few spares of each so if I had to destroy some for verification I could - but I think they'll be ok.
I just got some new 2SA1306 (toshiba) that look nothing like the originals or it's new compliment. They are also missing the Hfe group code. And the date code seems to be printed wrong. And the plastic doesn't look like any other toshiba transistor I have seen - new or old. And the printing is much to easy to read, etc - I haven't tried to wipe it off yet. . .
So compared to this - the Sanken parts look totally correct.
I just got some new 2SA1306 (toshiba) that look nothing like the originals or it's new compliment. They are also missing the Hfe group code. And the date code seems to be printed wrong. And the plastic doesn't look like any other toshiba transistor I have seen - new or old. And the printing is much to easy to read, etc - I haven't tried to wipe it off yet. . .
So compared to this - the Sanken parts look totally correct.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.