Sansui 9090 restoration. Need advice on matching transistors.

steve1953

steve1953

Active Member
I bought a kit to rebuild the F-2436-1 board. After some further research I found that TR01, TR02, TR03, and TR04 being at the front end of the amp needs to have matched sets. Does anyone know of a place that sells matching transistors ( KSC1845' s were the new replacements in the kit) ? Or do you just buy a bunch and try to match them yourself ? I also plan on changing all the output transistors , so the same matching problem comes up there.
 
Register to hide this ad
You can get 100 KSC1845 for less than 8 bucks and match them based on hfe - at home. I've never matched new outputs. I don't think it's required for power transistors.
 
steve1953 said:
I bought a kit to rebuild the F-2436-1 board. After some further research I found that TR01, TR02, TR03, and TR04 being at the front end of the amp needs to have matched sets. Does anyone know of a place that sells matching transistors ( KSC1845' s were the new replacements in the kit) ? Or do you just buy a bunch and try to match them yourself ? I also plan on changing all the output transistors , so the same matching problem comes up there.
Click to expand...
No need to match output transistors. As far as the input pairs, the closer matched, the more stable DC offset will be. However since DC offset is adjustable, it isn't really all that critical. The offset will still drift around because of air currents since they are separate transistors thus being slightly different temperatures, but will not hurt anything, nor be noticeable out the speakers. KSC1845 are cheap and if you buy them on TApe (KSC1845FTA), they are usually pretty consistent being in the same run/batch. 20 of them will get you enough good pairs matching them yourself for hFE and/or vBE.
 
Danny Miller said:
the 8080/9090 driver board has very delicate traces and they will lift very easily if you are not careful ..least amount of heat possible and make it fast..
Click to expand...
Yes i ran into that problem already, I had to give up using de-soldering wick and built a sucking station using a old vacumm pump to do it faster.
 

Attachments

  • DSCN7638.JPG
    DSCN7638.JPG
    99.5 KB · Views: 36
Skywatcher said:
No need to match output transistors. As far as the input pairs, the closer matched, the more stable DC offset will be. However since DC offset is adjustable, it isn't really all that critical. The offset will still drift around because of air currents since they are separate transistors thus being slightly different temperatures, but will not hurt anything, nor be noticeable out the speakers. KSC1845 are cheap and if you buy them on TApe (KSC1845FTA), they are usually pretty consistent being in the same run/batch. 20 of them will get you enough good pairs matching them yourself for hFE and/or vBE.
Click to expand...
Ok, thanks for the info. This is my first restoration and it just has to work or my wife might throw it out. Haha
 
steve1953 said:
Yes i ran into that problem already, I had to give up using de-soldering wick and built a sucking station using a old vacumm pump to do it faster.
Click to expand...
You'll suck the components and all off the boards with that thing. :rflmao: But it makes for a "clean start". :D
 
steve1953 said:
Ok, thanks for the info. This is my first restoration and it just has to work or my wife might throw it out. Haha
Click to expand...
Check the transistors in the kit. You'll probably find they will be "close enough" as long as they are on tape instead of BUlk (KSC1845FBU) As long as that aren't wildly different, they'll be fine. Note the pinout. With the flat side toward you they are ECB from left to right. What is in there may be the opposite (BCE) so watch it.
 
I've received "tape feed" reels of components (Fairchild/OnSemi) which have varied widely in gain. When I had time, I would pull, gain test, and mark the tapes for gain to save time when going elbows-deep into a project. For example, a "FT" reel of KSA992s had devices which varied from the 290s to 370s for gain. It's underscored my belief that checking is worth the time.
 
Watthour said:
I've received "tape feed" reels of components (Fairchild/OnSemi) which have varied widely in gain. When I had time, I would pull, gain test, and mark the tapes for gain to save time when going elbows-deep into a project. For example, a "FT" reel of KSA992s had devices which varied from the 290s to 370s for gain. It's underscored my belief that checking is worth the time.
Click to expand...
Yes, some batches are more consistent than others. The last 100 992s I got were pretty consistent between 416 and 425 with a lot of them matching almost perfectly in a row. A few were a little higher ranging up to 436. Only 3 were at 395/397/401, all grouped together. I test them all at the same time so their temperatures are the same for consistent readings. The "oddballs" get used where matching isn't necessary.
 
Well I have to admit I am at a disadvantage because I have a limited storage capability. Even though the components are small, reels and batches end up as piles.I had to clean out my basement just to get access to my pile of receivers. I ran across stuff I forgot I even had. I am not to organized in parts bins or inventory but I hope to fix this stuff if possible.
 
I'll second the least amount of heat to the driver board that Danny Miller commented on. I lifted two on mine while re-capping it. Although they were easy fixes, it did take time to do it right. That board was a real P.I.T.A. to do!
 
You must log in or register to reply here.
Back
Top Bottom