Announcement

Collapse
No announcement yet.

Choosing replacement output transistors(2sc3856,2sa1492) and possible fakes

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Choosing replacement output transistors(2sc3856,2sa1492) and possible fakes

    I'm dragging out an old project: a Sony STR-AV970X stereo amp that I "fixed" a dead channel on a couple years ago by replacing the output transistors (2sc3856,2sa1492) with NTE equivalents. Ever since it's worked but had a funny distortion. Checking with scope and a signal gen (which I didn't have back then) a clean sine wave in becomes round-on-the-top/sharkfin-on-the-bottom in the power amp. I didn't know to replace all the drivers and power resistors then so I'm going to do that now, but also...I think it might be a good idea, from what I hear, to scrap the NTEs.

    Which brings me to the real point of my post: what to replace them with? I found these http://www.electronix.com/catalog/pr...oducts_id/3204 but it seems suspicious that these out of production transistors that I can't find anywhere else, and couldn't find 2 years ago, are available here for $1.50. I'm guessing there's a readily available part that will work just fine, but am still a little unsure when choosing subs for bigger power transistors. Are voltage, current, and power ratings, hfe and package all you need to know to pick a substitue?

  • #2
    MCM has them, Sanken brand, in stock for $3.19 and $3.39 respectively. I have always had good luck with MCM, and don't recall getting a bum part from them ever.

    www.mcminone.com

    B&D Enterprises also has them in stock for about 50-60 cents less. I have not done a lot of biz with B&D, but so far I got good parts. In fact the only thing I ever order from them is the higher power semis.

    www.bdent.com

    Unlike tubes where the more or less equivalent 6L6 sounds different from the EL34, power transistors can be rather liberally substituted. SO yes, ratings and package pretty much sum it up.

    I rant against NTE all the time, but the reason is that the parts are NOT exact replacements, they are approximates. ANd in a power amp where the parts are expected to share current, they MUST be the same. SO when I see an NTE part on a row of something else - Sanken perhaps - I know that that odd transistor will either be trying to hog the current from the others, or it will not be turning on when they do, and so might as well not even be there. If you just HAVE to use NTE, replace the whole row of parts, not just the bad one.

    I am on Electronix mailing list too. They have always impressed me as one of those cheap flea market kinds of places - sort of like a dollar store. I would not buy from there for this.

    That said, I would tend to doubt it is the output xstrs causing this distortion. That is just an opinion sight unseen, of course. I would be looking for damage further back from when the thing blew in the first place. Check the drivers, and all resistors associated. Also try disconnecting any limit xstrs sampling off the emitters of the outputs. A lot of times distortion is introduced by the front of the power amp trying to compensate for the late stages inability to produce. In other words, for example, if the positive side cannot rise over a volt, then the feedback circuit will make the input stage increase the gain on that part of the waveform to compensate. The results can look pretty weird.
    Education is what you're left with after you have forgotten what you have learned.

    Comment


    • #3
      Thanks Enzo. I guess I didn't know what the issues with NTEs were, just that they had some.

      That said, I would tend to doubt it is the output xstrs causing this distortion. That is just an opinion sight unseen, of course. I would be looking for damage further back from when the thing blew in the first place. Check the drivers, and all resistors associated. Also try disconnecting any limit xstrs sampling off the emitters of the outputs. A lot of times distortion is introduced by the front of the power amp trying to compensate for the late stages inability to produce. In other words, for example, if the positive side cannot rise over a volt, then the feedback circuit will make the input stage increase the gain on that part of the waveform to compensate. The results can look pretty weird.
      I will definitely look into those things. Thanks again.

      Comment

      Working...
      X