2SC3281/2SA1302 replacements

Discussion in 'The Projects Forum' started by Blatboy, May 4, 2014.

  1. Blatboy

    Thread Starter Member

    Jun 11, 2012
    46
    0
    Greetings. I'm building a power supply, using the book "TAB Electronics Guide to Understanding Electricity and Electronics" by G. Randy Slone. (Note this thread, as someone else here was also using this book.)

    the power supply calls for 2SC3281 and 2SA1302 transistors, which don't seem to be in production any more, and I've been reading about counterfeits that don't work.

    That said, (and obviously I've very new to electronics here) what would be a good replacement for that pair?

    Thanks so much for your time.
     
  2. #12

    Expert

    Nov 30, 2010
    16,257
    6,763
    NTE (previously ECG) is a good place to look for replacements. Not good on price, but simple to use.

    http://nte01.nteinc.com/nte%5CNTExRefSemiProd.nsf/$all/64E85DD67A0CFCE585257910007ECEFA?OpenDocument
     
    Blatboy likes this.
  3. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    (from my trusty old semiconductor reference book);

    2SC3281 silicon NPN HiFi-NF-E 200/200v 15A 150W 30MHz TOP-3 flatpack (equiv part 2SC4029)

    2SA1302 silicon PNP HiFi-NF-E 200v 15A 150W 30MHz TOP-3 flatpack (equiv part 2SA1553)

    They look like a matched pair of 200v 15A 150W flatpack power trannies.

    You might have some good luck with the MJE series parts.
     
    Blatboy likes this.
  4. Blatboy

    Thread Starter Member

    Jun 11, 2012
    46
    0
    Wow. That was fast. Thanks!
     
  5. Experimentonomen

    Member

    Feb 16, 2011
    331
    46
    STOP! Be very weary with NTE parts, they are known to be remarked devices that does not match the original parts spec what so ever.

    The proper replacement parts are MJL3281 and MJL1302 from Onsemi
     
    Blatboy likes this.
  6. Blatboy

    Thread Starter Member

    Jun 11, 2012
    46
    0
    oops... hmmm. Well, the deed is done. I move fast. It's a character flaw, or the fact that I'm where the town where the phrase "New York Second" originated...

    Hopefully I'll be ok. It's for a power supply...
     
  7. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    Don't forget the HUGE heatsinks and fans. :)
     
  8. Blatboy

    Thread Starter Member

    Jun 11, 2012
    46
    0
    Yeah, I'll be attaching it to the aluminum chassis for heatsinking... I also picked up some mica and thermal compound. I think I got the wrong form factor for the mica, but I bet I can make it work...
     
  9. THE_RB

    AAC Fanatic!

    Feb 11, 2008
    5,435
    1,305
    The metal chassis is not "heatsinking" when you are using 150W 15A power transistors! :eek:

    What output current will you be running and what is the linear voltage drop (difference between Vin and Vout)? That can be used to calculate the worst case heat dissipation in the power transistors.

    If you post a schematic diagram that will help a lot. :)
     
  10. Blatboy

    Thread Starter Member

    Jun 11, 2012
    46
    0
    Ok... I'm new to this..and am more or less going by "the book." The book, and it's late author, seem to be pretty good, so I feel I can trust it...

    This does give me an opportunity to try to explain what the project is, which may be better for me than for you. In the end, the question of whether using the chassis for heat dissipation is wise or not may be overshadowed by my obvious ignorance and the fact that I'm sure that there are better designs for a power supply. This design seems good for learning though. (There is a chapter later in the book with some mods to allow for better voltage regulation...if I'm understanding correctly.)

    I'm not 100% sure I'm giving you the exact info you're asking about... and I'm sure my language/jargon will be off...but here goes:

    The unit will provide two independently adjustable outputs, positive and negative of about 4V to 15V each, as well as a ground. It goes into current limit mode at about 1.5A.

    Here's my attempt to describe what happens with my limited knowledge:

    Once the 120V AC signal has gone through the transformer, bridge rectifier, and filter capacitors, we get +34V and -34V raw DC lines before it goes into the (please forgive me) transistor-y section. haha. This section, regulates? the signal... also brings the pots into play where the user controls the output. Again, please forgive my ignorance.

    For the "transistor-y section":

    Materials:

    [​IMG]

    Schematic:

    [​IMG]

    Regarding the schematic... they never put the schematic of the whole project in one place in the book... unless I missed it.

    I can't imagine that a book like this would be put out if it gave unsafe information. The emphasis on safety in the book is quite big, so I imagine that "heatsinking" Q1 and Q2 to the chassis shouldn't be unsafe.

    That said, thanks again for your help...and forcing me to attempt to explain what the book is teaching... which I still need to re-read about 50,000 times. Especially that transistor chapter.

    Check, please! ;)
     
    Last edited: May 7, 2014
  11. #12

    Expert

    Nov 30, 2010
    16,257
    6,763
    The power you need to get rid of is 45 watts at the worst. This puts you in the position of learning heat flow. Resistance to heat flow is called, "theta" and this resistance is always defined as between two places. Theta junction to case (of the transistor), theta case to heat sink, theta heat sink to air, all add together. Watts times thetas = temperature increase at the silicon chip. The datasheet of the transistors will tell you most of this, like maximum temperature allowed and the heat flow resistance of the transistor.

    Long story short: You're going to be using one or two chunks of aluminum about the size of your fist.
     
    Blatboy likes this.
  12. Blatboy

    Thread Starter Member

    Jun 11, 2012
    46
    0
    ya think? Really? Hmmm. Wonder if I understood the first part of that design correctly (as far as Vin goes). That said, I've got a nice metal enclosure on order.

    You really think the book is that far off?

    Maybe I need to look and double check I understand the front end of that circuit.
     
  13. Blatboy

    Thread Starter Member

    Jun 11, 2012
    46
    0
    [​IMG]

    F1 is 250V slo blow fuse
    BR1 is 6amp, 200V PIV bridge rectifier
    T1 and T2 are 120V primary, 24v secondary 2A transformers
    C1 and C2 are 4400-uF 50-WVDC electrolytic capacitors (I'm using 4 2200 uF...two each in parallel)

    obviously, the side of the schematic with C1 and C2 isn't entirely correct in the final design, which was shown in the earlier schematic...
     
  14. studiot

    AAC Fanatic!

    Nov 9, 2007
    5,005
    513
    Well that design is similar to buying a pair of size 12 hobnail boots for a 5 year boy.

    The 2SC4029 / 2SA1553 pair will work just fine in the circuit. They offer similar current gain to the originals (around 100 at 1.5 amps so will require feeding 0.015 amps).
    However the regulation will not be efficient because the design is very simple.

    What is happening is that a great deal of voltage is being thrown away to achieve the end results.

    The worst power case will be when the power supply is delivering minimum voltage at maximum current.
    Yes I did say minimum.

    So you have 34 volts in minus 4 volts out time 1.5 amps = 45 watts dissipated (lost) in the output each output transistor, making 90 watts in all.

    Have you ever felt the temperature of a 100w greenhous heater?

    Your chassis will get this hot under these conditions.
     
    Last edited: May 7, 2014
    Blatboy likes this.
  15. Blatboy

    Thread Starter Member

    Jun 11, 2012
    46
    0
    Interesting. I wonder if the mods he has later in the book deal with that. He does mention that it's a basic design...
     
  16. Blatboy

    Thread Starter Member

    Jun 11, 2012
    46
    0
    Later in the book he talks about replacing Q3 and Q4 with Darlington pairs as one possible improvement.

    Another improvement he talks about involves this schematic, showing only the positive regulator...

    [​IMG]

    This seems to be more about regulation than keeping Q1 and Q2 from, er, working to hard/hot, correct?

    Here's two more improvements he mentions, with a final one being a 4 output option:

    [​IMG]
     
  17. bertus

    Administrator

    Apr 5, 2008
    15,646
    2,344
    Hello,

    Have a good look at the transformers connections.
    I made some remarks in the schematic you posted:

    [​IMG]

    Bertus
     
    Blatboy likes this.
  18. Blatboy

    Thread Starter Member

    Jun 11, 2012
    46
    0
    yeah... that's because he's using 2 24V transformers. (which obviously you know, but I'm reiterating in order to be sure I understand as well.) He makes a point to really show how to wire them so they are in phase.
     
  19. Blatboy

    Thread Starter Member

    Jun 11, 2012
    46
    0
    he mentions that fig 12-3 is the "best method of improving the lab power supply" and to me looks doable... would that help the issue?
     
  20. studiot

    AAC Fanatic!

    Nov 9, 2007
    5,005
    513
    I assume the book also explains some of the circuits?

    Building a regulated supply as a learning exercise is a good way to understand much electronics and you (hopefully) end up with a useful piece of kit.

    I have certainly built more PSUs than another other type of circuit over the years.

    The problem is this.

    The gain of the pass transistors is low (which is why Darlingtons are recommended as an improvement) and the supply differential amp gain is also low (Which is why he recommended an op amp).
    To compensate for this the raw input supply is high so that Q3 and Q4 can draw enough current through R1 and R2 to supply the pass transistor base drive and and also their own needs.

    So in my worst case the raw supply is 54 watts per rail, of which 6 watts is delivered to the load, making the efficiency (6/54 x100)% = 11% efficient.

    The more you can lower the raw dc input the more efficient you can make the unit, but the more sophisticated (=high gain) the regulator design needs to be.
     
    Last edited: May 7, 2014
    Blatboy likes this.
Loading...