People wrong about transistor heating!!!

Discussion in 'General Electronics Chat' started by elecbeg, Nov 29, 2009.

  1. elecbeg

    Thread Starter Active Member

    Sep 7, 2009
    30
    2
    I posted on here about a week ago a circuit about testing a 2N3055 transistor. In it I said I was goin to put about 2A across the collector-emitter with a steady none pulsing DC.

    I had loads of replies saying that wasn't possible as it would overheat and burn out! even with a heatsink and fan on it.

    Now I decided to actually build the circuit and test it anyway with a temp probe, and with a basic small heatsink and a small fan on it, it stayed below 25 degrees!!

    I have learnt a lot from this forum, but I feel that some people are giving advice from what they have read, rather than from experience!
    It is miss leading and if it wasn't for the fact of I realised what they was saying was stupid. I never would have built it and would have passed on the wrong info to others.

    So this is just advice to all new people to the forum, there is a great deal of knowledge to be gained form this forum, and very good advice, but if you doubt what you are being told, even from several senior members, then experiment yourself just to make sure.
     
  2. hgmjr

    Moderator

    Jan 28, 2005
    9,030
    214
  3. elecbeg

    Thread Starter Active Member

    Sep 7, 2009
    30
    2
    yep, I have found out the info I needed to make it work better. But the only thing that people kept goin on about was the temp, I had already looked into that and found it wasn't a problem. As I replied several times in the thread.
     
  4. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    The transistor will not overheat because it is saturated with a low voltage drop. If it was linear then it would get hot.

    Your idea of an hFE tester does not work for a saturated transistor.
     
  5. steveb

    Senior Member

    Jul 3, 2008
    2,433
    469
    Somehow I missed that original thread. I scanned it very quickly; so, I could be missing something. It looked like you were testing the transistor at 2 Amps with a 3 ohm collector resistor and a 6 V power supply. This situation is going to put the transistor at (or near) saturation, so naturally the temperature rise will be small because power is only about 2A * 0.2V = 0.4 W.

    Is this the way you want to test your transistor? - In saturation? This seems very unusual, as does your whole motivation for doing the measurement. But, perhaps I just don't understand what your goal is.
     
  6. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    He wanted to test the gain of transistors but doesn't know that a saturated transistor cannot be tested for gain.
     
  7. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    The advice you got was based on solid science and experience. I think you are going off half-cocked. Instead of criticizing the mentors here, you might want to ask what you are doing wrong.:rolleyes:
     
  8. hobbyist

    Distinguished Member

    Aug 10, 2008
    764
    56
    If your'e using your transistor in the linear region, per application,
    then you want to determine for each transistor being used per board
    what base current would be necessary to be able to supply said collector current per usage of each individual baord in question.
     
  9. elecbeg

    Thread Starter Active Member

    Sep 7, 2009
    30
    2
    lol, talk about back tracking, on the origanal thread you never mentioned it would go into saturation!

    Infact it was mentioned many times, with the circuit I shown, that it would overheat. You should really read the origanal thread if your goin to try and take the mic.

    Its information like that, that I was after in the orginal thread, so instead of just goin on at me again, why don't you read up what was said and see that when I kept saying the temp wouldn't be a problem, I ment it.
    Now if thats cus its in saturation then fair enough, but why didnt someone say that instead of incorrectly saying it would overheat.
    So don't get your knickers in a twist, what I have said on the last thread and this one are both true, you are now giving the opposite advice from before.
     
  10. hobbyist

    Distinguished Member

    Aug 10, 2008
    764
    56

    You need to use the quote button
    so everyone knows what post your referring too.

    I don't think your referring to my last post,

    I was posting that I have a slight understanding of what you were trying to acomplish in the first place.

    So who'se post were you referring too???
    please use the quote button.

    So we all kbnow who youir replying back too thankyou..
     
  11. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    If your transistor is typical, it should saturate, or at least have Vce low enough that it will not get very hot. If it is at, or maybe slightly below, the low end of the beta spec, it could dissipate 3 Watts maximum. If it were me, I would use a heat sink that will keep the transistor from overheating if it has low beta. Beta for worst case dissipation is about 17 @Vce≈3V.
     
  12. maxpower097

    Well-Known Member

    Feb 20, 2009
    795
    388
    Just because you can do something doesn't mean you should. In a hobby environment it may work fine and be perfectly acceptable to push limits and work on the edge like that. But there are many people here getting advice on commercial applications where failure is not an option and the part may need to function for years and years. The part may work fine on your bench but thats no good when you manufacture 10,000 boards that all burn out in 6 months. So think about when they say you can't do that, there may be more to it then you just physically can't do it.
     
  13. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    IMHO, it is acceptable to give different advice to hobbyists than to people who are designing commercial products. I'm an old retired EE, with years of design experience on products that went into production, and I would never have done a test like this on a component that was going into a product. In fact, I don't ever recall needing a beta tester in my work. However, I did things very similar to this when I was just beginning to learn about transistors, and it helped my understanding more than weeks of reading books would have.

    EDIT: After rereading your post, I don't think this contradicts what you said. It's just a little additional perspective on it.
     
  14. studiot

    AAC Fanatic!

    Nov 9, 2007
    5,005
    513
    This thread has a disappointingly nasty flavour for no good reason.

    I warned about possible overheating when asked about a transistor tester.

    I have to say that I understand the purpose of transistor testers is to test transistors, perhaps many transsitors.
    So I assume that there will be a range of gains measured, perhaps including some faulty transistors.

    The 3055 is not a noted star performer, gains from below 10 to maybe 50 if you are lucky being available where several amps of collector current are being considered.

    The thing to do is to calculate the effect of such a range of gains, something I assume to be within the capability of someone contemplating measuring said gains.

    I have done this and broadly agree with Ron.
    For a 6 volt supply in the active region the base resistor will supply 61 milliamps.
    Using this figure I have tabulated the resulting collector current and power dissipation for various gains.

    Gain - Ic - Power (watts)

    05 - 0.305 - 1.55
    10 - 0.610 - 2.54
    20 - 1.220 - 2.85
    30 - 1.830 - 0.93
    50 - not attainable

    You can see from the calculations that a low gain 3055 will operate in the active region, an abnormally high gain one will be pushed straight into saturation as it cannot develop the current through the collector resistor. A simple application of ohms law will show at what current the voltage across the 3 ohm resistor would exceed the supply ( obviously impossible)
     
  15. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    As most know, maximum dissipation for an active transistor with a resistive load, at steady state, will occur when Vce=Vcc/2, and will equal ((Vcc/2)^2)/Rload. That's why I said previously that our OP's transistor will dissipate a maximum of ((6V/2)^2)/3Ω=3 Watts.
     
  16. studiot

    AAC Fanatic!

    Nov 9, 2007
    5,005
    513
    I also wondered, right at the outset, but gave credit for the OP providing a sufficiently beefy supply.

    But all real supplies have internal resistance and if the 6v supply struggles to provide the necessary amps this will distort the situation, driving a transistor of even low gain into saturation.

    I attach further calculations showing what happens if the supply has an internal resistance of 1.5Ω - even a gain of 20 will cause the transistor to bottom. In this case the max current drawn barely gets over 1 amp.

    So it is possible that the transistor did not warm up simply because the supply internal resistance made it impossible for it to operate properly.

    We are not told about the supply or the actual measurements, which would confirm or deny this.

    These sets of figures concur exactly with Ron's, although I haven't tabled the case of Vce= exactly half Vcc.
     
    Last edited: Dec 1, 2009
  17. elecbeg

    Thread Starter Active Member

    Sep 7, 2009
    30
    2
    Thank you all for your replies, I am sorry that this thread does/did have a bad vibe to it. Rather than bring that back up as to why, I would like to try to answer some questions.
    The supply I am using at present is a shop bought variable power supply with voltage and amp digital meters on.
    The supply I will be constructing in the end will depend on how everything goes.
    Yes, the reason that I am building it is to measure different transistors, the reason for this is that I want to experiment with power going through multiple coils creating magnetic fields. I want to build a tester so I know what the presice gain is for each different transistor, so I will be able to better match the power going through the multiple coils. Hope this makes sence.

    I am still reading and absorbing the rest of the information posted, if I have any questions about them I will ask, otherwise thank you for the info.
     
Loading...