First Power Amplifier Build - Help Needed

Discussion in 'The Projects Forum' started by JayJones, Mar 22, 2016.

  1. JayJones

    Thread Starter New Member

    Jun 17, 2015
    15
    0
    Hello everyone,

    I'm in the process of building my very first power amplifier, and I've blown up my output stage twice already.

    I was wondering if anyone could take a peak at the attached LTspice schematic and let me know which parameter of a TIP41/TIP42 that I'm exceeding?

    I had both positive and negative rails fused with 1A fast blow fuses, so I'm thinking it's voltage related, not current but I could be wrong.

    I also had a thermocouple clamped to the case of the output transistors and they are in the 42-46 degree celcius range when running a 5vpp sine wave into the input and getting 50vpp on the output into an 10 ohm test load (resistor).

    Heatsinks are Aavid Thermalloy HS350-ND (3.7C/W) - Maybe too small?

    I originally had 0.1 ohm resistors for R9 R10 R15 R16 but have changed them to 0.47 Ohm in hopes of spreading the load more equally.

    Is there any chance I'm getting back EMF when connected a to speaker that would cause the output section to blow?

    Any help would be greatly appreciated!

    Thank you
     
  2. bertus

    Administrator

    Apr 5, 2008
    15,645
    2,344
    Hello,

    Can you post a schematic in a graphic format on the forum, like .png?

    Bertus
     
  3. JayJones

    Thread Starter New Member

    Jun 17, 2015
    15
    0
    PowerAmplifier.png
    Here's the LTspice schematic.
     
  4. bertus

    Administrator

    Apr 5, 2008
    15,645
    2,344
    Hello,

    What is the function of C9 and C11 at the output stage?

    Bertus
     
  5. JayJones

    Thread Starter New Member

    Jun 17, 2015
    15
    0
    I had added them as stabilization caps while breadboarding. They may not be required?
    Like I mentioned, this is my very first amplifier build.
    Thank you
     
  6. MrChips

    Moderator

    Oct 2, 2009
    12,421
    3,357
    If you are testing a power amp build for the first time, you need to take some precautions and test one thing at a time.
    You need to check all DC bias voltages without a load.

    - Don't connect a loudspeaker.
    - No need to test dual output pairs. Remove one pair.
    - Insert current limiting resistors in the collector legs of the output transistors.
    - Disconnect the feedback at R3 and connect R3 to GND for testing.
    - Measure all DC voltages at your transistors and verify that the circuit balances to zero volts output.
     
  7. blocco a spirale

    AAC Fanatic!

    Jun 18, 2008
    1,438
    368
    I can't see anything obviously wrong with the schematic but it makes sense to test it with a resistive load rather than a loudspeaker and if it still blows the output transistors, the back-emf theory goes out the window.
     
  8. dannyf

    Well-Known Member

    Sep 13, 2015
    1,771
    359
    Power up your circuit through a large value power resistor. Check DC working points vs. Expectation.
     
  9. Lestraveled

    Well-Known Member

    May 19, 2014
    1,957
    1,215
    OK
    Your output transistors are rated correctly. You fused the power supply rails. So, I suspect that the method of failure might be excessive base current. This is just a wild guess after looking at your schematic and data sheets for 2 minutes.

    Follow MrChips suggestions. Also, put a 10 ohm resistor in series with the bases of the output transistors. Use this resistor to monitor base current. Or use a some other kind of current sensor.
     
  10. JayJones

    Thread Starter New Member

    Jun 17, 2015
    15
    0
    It has blown once with a resistive load, and once with a speaker attached. It doesn't blow right away, but after a short while of passing audio.

    I'll see if I can add in a base resistor somehow. The current should be around 25mA or so judging by the LTspice sim.

    I'm wondering if the emitter resistors I initially used were too small at 0.1 ohm? It has not blown since I changed them to 0.47 ohm and the heat on the heatsink seems to be spreading out more evenly between the output transistors.

    Thanks again for the replies!
     
  11. dannyf

    Well-Known Member

    Sep 13, 2015
    1,771
    359
    The two capacitors on the output are killing your amp. It must be oscillating like crazy.
     
  12. JayJones

    Thread Starter New Member

    Jun 17, 2015
    15
    0
    It doesn't oscillate now, but I can try removing them.
    It oscillated like crazy without them when breadboarded, but now that they're on a PCB, they may not be required at all.
     
  13. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,518
    1,247
    What does "blow" mean? Are the output transistors overheating? Is it always the same one(s)? If it makes audio for a while then dies, that almost certainly is a thermal problem, and that means current.

    Separate from that (or not?) I think R7 is usually two resistors tied to the output, like the output transistor ballast resistors.

    ak
     
  14. JayJones

    Thread Starter New Member

    Jun 17, 2015
    15
    0
    When they have blown, the collector and emitter have been shorted together when tested with an ohm meter, and the 1A or 1.5A fuses have blown at the same time also. It has always been at least one NPN and one PNP that have blown. The first time they blew I believe three out of the four output transistors had shorted internally. I was under the impression that they should be able to handle more current than an amp or two, so maybe it is more heat related? Is it possible to have a case temperature of say 50 degrees C but the internal junction temperature is actually exceeding 150 degrees?
     
  15. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,518
    1,247
    Yes. Check the datasheet.

    What steps did you go through to adjust the static current through the output stage?

    Heatsinks - 3.7 C/W x 30 W = 115C case temperature. Check the datasheet for the delta-T from junction to case. Yes, the heatsinks are too small.

    50 Vpp is 25 Vp is 17.7 Vrms is 1.7 Arms into 10 ohms.

    Inductive kick from the speaker is not a problem.

    ak
     
  16. Plamen

    Member

    Mar 29, 2015
    34
    4
    Petkan:
    First start with no load and no input signal applied and if possible lower power supply voltages.
    The output should split the power rails - in case of their symmetry - sit close to 0. The overall consumption should be low (below 0.1A).
    Idling (biasing) current could set by the pot to minimum at this stage (lower resistance B-E). You can evaluate the current through each output transistor indirectly by the voltage drop across its emitter resistor.
    Once steady state idling is confirmed, apply input sine initially at very low level and observe by scope the output.
    Keep monitoring the consumption and gradually increase the input level while observing the output waveform.
    If the output follows the applied input undistorted keep increasing the level; (so far unloaded) until you see slipping.
    If the output is unstable do not proceed further and focus on assembly details.
    Bear in mind the layout of the real circuit is very important. The power wires from the bipolar source (transformer, rectifier, filter) should be connected close to the output transistors and should have local decoupling caps to GND (ceramic and electrolytic).
    Current carrying wires produce voltage drops. Heavy current carrying ones more so.Voltage drops could be reduced by shortening the wires and increasing their cross section and "eliminated to great extent by "star point" connection.
    For instance the wires from the rectifier to filtering caps carry heavy pulse current. It creates voltage drop in the wire connecting the transformer center tap to the filter capacitors. Let us call such a wire "hot" wire. If your circuit gets grounded at numerous points across hot wire - GND voltage difference is generated between them. All GND points in your circuit should be star point connected to one spot and then led to filter caps GND. In this way the GND difference will be eliminated from your circuit.
    Input should be applied through shielded wires connected to GND point next to input stages. The shield is to be grounded at the receiving end only and floating at the input terminals. This is particularly important for low level signal sources like magnetic heads or turn table pick ups,
     
  17. JayJones

    Thread Starter New Member

    Jun 17, 2015
    15
    0
    Basically, I just adjusted the 1k potentiometer U1 until the crossover distortion looked minimized visually on the scope.
    The idle current of the whole circuit without an input signal applied is between 25-30mA right now.
     
  18. AnalogKid

    Distinguished Member

    Aug 1, 2013
    4,518
    1,247
    The TIP41 is a 6 A part, in a 2 A application. Why do you have two in parallel?

    ak
     
  19. JayJones

    Thread Starter New Member

    Jun 17, 2015
    15
    0
    I was trying to spread the load in order to keep the heat down and increase reliability. It's obviously not working too good for me so far :)
     
  20. Plamen

    Member

    Mar 29, 2015
    34
    4
    Petkan:
    After setting the idling current do you still overheat?.
    Linear amplifiers have low efficiency like 30% and less. Heat dissipation rises with output power.
    With fixed power rails the difference in voltage between the rails and across the load is absorbed by the output power transistors.
    At low levels efficiency declines.
    Some people make the power rails automatically spread further apart when input level increases.
    This allows cooler operation at lower levels.
     
Loading...