• State Resolved
  • Locked Locked
  • Replies 20 replies
  • Answers 3 answers
  • Subscribers 31 subscribers
  • Views 7990 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

LM386 maximum output current

John Woodgate
Prodigy 140 points
Other Parts Discussed in Thread: LM386

The LM386 data sheet doesn't specify a maximum output current. For a special application, I need guidance on how much current it will deliver with a speech signal having a crest factor of 4, so device dissipation is not a big issue. I propose Vsupply = 12 V for the LM386N-1 and 18 V for the LM386N-4.

Please say whether a figure is peak or r.m.s.

  • 0
    TI__Genius 9175 points

    Hello,

    The pots given in the datasheet recommend the maximum RMS output power (thus RMS current up) to 10% THD.

    At 12V the maximum output current for 10% THD is 300mA. I would not suggest asking for more current from the part due to very hard clipping beyond 300mA. This also will be pushing the part quite hard into maximum dissipation; however, if the audio source is very dynamic, then it shouldn’t be a big issue.

    At 16V, the maximum current for 10% THD is around 320mA. Again this will be pushing the part hard.

    If you look at the plot “Device Dissipation Vs. output power 8ohm load” increasing the source voltage from 12 to 16V does not increase output current. The part is consuming the extra energy, increasing its dissipation sharply. Therefore, there is no advantage to running at 18V and in fact, you are more likely to damage the part for the same output current you could achieve at 12V.

    This also tells me that the 300mA is all it can give for the same distortion figure, no matter what the supply voltage or load.

    Regards,

    Matt

  • 0 in reply to Matthew Beardsworth
    Prodigy 140 points

    Thank you, but your response is not too helpful. First, you mention 'rms power', which is a misnomer; the product of rms volatge and current is average power. I'm surprised you didn't know that.

    Then you cite the data sheet. Of course I looked at it and did the calculations, but the results are not realistic compared with the results from other data sheets that include a 'maximum output current' value much larger than the 'ouput power' values indicate.

    The results you mention for 12 V and 16 V supply are due to **voltage clipping**, not current clipping.  With a 4 ohm or 2 ohm load, the current can be much higher than 300 mA without distortion, but I want to know what the safe limit is.

  • 0 in reply to John Woodgate
    TI__Genius 9175 points

    Hello,

    My mistake, you are correct in that RMS voltage x RMS current = Avg. Power. That’s who whole point of using RMS in the first place!

    As for voltage clipping vs current clipping, they are interdependent. However, when comparing between a 4ohm load and 8ohm load in the datasheet, both reach the same amount of THD+N at around 300mA.

    What type of load are you trying to drive with this part? What is its impedance?

    Regards,

    Matt

  • 0 in reply to Matthew Beardsworth
    Prodigy 140 points

    Voltage and current clipping are not interdependent if one uses the correct definitions. Without being too formal, voltage clipping occurs when the output electrode(s) of one or both output stage devices hit the rail, or nearly. Of course, with a resistive load, the current is also 'flat-topped', but that is an effect, not a cause. Current clipping occurs when over-current protection limits the output current. If there is no such protection, the output current can increase indefinitely until the die or the bond wire melts.  I do not want that to happen.

    The loads are audio-frequency induction loops, for communicating speech to hearing aids. For example, a 5 m square of wire laid out on the floor. So the impedance is inductive and resistive. For best use of the available voltage, the resistance shoudl be as low as possible. The amplifier is configured to sense its output current, applied as negative feedback, thus keeping the output current independent of the load impedance, giving a flat frequency response (typically 100 Hz to 5 kHz).

    The LM386 will not drive a 5 m square loop, of course, but there are applications for smaller loops. The low output current and relatively high voltage allow the use of multi-turn loops in spite of the much larger inductance than that of one turn.

  • 0 in reply to John Woodgate
    TI__Genius 9175 points

    Have you considered the LME49610? This is a great audio buffer that can source 750mA max. Place it inside of an op amp and its performance is increased and turns this into a great little power amplifier.

    I think this would be a better choice since you don’t have to run it to its maximum capabilities like the LM386. The distortion would also be orders of magnitudes lower.

  • 0 in reply to Matthew Beardsworth
    Prodigy 140 points

    I will certainly look at the LM49610, but for this project there is a preference for 'commodity' parts. I don't *have* to run the LM386 at maximum, I need to know the maximum so as to not exceed it.

    Thank you for your continued interest.

  • 0 in reply to Matthew Beardsworth
    Intellectual 380 points

    Sorry to bring up this old thread but does the LM386 IC experience an increase in THD % before voltage clipping occurs? Reason being I just measured close to 1w into 8 ohm @ 12v Vs before voltage clipping started (2.8v RMS), I thought these chips were good for upto 700mW output max?

  • 0 in reply to John Jones
    Prodigy 140 points

    700 mW with a 9 V supply. With a higher supply voltage, of course you get more output.

  • 0 in reply to John Woodgate
    Intellectual 380 points
    So are thermals and total harmonic distortion at those levels why they only rate it for 700mW @ 9v into 8 ohm?
  • 0 in reply to John Jones
    Prodigy 140 points
    The data sheet just gives examples. Obviously, it can't give examples for every permitted supply voltage and every load resistance (which is not just 4, 8, 16, 32 ohms because all intermediate values can be used). The rest of the data sheet gives the information you need to design, say, an amplifier with an 8 V supply and a 5 ohm load. Take care not to exceed the package dissipation.
  • 0 in reply to John Woodgate
    Intellectual 380 points
    Well the datasheet states 1.25w as the maximum but doesn't that seem a little bit too high for an 8 pin dip package? The second source 386 from JRC states 700mW package dissipation as maximum.

    Can output total harmonic distortion increase before any voltage clipping appears on an oscilloscope when measuring across the output load?
  • 0 in reply to John Jones
    Prodigy 140 points
    You can see what happens in the relevant ' Distortion vs output power' graph in the data sheet. The very steep rise in distortion is caused by clipping; it creates odd harmonic distortion because the waveform is clipped symmetrically.
  • 0 in reply to John Woodgate
    Intellectual 380 points
    So as long as I'm just below clipping then I can assume the THD levels are going to be low too? I'm trying to determine where to stop taking power measurements, at clipping or maybe a bit lower if the THD is going to be high enough to hear.
  • 0 in reply to John Jones
    Prodigy 140 points
    Just-visible clipping is often said to result in 2% THD. Most people can't hear that, despite popular belief. You can stop taking power measurements at clipping, with confidence.
  • 0 in reply to John Woodgate
    Intellectual 380 points
    Thanks! I had no idea, so there is some increase in THD before actual clipping occurs? Do you know of any good threads/reading materiel on the subject (preferably with illustrations if the latter).
  • 0 in reply to John Jones
    Prodigy 140 points
    Look at the graph in the data sheet. The increase in THD is very sharp, from a value well below 1%, which is negligible. For further reading, any good audio textbook should help.
  • 0 in reply to John Woodgate
    Intellectual 380 points
    But isn't that only applicable for 6v supply voltage?
  • 0 in reply to John Jones
    Prodigy 140 points
    It's typical of what happens at any supply voltage and even to any well-designed audio amplifier, excluding Class D.
  • 0 in reply to John Woodgate
    Intellectual 380 points
    I meant that the chart in the datasheet shows the distortion levels when powered with 6v into 8 ohm, won't it be different with a 12v supply?
  • 0 in reply to John Jones
    Prodigy 140 points
    The curve will be shifted to the right, showing clipping at a higher output power. But the general shape will be the same; negligible THD up to a point where it steeply increases as clipping occurs.