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.

LM3886: Failure modes and low impedance performance

Part Number: LM3886

Tool/software:

Hello,

I am using the LM3886 in a unique scenario where it configured as a single supply application circuit with the V+ supply tied to GND and V- tied to -48VDC relative to GND. The output impedance that the power amp sees in our application is approximately 2.5 ohms. We are using a 1.4mF series capacitance instead of 4.7mF and this feeds into the primary winding of a 50W custom impedance matching transformer. This transformer then feeds into a load with resistive and capacitive components.

In some instances, we have seen two types of failures in the field.
1) The pin 3 output shorts to -48VDC as a failure mode. This appears more common.
2) The pin 3 output shorts to GND as a failure mode.

Can you provide insight as to what the root causes could be for both types of failure modes, and then how to mitigate? The output power we were seeing was often between 25W and 40W.

As a related follow up question, I mentioned that we measured a primary impedance of 2.5 ohms. In these cases the input tone frequencies were below 1kHz.
Can you provide details as to what the maximum power dissipation of the power amp can be and how to calculate it? From the LM3886 datasheet, I notice maximum power dissipation information for 4ohms and 8ohms in Figures 35 and 36 respectively as a function of supply voltage. What about below 4 ohms? I notice information in Figure 33 regarding maximum power output as a function of load resistance, but can you provide similar details for maximum power dissipation across the amplifier itself? What would be a safe range for the circumstances I am describing?

Thank you,

  • Hi,

    2.5Ohm load seems quite low for conventional speaker amplifiers, that's why most of the data you see in this kind of amplifier spec is for either 4 or 8 Ohms.

    Perhaps the current/temperature peak transients are faster/larger than what the protection can handle and thus the output transistors get damaged.

    Best regards,
    -Ivan Salazar
    Applications Engineer

  • Hello,

    I should add that we connect these power amplifiers to a number of static loads with unknown impedance. The end result is that the power amp itself can see loads that vary from around 2 ohms to around 6 ohms, depending on where and how it is installed. We are not using classic resistive audio loads in our unique application. Our load is mostly resistive but also with a strong capacitive component.

    Do you have any comments on what sorts of conditions could cause it to be more likely to fail with the output short to -Vcc versus +Vcc?

  • Hi,

    Perhaps it is related to how the back EMF is produced from the actuator. I have no clear idea why one side of the output can get damaged more than the other.

    Best regards,
    -Ivan Salazar
    Applications Engineer