LM5013-Q1: Thermal cut-out

Joe Mann

Part Number: LM5013-Q1

Tool/software:

Hi,

I'm having issues with the LM5013-Q1 going into thermal shutdown well below the design current of 3.2A. It shuts down after 54 seconds from 30C ambient at a continuous 2A.

I read elsewhere that the Vin and Ground pins require a reasonable amount of copper to transfer heat. Granted my layout hasn't accounted for this, but I'm dubious as to weather improving those copper areas will allow stable operation at 3.2A. I tried soldering some heavy copper wire to the Vin area between C66 & C67 to try and improve thermal dissipation but the time to thermal shutdown from ambient was only improved by 6 seconds (1 minute). This may well just be coincidental anyway.

Here's my schematic:

Here is the layout. PCB is a two layer, top and bottom copper pours are 0V/ground.

7 months ago

0 EricLee 7 months ago

TI__Guru 66751 points

Hello Joe

Please make it sure DAP is properly soldered. Also, you have to use multiple thermal vias underneath the IC, which are connected to the DAP/EP of the device and a large copper polygon on the bottom layer.

In this example 6 vias are used underneath the IC . You can add more if no issue.

-EL

0 Joe Mann 7 months ago in reply to EricLee

Prodigy 40 points

I will try a revised layout as suggested.

My calcs show that the converter should be dissipating about 0.73W which should translate to a temperature rise, worst case, of 26C. This roughly squares with what I'm seeing with my thermal camera so I am slightly surprised that it seems to be going into thermal shutdown. The case temperature reaches about 62C when it does, is it possible the junction temperature threshold has been reached at this stage?

Can I just confirm that there aren't any aspects of the circuit design which would cause the converter to operate at a much lower efficiency, therefore increasing thermal power dissipation?

0 EricLee 7 months ago in reply to Joe Mann

TI__Guru 66751 points

Hello Joe

Joe Mann said:
Can I just confirm that there aren't any aspects of the circuit design which would cause the converter to operate at a much lower efficiency, therefore increasing thermal power dissipation?

I haven't seen. Unless the device is damaged, the power dissipations should be similar.

Joe Mann said:
The case temperature reaches about 62C when it does, is it possible the junction temperature threshold has been reached at this stage?

Junction to top char parameter is only 0.3°C/W. Estimated junction temperature is only 62.21C if power dissipation is 0.7W. (=62+0.3 * 0.7)

0 Joe Mann 7 months ago in reply to EricLee

Prodigy 40 points

Hi,

So is it possibly that something other than thermal shutdown is occurring here? Seems highly unlikely given the case temperature.

0 EricLee 7 months ago in reply to Joe Mann

TI__Guru 66751 points

Hello Joe

Would you please let me know how did you confirm your converter go into thermal shutdown ?

-EL

0 Joe Mann 7 months ago in reply to EricLee

Prodigy 40 points

Hi,

It was an assumption on my part as it correlated with increased load and temperature. Do you know what else could be making it shutdown?

0 EricLee 7 months ago in reply to Joe Mann

TI__Guru 66751 points

Hello Joe.

1) Can you double check D43 direction ?

2) If the issue was not thermal shutdown, it should be current limit, please check if the inductor is saturated, and also measure the inductor current using current probe.

-EL

0 EricLee 6 months ago in reply to EricLee

TI__Guru 66751 points

I am closing this thread since I have not heard from you for sometime.
- EL

Interface

Interface forum

LM5013-Q1: Thermal cut-out