LNK304 fails after haft hour

Posted by: behnam_sh on 08/16/2016

Hi every body

I used LNK304 in some of my boards convert 220vAC to 5vDC and it was working fine for about 2-3weeks. but after some weeks, now they shutdown after about half hour of working and then after 1-2 mins turns on again! and this loop repeats for ever.

The schematic is attached.

I guess the problem is about heating. maybe some components are lost their quality! (I've changed the inductor but is still here)

Can anyone help me on this please?

Thanks

Files

Altium Schematic. Transformer starting point wrong in schematic But corrected in PCB design (42.25 KB)

Comments

hi, Any one there?

Hi behnam_sh,

Thank you for choosing Power Integrations.

Could you please provide further information on the design that you are doing?

1. What is the output current of your power supply/maximum output power?

2. What is the ambient temperature during your tests?

3. Could you please attach a copy of your PIxls/PIexpert design file?

4. If possible could you also provide a copy of your PCB layout?

Thanks.

HI, thanks for your reply
The pichture of PCB attached,

- I use it in room temperature, 4-5 hour in day

- the output voltage set on 6v and current is 40mA maximum

- I used the circuit in datasheet, does not used PI expert

Attachment	Size
Capture2.PNG (82.72 KB)	82.72 KB

Hi behnam_sh,

It would seem that the device is in over temperature protection. Could you measure the case temperature of the device?

Suggestions:

1. Check if the device is close to another device which has high temperature.

2. Try increasing the inductance to around 1100uH. After simulating your design using PIexpert, I found that the device will operate at a duty cycle below recommended minimum at Vmax.

3. Could you send a picture of your actual power supply board?

4. Ensure also that your inductor is properly rated. Check if there is inductor core saturation.

Thank you so much PI-Traveler

I attached both sides of actual board

I'm agree that the problem is with over temerature, but maybe any other component such as 1mH Inductor make it! I tried with some other inductors (some large radial 1A & 3A !!!), the only that works is one in attached picture!

Do you think it would be because of 5v output that I adjusted by feedback resistors? it goiing to fix if I increase it to 12v?

(I changed inducted with 1100uH, problem is still here)

Attachment	Size
photo_2016-09-15_16-49-02.jpg (127.78 KB)	127.78 KB
photo_2016-09-15_16-44-01.jpg (100.74 KB)	100.74 KB

Did you measure the LNK304 temperature? I can see from the photo you sent that the LNK304 source pins do not have enough copper heatsink. Could you try to externall force air into the LNK304 using a fan? Check if the protection will still kick in.

Thanks traveler

No I did not measure the temperature yet, but touched and it goes too hot!

So you think the protection makes it shutdown not other components heating? I have tested it in a cooler environment and seems the shutdowns happens much lower times.

I designed new PCB with large amount of copper, I will test that point, but want to know if any other issue may be in my PCB now such as components saturation or heating, or any other structural problem

You could check the inductor current waveform during normal operation to see if there is any saturation. by measuring the temperature on the inductor you could also get clues if the inductor is saturating.

Larger Headsink PAD with no soldermask made in cooler and no reset happened any more. Also enough large Inductor used to prevent saturation coused by heat and current

Thanks for support