Output of SMPS start rising very slowly
Hi Experts,
We are using LNK623DG for our 5V, 900mA (4.5W) SMPS. We are using the following designs for last couple of years and shipped more than 70K units so far without problem so we believe no problem with design (schematic attached). However in recent production lot we are facing a problem that we believe is due to some faulty component but not able to figure out which one is faulty.
Problem description:
When power up SMPS we immediately used to get voltage around 4.9-5V. But in recent production lot we get voltage anywhere around 3V-3.8V on powerup, it increases very slowly and eventually reaches to 4.6-4.8V after several minutes.
1) We suspected dry joints, did touchup of all components.
a) Sometime behaviour remains same (low output voltage and increases slowly)
b) sometime we get correct 4.8-4.9V after touchup but if we leave SMPS off for 24 hours and check next day we see same behaviour (low output voltage and increases slowly)
2) We suspected Input capacitor (4.7uF/450V - EC1&EC2), replaced them with the capacitor working perfectly fine from our previous production lot. Result: No real improvement
3) We replaced 1uF (C2) capacitor from different brand. It seems to helped reaching 4.8-5.0V immediately on power up. We thought we found the faulty component but if we leave it off for 24hours and switch it on again, we see same behavior (low output voltage and increases slowly). So again no help
4) We replaced LNK IC from different lot. that also seems to help reaching 4.8-5.0V immediately on power up but next day again same behaviour (low output voltage and increases slowly). So again no help
5) Tried to replace output capacitor 470uF (no real change observed)
6) Hard to doubt on PCB, checked all traces with continuity test and did not find anything wrong in pcb either.
Now running of ideas how should we proceed to identify problematic component. because what seems to work immediacy after replacement does not work next day.
Our original designer who made this design is also switched the job and is not with us currently. Looking for your expert advice to isolate the faulty component.
Comments
Dear Sir,
Please find the requested details. (BOM, Gerber and Transformer specs). Hope this helps.
Do not have PI Expert file. Design take care of universal supply 90V-265V. We are testing at 230V
| Attachment | Size |
|---|---|
| Gerber Data (50.57 KB) | 50.57 KB |
| Bill Of Material (96.69 KB) | 96.69 KB |
| Transformer Specs (546.63 KB) | 546.63 KB |
Hi Vipin,
Good day. Since you don't have a PI Expert file, I have made an initial design using our PI Expert Online tool. I have entered your input and output specifications and the total value of your input capacitance, then replicated your transformer number of turns. According to it, your primary inductance (1.604 mH) is too low compared to the required primary inductance that is required to deliver an output power of 4.5 W. This might result in the current limit being triggered after leading edge blanking and might be the one causing the slow rise in the output voltage. It also flags a warning on the peak flux density because it exceeds the recommended value of 3100 gauss, which might cause core saturation during operation. I suggest optimizing the transformer first to rule out that it is the one causing the issue. Kindly refer to pages 5 and 6 of the attached file for your reference regarding the suggested transformer specifications. Please note that the pin allocations of the transformer correspond to the pin connection in your schematic.
If you wish to modify it further, especially the wire gauges, here's the link to its PI Expert Online spreadsheet design: https://piexpertonline.power.com/piexpert/design/select?share=2be19b05289d7d5187822d5b8beab95ebd33efbf42e43d3da84618e5148fb71c.
Feel free to contact us if you require further assistance with this issue or have any more questions or concerns.
Thank you and have a great day.
Regards,
Tommy
| Attachment | Size |
|---|---|
| LinkSwitch-CV_4.5W_5V_900mA.pdf (1.3 MB) | 1.3 MB |
Hello Vipin,
Good morning and thank you for continuously choosing Power Integrations. I'm sorry to hear that you're experiencing this kind of issue.
If you could share the PCB layout (Gerber files preferred), BOM, transformer file (with primary inductance and number of turns of each winding), and the PI Expert file (if available on your side), it would help our investigation on what might be causing this slow rise in output voltage. Also, what is the input voltage range of the design?
I will wait for your response. Thank you.
Regards,
Tommy