Input EMI Filter issue with TNY287 Circuit
I designed a circuit in PI Expert with following specs :
Input = Universal (85 VAC to 265 VAC),
Output-1 (Main) : 3.3V @ 1.2A,
Output-2 : 8.0V @ 0.3A,
Isolated, with TL431 and Opto-coupler feedback.
I selected a Common-mode EMI filter, and the PI Expert software suggests a 6 milli-Henry CM Choke. That is a huge value. My design is supposed to be a very compact PSU, almost like a module, to be used in a custom consumer electronics device. The CM-Choke alone increases the height / width by a magnitude.
I have done another design previously with a TNY284 chip, and a similar high-value Inductor was suggested (PI Filter) - i think it was a 1mH one. And yet it failed CE conducted emissions testing, then we later used a 15uH (micro-Henry) CM-Choke with appropriate caps to just meet the CE requirements.
This leads me to think that there can be other possibilities apart from the 6mH CM-Choke suggested by the PI Expert. Can you please have a look and suggest a choke that is in the uH range (thus smaller size) AND associated components for EMI filtering ?
I am attaching my Circuit as made in PI Expert here (both, UDS file as well as a PDF file), for reference.
Thanks.
Zeeshan Ashraf
Comments
Thank you Jedidiah.
I understand that PI Expert gave me a suggested value for the Input EMI circuitry. The Transformer we get from a listed vendor of ours (a Transformer factory in China), based on the Transformer specs as provided by PI Expert software. Though i have not cut through the transformer physically to check if it is made exactly as per the details, however, it did not give us any problems so far in our previous design. So i am confident the transformer is not much of a problem here, maybe a little bit but not too much to worry.
The layout, well, that is very tight and we do ensure to make the loop areas very small with a little more than ample trace widths, to ensure very low resistance current paths.
As a reference, i am attaching the other PSU schematic i mentioned in my post, for one of our other products. This was also designed in PI Expert initially, PI Expert gave us a Single-ended Inductor (1mH) with associated componentry as the Input EMI filter, but then, after it failed CE, we were suggested to use the Common-Mode choke (15uH) with the caps (6.8uF -AND- 10uF) to correct for it. This made the conducted emissions within the limits and we passed the tests. Now i am not saying there is an error in the PI Expert suggestions, i do understand there can be more than one way to get it right.
So, for my current design, it suggests 6mH and the part number it gives, i checked it, its too huge to fit our PCB as well as confined space. So i am hoping maybe you guys can suggest an Input EMI filter than has physically smaller magnetics and capacitances, to just meet the CE Conducted emissions limits and yet meet our size constraints.
Thank you.
| Attachment | Size |
|---|---|
| EC_A2D.png (151.47 KB) | 151.47 KB |
Hi TopGunkPk,
Thank you for this. EMI is hard to predict as the solution to solve it varies in every design even with same output wattage but different board density. We can try to improve the suggested EMI filters but variations on the outcome of the EMI data can't be predicted properly.
Regards,
Jedidiah
Hi TopGunPk,
6mH Common Mode Choke might have worked properly for this design to pass EMI requirements. This is just suggested values by the software but there are several factors that affect EMI aside from CM Choke like layout, transformer construction, leakage inductance, and others. You may not use the suggested 6mH Common Mode Choke and optimized your design with low profile CM Choke or low inductance CM Choke and checked the EMI. Several EMI filters can be adjusted in case you are having problem when low profile/low inductance Choke was used like, increasing y-cap, adjusting snubber values, and make the crictical loop layout in PCB as close as possible.
Regards,
Jedidiah