解决方案搜寻 技术支持

Problems in Production with TNY268

Posted by: Fred Vines on

Dear Sirs,


I have two questions regarding our power supply designed with TNY268 (Please find attached the circuit diagram).


Any information from you could be very helpful to us:




1) The TNY268 worked very well during the prototyping phase. We assembled about 17 prototypes, and we didn't have any problem with the boards. In the prototyping phase, we even did not have all due soldering and ESD protections.


But in the production phase, more than 10% of the power supplies are failing in the first test. The TNY268 simply doesn't operate. After we change the component for a good one, the boards work fine.


We could not find any error in design (overvoltage, overload, bad layout...) or error in production process (bad temperature profile, improper ESD protection...) that could justify such many failures. We know that the company which is assembling the boards have very good control for ESD and temperature profile.


- Should I question the origin of the TNY268 ICs acquired by the company which is buying the ICs and assembling this board to us?

- Do you have any information, former experience or recommendations that could help us in this case?




2) We are about to review the design of this power supply, and we found out that TNY268 is not recommended for new designs. Do you recommend we change our IC for a more recent product, or will this IC still be available for a long time?




Thank you very much!


Regards,

Fred

Sao Paulo, Brazil

Files

Attachment 大小
diagram_power_supply_tny_268.jpeg 397.02 KB

评论

Submitted by John Carpenter on 11/20/2008
Try installing a 1 k Ohm resistor in parallel with the optocoupler diode (pins 1 & 2 of CI3). The TL431 bias current may disable the TNY268. There are no plans to discontinue the TNY268, however the TNY III devices may be less expensive. Z. Cochrane
Submitted by PI-Chekov on 11/21/2008

Just to expand the explanation the TL431 requires a minimum current for correct operation (0.4 to 1 mA). Without a parallel resistor this current will flow through the opto and depending on the CTR of the opto coupler will exceed the disable current of the EN/UV pin, stopping the device for switching. Adding a 1 k in parallel to the opto supplies this current the TL431 without going through the opto.

Cheers

PI-Chekov