TOPSwitch-GX Family internal self protection current limit

Hi, Thanks for sharing the information.

The drain current should be well limited by the current limit, the reason of getting TOP243 damaged with lower capacitor should be further investigated. It may not related to the current limit.
Did you design your power supply using our PI-expert software? Would you please attach them, that I can have a check for you?

Hi PI-Terry,
how are you?
We have already discussed topic "Flyback PQ20/16 and TOPSwitch-GX family" and concluded design isn't good with NS=2.
But according to AN-17 there are three iteration variables: NS, KRP and number of primary winding layers.
The design that we have considered had problems for NS = 2 and KRP = 0.6 (continuous operation).
If user manually sets KRP = 1 (discontinuous mode) and NS = 2 with many other modifications as in the real design obtained solution has a few warnings but design passes! Calculated transformer looks similar then as real.
According to the program design could work.
How to submit uds file - forum supports only text or photos?

The original design was made 10 years ago with available tools so mistakes are possible.
For example maximum line capacitor for the same volume and voltage was 15uF and today is 22uF.
The declared consumption can occur at one or another secondary 24V (first or second external load) while 5V powered CPU.

Getting back to the topic. With experiment we have found it is possible to destroy the IC if the capacity of the high voltage DC bus capacitor becomes too small (1uF instead 15uF). That sometimes really happened because mounted capacitors in the past weren't long life and may lost capacity if the ambient temperature was high. With better capacitors problem is no longer visible.

Please tell me if you have any idea what was going on.

Best regards, Zeljko.

Hi,

Thanks for sharing all the experience. It is good to know that you solve the "Flyback PQ20/16 and TOPSwitch-GX family" issue. Yes, you could be able to make a legal design if you tune the design by choosing different operation mode, for example the KP or KRP. However, if there are some warnings that still showing in your design, I would recommend you to carefully check what are the warnings, and perform necessary bench tests under different conditions. Just make sure the design is solid and works well under some extreme conditions (for example, output short circuits, higher input line voltage startup...).
Back to the topic, it is not so clear to me that the 1 uF capacitor caused the damage of the IC. Is there any inductor that you used in your circuits from the AC input to the transformers? Is there any possibility to measure the drain current and voltage during the damage? That will be helpful to check the reason of the damage. There is also a trick to avoid this issue, it is to use two capacitor in parallel with equivalent capacitor value. If one single capacitor failed, the other will still work to protect the IC.
Hope this is helpful for you.

Hi.
The common mode choke 22mH is placed between line and bridge. I've noticed that simulation force 6mH.
22mH is chossed because it is a component from our warehouse.
Does it create problems if the value of the input capacitor fall?
Experiments were made earlier before choosing of long life capacitors. On the printed circuit board barely has room for one capacitor so that the idea of the two is good but does not help. Thanks anyway.
Best regards, Zeljko.

Hi:

If the input capacitor is low, this may create some performance issues, for example the output voltage ripple, efficiency ...ect. But I should not cause the damage of the IC with lower input capacitor.

With appropriate capacitors potential problem is related to the lifetime of the product.
Thank you for your efforts.
Best Regards, Zeljko.