Power supply

[Ian1954]

I have designed a power supply using a TOP268 controller from Power Integrations. The supply generally works fine but has blown up 3 times during switch on. Has anyone had a similar experience and if so did you find the cause of the problem?

 

[itsmoked]

Got a link to what you're talking about?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[MacGyverS2000]

WHAT part has blown...?

Dan - Owner

http://www.Hi-TecDesigns.com

 

[Ian1954]

The failed part is the TOP268. I've not used this part before. Never had a problem like this with other supplies I've designed. The fault occurs rarely. I've switched it on and off many times with no problem. The circuit can be found on the Power Integrations web-site,

https://ac-dc.power.com/design-supp...on-notes/an-47-topswitch-jx-application-note/

 

[MacGyverS2000]

Have you put a scope on any part of it? You appear to make the assumption the part itself is faulty... I would wager your surrounding design is faulty, but the eventual victim is the chip. Switching power anything will involve voltage/current spikes... are you sure none of those being created exceed the part max tolerances? Is there any physical evidence that could lead you to believe it's something ancillary to the chip, such as poor PCB design with low- and high-voltage traces too close to each other causing arcing?

Dan - Owner

http://www.Hi-TecDesigns.com

 

[Ian1954]

It could well be the external circuit causing the distress to the TOP268. I have scoped it. The drain to source waveform looks good. Very little overshoot. Maximum voltage is 420volts to the transient peak, 400volts is the DC bus plus the reflected secondary voltage. Only 20volts due to leakage inductance. The device is rated at 705volts minimum. Can't detect any turn on transients going above 325V dc, the rectified incoming mains.

 

[itsmoked]

The other killer is that device finding itself in the linear region - huge power dissipation - for a moment anyway.

Any chance of some low frequency stuff while starting up or shutting down?

Keith Cress
kcress -

http://www.flaminsystems.com

 

[felixc]

Out of curiosity, have you added some inductor or filter between the filter cap and the chip to smoothen the noise? I've seen circuits blow because of it. No path for the spikes to be absorbed.

 

[Ian1954]

There isn't any added inductance. The supply is direct from the 3 phase supply. A simple 3 phase bridge to provide the DC and 20uF of smoothing capacitance. It could be resonance with the line inductance but the problem occurs so infrequently that I don't think that's the case. I have switched the power On and Off many, many times over the past 2 weeks and have not had a single failure. I have been deliberately trying to provoke one in order to show where the problem lies.

 

[IRstuff]

HAve you done any sort of failure analysis on the failed parts?

TTFN
faq731-376

Need help writing a question or understanding a reply? forum1529


Of course I can. I can do anything. I can do absolutely anything. I'm an expert!
There is a homework forum hosted by engineering.com:

http://www.engineering.com/AskForum/aff/32.aspx

 

[Ian1954]

A piece of encapsulate is blown out leaving a hole in the device. The hole appears where I would think the internal power MOSFET is positioned. The failures, when they occurred, were instant on applying power. It certainly has the characteristic of a transient overvoltage causing avalanching of the MOSFET but I have monitored the drain source voltage with a fast oscilloscope and have seen nothing in excess of 420volts. This is the expected voltage when the MOSFET is switching.

 

[daveross100]

http://www.mouser.com/ds/2/328/topjx_family_datasheet-11851.pdf

Add CL and TVS to prevent any over load.

Any question contact :daveross100@gmail.com