Favorite Buck Regulators

that

good

doing

what

Joerg has continual *KABLAM* issues. I think he has a black thumb ;-) ...Jim Thompson

We had a battery that provided around 20V and it was a small one. Towards the end of the discharge cycle the natural battery thing occurred, the impedance goes up. If you then switched on the line to the battery ... *POP* ... the chip blew a crater. Communication with TI did not provide a solution and so I designed it out.

If you are using them at lower voltages and/or with a low impedance power source they may be ok. But I will not use them.

variety

parts

that

good

doing

may

what

No, this one was more like .. tzt ... *POP* ... very fast.

For some reason almost every time this happens the support folks scratch their heads, things come to light that weren't quite made clear in the datasheet, claims that were in the datasheets appeared to be written by Marketing, and lots of other things depending on the case. The topper was one large group that assembled at the mfg, the big old chip design drawings came onto the table, and finally one guy exclaimed "Ooooh S..T!". I had inadvertently unearthed yet another gotcha but this one was serious. The group then went into a sort of damage control mode.

It's from that cheap, twisted hammer (with a broken face) that he brought with him from Europe! ;-)

Just be happy that you actually got some people together that can tell you. Some components that I (have to) work with have undocumented features or no complete specs at all. I'm currently writing some software to control a camera module with a lot of image processing features. I have a datasheet for the camera chip under NDA but it seemed the software that comes with the development kit writes a lot of undocumented registers as well. The manufacturer won't disclose what those registers do. The lense focus control is another story. There is no documentation at all :-) I managed to find out which chip it uses and how its connected. But AD doesn't provide the datasheet. I simply tried if it would work the same as a similar chip from AD and it does.

sounds like the battery was acting as a shunt clamp. Sounds like a job for a snubber/Zener..

Jamie

Well, I'd expect such needs to be mentioned in the datasheet. Even more so as the TI part was (still is ...) advertised for the micro-power market.

The least I'd expect is that after such episodes someone revisits the chip design and adds the essential information to the datasheet.

The LM2575 is a 5-terminal TO-220-sized package, and they call that the "simple Switcher'. Maybe they had an earlier version, but the general scheme makes me think it would be hard to do in 3 terminals. You need input, ground and output, and a sense connection, so it seems it needs at least 4 terminals.

I have been using the LM2575S-5.0 in a product for a while, and have had no problems with it so far.

Jon

You are correct, of course. Too much to-220 stuff floating around in my head.