[sdiy] circuit protection question

[blacet at blacet.com]

The diodes are much faster than the polyfuse but the diodes will
eventually heat up and possibly fail. By that time though, the polys have
activated, limiting the current and voltage to a very low level.

I'm suprised that you got an op amp to smoke! That really should not happen.

The polys come in different current ratings. You need to use the correct
size for the current draw of the module, otherwise you will get too much
voltage drop. I stock 100mA, 170mA and 300mA.
The 300mA was necessary for the Hex Zone because of the 100mA power
consumption from the LCD backlight on the + rail.

> On Tue, Dec 14, 2010 at 9:02 AM,  <blacet at blacet.com> wrote:
>>
>> Why don't any other companies use this? Too cheap? I dunno.
>>
>> People WILL plug things in backwards, it's guaranteed.
>>
>> John "the Guru" Blacet
>>
>
>
> John and all,
>
> I use your technique on all my modules.
> I've come up with a few questions I'm wondering about:
>
> It seems to me that the reverse diodes are doing the protective work
> for the module. What are the polyfuses doing? I'm figuring that they
> protect the PSU? I wondered if they protected the circuit in different
> ways. So I tried an experiment and put an op amp in backwards. It
> smoked.
> I'm also a bit foggy on the value of the polyfuse and figure it
> depends on the current draw of the particular module. Looking at a
> Bourns data sheet, generally, the lower the trip point the higher the
> resistance of the polyfuse in it's normal operating state. Doesn't
> adding resistance in series with the power lines make them less stiff?
>
> Semi related:
> How about if someone plugs the power cable in and they are off to the
> side one pin?
> Say...+15V header is now connected to PSU ground and PSU ground and
> -15V are connected to the header ground and the -15V header pin is
> unconnected (the power connector is shifted one over where it should
> be).
>
> Your article convinced me to use your technique years ago. I'm just a
> bit embarrassed that I don't know everything that is going on with
> this.
>
> p.
>