Most suitable driver chip

They do. The absolute maximums are precisely those. Since we're talking Microchip here I'll quote the first of their datasheets that comes to hand, namely a PIC12F675, though I'm sure the wording will be identical in your chip's datasheet:

Stresses above those listed under 'Absolute Maximum Ratings' may cause permanent damage to the device. This is a stress rating only and functional operation of the device at those or any other conditions above those indicated in the operations listings of this specification is not implied. Exposure to maximum rating conditions for extended perioids may affect device reliability.

The specifications you want to design to are listed as 'DC characteristics' later on in the data sheet. In the case of the device I am looking at, for instance, Vcc is quoted as going up to

Think of these as operating and non-operating limits. Within the more conservative limits specified in the data sheet the device is guaranteed to perform as described. Outside of those the device will not be damaged in the short term but it is not guaranteed to operate correctly under those conditions.

In your particular case of sourcing/sinking too much current through the pin you'd expect to see voltages lowering or rising from their nominal values, possibly into an undefined region or worse, clearly into the 'wrong' logic state. Prolonged exposure to these conditions stresses the device and may cause it to fail prematurely.

Wait until you learn how fuses work.

That's why you need to read the whole data sheet. Also it helps if you know what the commonly used terms like absolute maximum means. Now you are beginning to. Sometimes these terms are explained, sometimes you will be expected to know them. Microchip did explain the term you just had to read a little further.

The rest of the information will likely be in the parts of the data sheet you haven't read.

Now for the rest of the bad news, some times data sheets have mistakes, are incomplete or are written with confusing turns of phrase.

Robert

To those of us who also deal with the analogue and thermal aspects of design the data sheets that provide some graphs of the output device capabilities will speak volumes about the fuzzy areas between fully acceptable and totally unacceptable.

To get an appreciation of the zones, plot the load power curve using the current through the load for a range of output voltages that will be applied to it as it switches logic state from 0 to 1. You then should calculate the thermal dissipation of the device based on the peak power and then see how long the device is likely to last. Quite often, if you want to ensure the device lasts a long time, you will find yourself using much lower values than suggested by the data sheet maxima.

Maybe, tho that is a newish device.

It could also mean they found the 95mA was all the part was reliable to. Not the first time a device has been 'derated' following field feedback !

-jg

Here's what I'm thinking of going with for the display:

Ah crap I spoke too soon. I want to be able to turn both LED's off by setting the uC pin to high impedence.

Maybe I should add a pull-up/pull-down resistor to the microcontroller and then replace one of the FET's with a BJT?

Have you considered a 'high side' driver array such as the TD62783 instead of discrete components? It makes a good complement to the

Welcome to the world of engineering in general. Best regards, Spehro Pefhany

WOW, I had to check the date on my post.

Catching up on usenet ?? ;-)

donald