ULPWU and current draw.

[s.jaglowski]

I'm using the ULPWU feature of the 16F688.

I'm experiencing far more current draw than I should be with things turned off and 100K pullups to VDD on inputs. External circuits should offer ~10uA.

However, I'm measuring 60uA over all. Also I note that if I code the machine such that it is in almost permenant sleep mode, waking to execute just a few instructions, the current drawn in sleep increases to 80uA. Additionally, with the small code set, the current drawn in sleep will rise when the voltage rises but not when a longer wake time is used (50ms). I think this may be something to do with the dynamics of the LCD meter I'm using, but I'm not sure.

The ULPWU has a 200nA current sink to draw from the RC load but shouldn't affect anything.

My question is, given there is digital CMOS circuitry in that area and a slow changing voltage on the input pin due to the RC and current generator, could this explain the additional current I am measuring?

Regards,

Steve Jaglowski

[PCM programmer]

I would look for modules that are still enabled when the PIC is put into
sleep mode. Look for modules that use amount of current:

1. Brownout Detect
2. Comparator
3. Comparator Vref
4. Weak pull-ups on Port A, with an external circuit holding the pin low.

Any of those could account for the 60 ua.

[s.jaglowski]

Hi thanks for the suggestion.

I'd already done that and theoretically all are off as are pull ups. I am going back however to turn them on to see if the current increases. Its the only way I can make sure I have actually turned them off.

I agree with the premise though; it must be something I think I've turned off but must actually be on.

.....................

OK. I tried out turning things on and sure enough the current increased and then decreased when I turned them off. So I know they are off.

What's going on?

Regards,

Steve Jaglowski

[s.jaglowski]

Just to close the thread with the solution to the problem.

It turns out, the source of the current was the receiver circuit of an IrDA device I was using.

Although the shutdown pin for the device was in the required state, it turns out the shutdown only operated the transmitter circuit. This feature was not mentioned in the data sheet. The receiver was automatically shut down when a bit was being transmitted (a high on the TX pin) to prevent local echo. I believed the output state on the pin didn't matter and coincidently had it low.

So I wasn't stupid (which I did think at one point); just ignorant of features not declared in data sheets.

Now who do I bill for the time this took me to resolve???

Regards,

Steve Jaglowski