Current consumption using 32KHz internal osc

Bean

I have an sx28 using the internal 32KHz oscillator.

The program is just generating a 1KHz clock on an output pin.

When I measure the current @ 5Volts it starts about 12ma then slowly drops to about 1.3ma ???
Slowly being about 10 seconds or so.

Is this normal ? Has anyone else tested this ?

Bean.
·

KenM

Good question Bean,

I don't know the answer other than.......

Last night I was doing essetially the same, running on the internal 32khz osc, with one LED output and using the analog comparitor on rb.1 & rb.2

I saw the exact same thing.....current would start out at about 13 mA, and slowly work its way down to 5 mA or so (in about 10 sec).· After a while, over a similar time period the current would drift back towards 13 mA.

Very interested in other replies/results.

Michael Chadwick

This sounds like a classic problem with CMOS circuits that have un-terminated inputs, in other words, inputs that aren't either pulled up to the supply voltage or pulled down to common.

The inputs are such a high input impedance that stray charge builds up on them and biases them into the nether region between the supplies. Usually the input is a totem pole arrangement of two MOS transistors. When the input voltage drifts inbetween the supply and ground, both transistors turn on, and start drawing current from the supply to ground. Not quite a short circuit, but can draw substantial current.

If the charge happens to be drawn off, by air current or what have you, then things go back to normal.

The solution is to make sure that any pins set up to be inputs either have the internal pullup resistors enabled, or put external pull up or pull down resistors on them.

James Newton

Michael is very probably correct.

http://www.sxlist.com/techref/logic/xtrapins.htm

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
---
James Newton, Host of SXList.com
james@sxlist.com 1-619-652-0593 fax:1-208-279-8767
SX FAQ / Code / Tutorials / Documentation:
http://www.sxlist.com Pick faster!

Bean

I can't believe I didn't think of that...
I bet that's it too.
Bean.
·

KenM

Micheal was correct......

Normally I will make unused pins outputs,·and I did not do that for an initial test using the analog comparitor.

After dealing with the unused pins (making them outputs and low) the current consumption was steady.

ken