Picaxe power itself down

erco

erco

Senior Member
@Goeytex: Jeepers, your voltage & current numbers are surprising and disappointing. Sounds like an awful lot of internal resistance on the output pins. Makes me wonder if you could even damage an output by shorting it to ground. Now where did I leave my paper clips...?

As I mentioned previously, I have used latching relays sucessfully to provide auto power-off using direct connection to uC I/O pins, although I was using a BASIC Stamp 2. That is also PIC-based, so I assumed a PicAxe would behave similarly. I'm away from home for a few days but when I get home I'll do some tests. Thanks for posting your findings.
 
G

Goeytex

Senior Member
It would be nice if someone could spare a few minutes time & confirm my test results.

Jeepers, your voltage & current numbers are surprising and disappointing.
Click to expand...
You will find "similar" results on most CMOS logic devices. For example a 74HC14N hex inverter with a 5v supply and a with a 330R load ..... the I/O source voltage will drop to ~4v, and instead of the expected 15ma, it will source about 12ma.

It is not reasonable to expect a CMOS logic device ( PIC/Picaxe) to supply a constant voltage on an I/0 pin over its specified current range.

This is why I seldom use a micro to source/sink more than about 5 milliamps.
 
Last edited:
hippy

hippy

Technical Support
Staff member
Goeytex said:
It would be nice if someone could spare a few minutes time & confirm my test results.
Click to expand...
The readings I took on an 08M2, C.2 pin, 5V PSU, output high into 330R, delivered comparable results to your own; pin output voltage was 3.94V and supply current draw increased by around 12mA compared to without the 330R.
 
G

Goeytex

Senior Member
Hmm so to blow your ear drums you could possibly add a transistor and get the full 28ma drive current?
 
D

Dippy

Moderator
Re: post#44

Does it 'jibe'?

You have a buzzer apparently rated as 28mA @ 5V. Is that right?
You power it from a PIC out-pin and it 'takes' 11mA. i.e. less than half the claimed rated value (@5V).
That implies that the pin Vout was a lot lower than 5V when loaded with buzzer.

Have you checked buzzer current on a 5V psu versus manuf claims?
Have you checked buzzer current linearity over a supply range of (e.g.) 3V to 5V?

PIC pins will droop, just as said like logic chip Vout.
An ancient example 4011 (Vdd=5V) suggests 4.6V @ <1mA and 2.5V @ <3.2mA
I find the figures supplied unsurprising and have no probs.
 
geezer88

geezer88

Senior Member
Goeytex said:
Another thing to consider is phantom powering of the Picaxe. Even if you remove the supply voltage, if any of the input pins have a high signal applied to them, the Picaxe may very well continue to operate. The same would be true if the source were removed to power down the Picaxe. Then if any input pins have a low applied to them, the Picaxe may continue to operate. The phantom powering is a result of the connection of internal ESD diodes to the supply/ source rails. Even a pwm signal applied to an input pin can power the Picaxe ( and many other chips with ESD protection) if there is some capacitance ( bypass capacitor)

Putting the Picaxe to sleep will likely not consume any more battery power than the external components necessary to allow the Picaxe to turn itself off.
Click to expand...
I know this is an old thread, but thanks to Goeytex for pointing this out. I've just used the surplus N-channel MOSFETs mentioned in another thread to hold power on a circuit until the Picaxe does it's chores and turns all off. Anyway, watching the current consumption, current would go to zero (1ma scale) and then in about a second, pop back up to 0.1ma. I had left the programming cable plugged into the 08M2. Removing the cable got rid of a phantom source, and now all works well.Gotta pay attention to the details, 'cause the Devil's in there.

tom
 
rq3

rq3

Senior Member
erco said:
Sounds like an awful lot of internal resistance on the output pins. Makes me wonder if you could even damage an output by shorting it to ground. Now where did I leave my paper clips...?
Click to expand...
My own measurements agree pretty closely with the data sheets for the 20M2 (16F1829). The N channel output drivers (sinking current to ground) have an impedance of about 30 ohms. The P channel output drivers (sourcing current to a grounded load) have an impedance of about 90 ohms. Nor is the impedance purely ohmic, it varies with the current, so is somewhat self regulating (but not to be trusted absolutely). In many cases you can connect an LED between two Picaxe pins directly, drive one pin high and the other low, and use no resistor at all.

Rip
 
You must log in or register to reply here.
Top