POR circuit - Xbee & 08M

stocky6409

Senior Member
Hi All

Still working on my little Xbee comms project ad am testing power scenarios.
Came across a little issue when the battery voltage gets very low.

The Picaxe 08m copes fine - but the Xbee tries to draw 50mA @ startup (before it can be disabled by the pin sleep mode).....which sort of stops the charging cct from ever getting the volts up enough to start the Picaxe which disables the Xbee......arrgghhhh

I have noted that if you hold the xbee RESET line LOW it doesnt "boot" and hence draw 50mA

so i think i have two options:

1) use a bigger picaxe with more i/o and switch the power to the xbee with the picaxe
2) some type of BOD or POR cct that holds the reset line of the Xbee LOW unitl VCC rises sufficiently.

The bigger Picaxe might be nice as a couple more i/o and freeing up an ADC lets me "self monitor" my battery status to limit power consumption during extreme low battery (ie disable xbee totally until chareg level improves).

Whats better at low voltage operation and low current consumption 14M or 20M?

Suggestions?


Stocky :)
 

westaust55

Moderator
14M vs 20M for varying Vcc vs Clock Freq

from the datasheets, both the 14M and 20M have the same low voltage capabilities vs clock frequency, namely:

Supply Voltage vs Clock Frequency
2.0 - 5.5 V -- FOSC < = 8 MHz: HFINTOSC, EC
2.0 - 5.5 V -- FOSC < = 4 MHz
3.0 - 5.5 V -- FOSC < = 10 MHz
4.5 - 5.5 V -- FOSC < = 20 MHz

I doubt there is any significant different in the current / power consumption of these two chip given the same supply voltage and clock frequency.
That will be more a funciton of the other loads/modules in the circuit/project.
 

westaust55

Moderator
Hi Stocky,

If you look at PIACXE Manual 1 (current V6.9) at page 13, under the heading:
PICAXE chip labels
you will see which PIC chip is used for which PICAXE part.

Then as Andrew says, it is off the the microchip website for the PIC datasheets.

While there is a lot of data in threads here by hippy and others including myself, it is still a good idea to download those datsheets.
I have virtually every PIC chip used for a PICAXE part on my PC's along with lots of other IC datsheets.
 

Dippy

Moderator
I agree.

And it is not unknown for:-
1. PICs to have different Electrical Specs between devices i.e.one size doesn't always fit all.
2. Microchip to change/modify specs after a while.
3. Microchip to make mistakes and put errors/workarounds in their Errata Sheets.

I can be time consuming and tedious, but it's always best to get the info straight from the horse's mouth, so to speak.
And, generally speaking, I would have thought that , where possible, getting data direct from the 'source' is a good idea.
 

slurp

Senior Member
I'm just playing with brown-out code for differing reasons, trying to stop a robot sitting round waiting for a user input when it's power is failing.... get it to crawl home as best it can.

The PCON Register is good for this. There's a little code here for power on vs. reset, if you insert a check on bit0 (after checking the power on reset condition) you can find out if there's been an brown out (the bit will need setting in software).

The PCON Register has been consistant between 08M and 18X picaxes, I hope the (microchip) manual give you a good context to the use of this.

regards,
colin
 

westaust55

Moderator
I'm just playing with brown-out code for differing reasons, trying to stop a robot sitting round waiting for a user input when it's power is failing.... get it to crawl home as best it can.
good concept. In the automotive world, that is what they refer to as:
"creep home mode"
 

slurp

Senior Member
good concept. In the automotive world, that is what they refer to as:
"creep home mode"
In this case I don't really have a creep/crawl strategy... I'll try to avoid the problem by specifying a suitable battery for the load/time of operation!

Usual process is to drop into startup conditions when power comes on and when the reset is pushed (18X ciruit). We'd have the operator positioning before the start button is pressed. On the occasions we've pushed the battery too far I don't want to be sitting around, I'd rather skip the start-up bit and try to keep driving on the basis of the sensors.

As the Brown-Out is usually under load so the position is one of desperation! It's unlikely to be of great benefit... but I'll give it a try ;)

regards,
Colin
 

stocky6409

Senior Member
My answer to the xbee POR issue was staring me in the face - I'm using a Micrel MIC5205 regulator and that has an enable pin - DOH!

So - I'm just going to use 2 of them - with one dedicated to the xbee having the enable pin switched by the picaxe. That way if the picaxe senses the battery voltage is too low it can hold off enabling the Xbee till the charge improves!
 

ciseco

Senior Member
You could use a simple transistor to switch the whole thing on/off through path to ground (cheaper than another reg). Seems to be ok for me, no need for a FET. This has been the only way to get the thing to go "really" low. The builtin xbee sleep modes are too great. We've got a lightswitch that sleeps waiting for interupts at around (dippy might correct me) 4ua. It's on a 16f688 (somewhat similiar to a 14m). We think it should last 6 years at 20 seconds of radio transmissions a day on a 2/3 AA 3v cell.

Just a thought.

Miles
________
buy digital scale
 
Last edited:
Top