Advice on using 9v battery

T Chap

For longest possible battery using a 9volt non rechargable and an SX28, would you recommend a 7805, or 5v Zener, or somke other method to conserve the battery? If possible, I was planning to put the SX to sleep unless one of two buttons are pressed. The SX sends a byte out via the Parallax 433mhz transmitter, and there's not much else in the box. I was hoping to hoping to get at least three months out of it as the buttons will hardly get used, maybe ten times a day if that. I felt like the killer would be the 7805.

Bean

Look for a regulator with a low quiescent current.

Bean.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Cheap used 4-digit LED display with driver IC·www.hc4led.com

Low power SD Data Logger www.sddatalogger.com
SX-Video Display Modules www.sxvm.com

There are only two guaranteed ways to become weathy.
Spend less than you make.
Make more than you spend.
·

Guenther Daubach

You might consider using an LM2574-5.0V switching regulator. The only external components required are a 330µH inductor, a Shotky Diode, like an SB150, and two elcaps (22µF and 330µF).

The advantage of a switching regulator is its higher efficiency compared to linear regulators. I'm using the LM2574 in many of my applications with good results. Although this is a step-down regulator, I found out that it even keeps the output voltage stable at 5V when the input voltage is below 5V but I did not test, how much current can be drawn in this case.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Greetings from Germany,

Günther

T Chap

Thanks Gunther I'll order some parts and test it out as well.

Guenther Daubach

Well, in the moment, I'm also working on a battery-powered SX-based device. I'm going to use a 9V battery, like you - so we can possibly share our experiences here. As a matter of fact, I'm also controlling a pair of Parallax 433 MHz receiver/transmitters plus a Parallax serial LCD display with this device.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Greetings from Germany,

G

T Chap

That is great, in that case I have a question already. The Tx has an active low pulldown pin. If the SX is in low power, I don't think it can be pulling anything down in that state right? The Tx goes from 10mA to 3mA in PD mode, and has an internal pullup. I was thinking that if it were not possible to have the SX keep it down when the SXC is in low power, then maybe use a pulldown resistor on it, obviously would need some math to get it to override it's own PU. Then when the SX comes up, have it pull the pin back up. Make sense?

ps Are you going to use internal osc?

Guenther Daubach

Yes, I will also need power down the modules while the SX sleeps, and you are right, the SX can't pull something low or high while sleeping. I have just checked the TX module, and found out that, although pin 8 (PDN) of the LINX TXM-433-LR-S is directly connected to the PDN header pin, you can't measure a meaningful resistance between the PDN header pin and the GND or the 5V header pins. On the other hand, there is a 4.7 kOhm resistor connected between pins 8 (PDN) and 7 (Vcc) of the TXM-433-LR-S. Without checking in detail, I assume that one of the on-board transistors is used to completely cut off the supply voltage of the TXM-433-LR-S. This is why the PDN header pin has a relatively high impedance. IOW, pulling this line low with something like 10 kOhm would nicely keep the Tx quiet while the SX sleeps, and pulling this line high with an SX output would not consume much power, so this is exactly what you and I are looking for. Nice design, Parallax!!!

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Greetings from Germany,

G

Guenther Daubach

Opps,

I clicked the submit button too fast. Concerning your PS: No, I'm going to use a 4 MHz crystal because the internal osc is not stable enough for the 9600 Baud I want to use for RF communication.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Greetings from Germany,

G

T Chap

Thanks again for the info Gunther. Glad to get the good advice. Seems like a fun project, or it wiil be when I get the 433 parts in!

This shows the 7805 which be replaced

Bean

G

Guenther Daubach

Terry (Bean),

to be honest, I never tried it, but I will as I could not find any information in the data sheet.

When you are right, this could cause even a sleeping SX draw remarkable current when its outputs pull low signals that are externally pulled up to Vdd, or vice versa when outputs pull high signals that are externally pulled low to Vss. So, it may be a good idea to configure the I/O pins as outputs before sending the SX to sleep (provided, the external components allow for it).

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Greetings from Germany,

G

T Chap

Any guesses as to the life of battery on an SX (sleep or awake) and some low current reguation?

17mA SX 28 at 4mhz awake
3mA 433 Tx module asleep

---

20mA not including unknown regulation variable

I am no math wiz, but is it correct that 1 amphour = 1 hour usage at the working voltage?

1000 mA draw = 1 amp = 1 hour use
100mA draw = 1 hour x 10
20mA draw = 10 hours x 5 = 50 hours

I may be way off here since I really have no clue to the real formulas, but if this is even close, then I am going to need a power switch on the remote Tx which I had hoped not to use.

T Chap

Here's another thought.... in my case, there are only two buttons that need to be pressed that wont get used but a few times a day. Why not instead of putting the SX to sleep, just have the regulator GND come from one of the buttons. There is no noticable dealy turning on.

Use some code like: if But1 = 0 then Tx some byte

Post Edited (originator99) : 9/6/2006 8:48:21 AM GMT

Guenther Daubach

Hey, why not...

cutting off power completely when no button is down is the best energy-saver I can imagine. The only point is that you may not release the buttons before the SX has done its job sending data via the transmitter. Maybe, a larger elcap across the SX Vdd-Vss pins can hold enough energy to complete the task when a "fast pusher" uses the device. One of these gold-cap parts with very high capacity might do a nice job. I used it many years ago (at pre-SX times) to keep a PIC controller alive for a while.

Concerning your amphour question:

As a rule of thumb, a device with a capacity of one amphour can supply one Ampere for one hour at a voltage within the specified voltage tolerance range. Your calculations are right, to my opinion.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Greetings from Germany,

G

T Chap

Or a feedback led, tell the customer to hold it till the led blinks. Have the led code after the Tx.

Guenther Daubach

Depends on the customers you have. To my experience, customers never read the manuals, never follow the instructions you give, and always push all available buttons at the same time, even when they were told before to only push one button at a time.

This simply follows the first rule of Murphey's Law: "If something can go wrong, it WILL go wrong".

IOW, I'd prefer a "keep-alive" circuit, like a buffer capacitor instead of a blinking LED, consuming additional power.

As a matter of fact, I'm always amazed seeing how many commercially available appliances use an LED as "Battery Low" indicator, adding more surge to the dying batteries .

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Greetings from Germany,

G

T Chap

Yes that never makes sense to make an led come on when the battery is low, I gues the inbverse is worse, have it on until dead? I'll plug in various caps to find a value that keeps it alive for the time required to send the required re-sync pulseout plus the single byte. The sample code shows the BS2 sending out a pulseout, then an "[noparse][[/noparse]" prior to the actual bytes. The Rx side the waits for "[noparse][[/noparse]" and then loads the follwing bytes and debugs them. I don't really get the need to do the "[noparse][[/noparse]", I'll try it without first.

In testing the SX powering up with a switch and then executing led blinking code, the start up delay is virtually unnoticable.

The only reason I thought about the LED was that it gives some feedback as to the battery state. I hate it when somethig isn't working, and I can't tell if it's the battery or not. With an LED, you know if you have a dead battery or not. I have sent remotes flying against a wall for this exact reason!

Mike Green

Discharging a trickle charged capacitor through an LED is a great way to provide an active indication of a weak battery. There's usually enough energy left for quite some time if you only draw microamps.

edgellmh

I have been looking at a low power application and found the following voltage regulator with very low quiescent current (90 uA): LP2954AIT

marshall edgell

Macgman2000

I usually use a bootstrap method. If the system needs to wake up on a button press I use a zener and current pass transistor for the regulator. The switches are momentary double throw, one set of contacts bias the zener regulator to supply the mcu with power. Once it boots up it uses an I/O line to hold the regulator on, long enough to blast out data packets via radio. Then the mcu unlatches and goes back to sleep. No sleep current.

Nick

Bean

Nick,
That's a good idea. But remember that in SLEEP mode all RAM is preserved.
When you lose power, you lose all the RAM conents too.

Bean.

▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
"A government big enough to give you everything you want, is big enough to take away everything you·have."·· Thomas Jefferson

"It is our choices, Harry, that show what we truly are, far more than our abilities."·Dumbledore from Harry Potter

www.iElectronicDesigns.com

·

Macgman2000

Yes very true, I forgot to mention that. LOL......serial EEPROM is an option [noparse];)[/noparse]

Nick

edgellmh

I am wondering if the transmitters you reference in this thread will transmit through walls. That is, would they be able to transmit water level information from my cisterns to a receiver inside the house.

marshall edgell

Macgman2000

The lower the frequency the better the range for the same TX power. If you can legally use low UHF chances are it will make it through walls better than 2.4Ghz. External antennas are also key to long range.

Nick

edgellmh

Do you have a recommended device?

marshall

Lord Steve

Macgman2000 said...
yadda yadda yadda current pass transistor...

What's a current pass transistor?

Macgman2000

marshall,

Try the website below. They offer 300/400 Mhz and 900 Mhz FSK modules.·I use them all the time because they are solid performers.


www.linxtechnologies.com·

Nick

Macgman2000

A current pass transistor is used to handle the current load that a voltage reference can not, basic building block of a voltage regulator. For example. If you place a 5.7v zener at the base of a BJT and use a resistor feedback from the collector to reverse bias it, it produces a 5.7v reference at the transistor base. The emitter side of the transistor is where the 5v regulated output goes·to the load. The transistor acts as a current pass since the zener can not supply high current while maintaining a 5.7v output.

Nick