How to design something that can last long (years)?

Go To Last Post
29 posts / 0 new
Author
Message
#1
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I'm trying to design a device that runs on battery that hopefully can run 20 years (change battery every 5 years is good too, but try to avoid that). The device is going to operate under the sun. what should aspect of the design should i be care. Let's say disregard the power consumption of mcu chip.

Battery selection: Battery Deration curve?
Voltage regulator: low quesiant current.
Temperauture that affect battery self discharge rate and leakage current.

Is there any current consumption stuff that i should be aware of that may consume power that i may not be aware of?

somebody suggest a vibration or an antenna to harvast some energy, or even solar panels. (Birds feces affect the solar panel a great deal too if i do solar panel)

Last Edited: Sun. Jun 20, 2010 - 05:35 AM
  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Don't forget that your battery has a self-discharge. Have you considered energy harvesting (from vibrations or an antenna) instead?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I don't think most batteries last 20 years even if not used. Some are prone to leakage if past their expiration date, especially alkalines.

Of course to get many years on a battery, you need low current consumption, or a very large battery. I think any voltage regulator would draw too much current. I think battery operated devices that run a long time between battery changes don't use regulators.

Alkaline cells drop their voltage a lot as they discharge, but other types of cells don't, like lithiums.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Leakage and very long MTBF for all components.

Ross McKenzie ValuSoft Melbourne Australia

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

timgoh0 wrote:
Don't forget that your battery has a self-discharge. Have you considered energy harvesting (from vibrations or an antenna) instead?

How do i get energy from vibration or an antenna?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

steve17 wrote:
I don't think most batteries last 20 years even if not used. Some are prone to leakage if past their expiration date, especially alkalines.

Of course to get many years on a battery, you need low current consumption, or a very large battery. I think any voltage regulator would draw too much current. I think battery operated devices that run a long time between battery changes don't use regulators.

Alkaline cells drop their voltage a lot as they discharge, but other types of cells don't, like lithiums.

Interesting, do they have a very low self discharge rate like alkaline?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

valusoft wrote:
Leakage and very long MTBF for all components.

True, but what kind of things i could do to minimize that?

BTW, the device is goinng to be operate under the sun. So i know Temperature will affect the discharge rate significantly too for a long time.

I know temp also affect the leakage current for a lot of electronics, is there anything else i should be put in a lot of consideration for?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Operate the components in a very conservative part of their ratings ... in other words do not dissipate 0.999 watts in a 1 watt rated item.

Ross McKenzie ValuSoft Melbourne Australia

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

valusoft wrote:
Operate the components in a very conservative part of their ratings ... in other words do not dissipate 0.999 watts in a 1 watt rated item.

That sounds good! Thanks Ross

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

With sun, and 20 year lifetime, lithium carbon-monofluoride is a good place to start. And currently I think, also where you end up.
They will work over a -40/+85 range, and self-discharge 10% over 10 years.
A 1Ah coin cell will give you 5uA to work with, though if you pulse the cell with higher load, your battery life goes down faster than you expect.

No electrolytic capacitors can be used. And packaging will be important; air will expand and contract, and moisture is going to kill the battery quickly.

/Kasper

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

cheungupdt wrote:
timgoh0 wrote:
Don't forget that your battery has a self-discharge. Have you considered energy harvesting (from vibrations or an antenna) instead?

How do i get energy from vibration or an antenna?

There are products such as those by http://www.adaptivenergy.com/. Linear Tech also makes the LTC3588 for piezoelectric elements, or the LTC3108 for TEG or possibly large rectified RF antennas.

Though your BOM cost may be quite expensive. Also, like the above posters have said, watch the reliability of the other components.

(Disclaimer: I've not used any of the above products in high reliability applications, so please contact the appropriate people and evaluate them carefully)

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

There's http://en.wikipedia.org/wiki/Rad... . It can last a LONG time.

There are pointy haired bald people.
Time flies when you have a bad prescaler selected.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Yes, they don't have an expiration date like chemical cells. They just have a half-life. ;)

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

KKP knows a lot more about lithium cells than I do. I know there are several "lithium" cells, each with a different chemistry. The ones I know all have a long shelf life, a good discharge curve and are as light as a feather.

The CR123A cell is a manganese dioxide lithium cell roughly the size of an AA cell. They are designed more for power than long life, but Panasonic guarantees they will have 80% of their original capacity after sitting on the shelf for 10 years.

Actually "shelf life" depends on their temperature. If they spend a lot of time under the hot sun, their shelf life will be considerably less. If they spend a lot of time in cold weather, their shelf life will be greater.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

daqq wrote:
There's http://en.wikipedia.org/wiki/Rad... . It can last a LONG time.
There are 3 Plessey automatic weather stations in the Rub al Khali desert of Saudi Arabia that in the late 70s and 80s (presumably still) were powered with Strontium-90 RTGs. They had HF transmitters and "sucked" heaps of power during transmissions ... compared to today's satellite comms.

Ross McKenzie ValuSoft Melbourne Australia

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

wow! Such a wealth of knowledge here! I'm so excited to hear so many stuff that is being suggested here. keep it coming! Thank you so much!!!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

There are quite a few types of Lithium batteries, most of them not known to most of us I guess, some are for speciality applications and probably very difficult to get.

Look here http://en.wikipedia.org/wiki/Lithium_battery

The Carbon Monofluoride one looks interesting :)

And why not use some solar cells?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

We have battery operated devices... and have been using the 1/2AA Lithiums. Roughly 1200maH and good 80ma pulse current. You will have to make sure your battery can provide enough current to run your circuit... the coin cells may or may not be able to... and typically have a maximum pulse current of 20ma (2477T can do 30ma) A battery will probably lose roughly 10% of it's life per year. So coin cells may not be for you depending on how much energy you need at any given moment.

To save lots of energy you have to duty cycle your processor... shut it down and wake it up only when it needs to be awake... and only power parts of the chip you need.

Company called Cymbet has energy harvesting chips which also have a built-in thin film lithium battery. You should be able to get more than enough energy from sunlight.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Another factor to consider is battery capacity. A battery can lose as much as half its capacity when it gets very cold. It will regain that 'lost' capacity once it warms up again, but that won't be much help if the system runs out of power at 4am on a cold winter night.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

I'm wondering why nobody mentioned solar cells. As the device is operating 'under the sun' a small solar cell seems natural to me. In addition they last a long time and are designed to be exposed to the sun. You can satisfy a large part of your energy needs with those.

Markus

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

If your energy source can produce at least 4 or 6 volts, try this:
http://www.aldinc.com/ald_energyharvesting.htm
with a low power-usage device. Advanced Linear Devices may soon have one that can run with one PV cell:
http://www.aldinc.com/ald_prultralowvoltage.htm
Linear Technology has energy harvesters:
http://parametric.linear.com/html/energy_harvesting

"Dare to be naïve." - Buckminster Fuller

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

So will the OP be telling us what exactly he's planning to design that requires a power source for that long?

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Quote:
The device is going to operate under the sun.
Designing solar recharge for a system is a well understood engineering task. Essentially, you start with the published insolation statistics for your locale, adjust for special conditions at the specific place of use. Next of course comes the load calculation, average, peak, etc. And the time durations: on 24/7/365? Seasonal insolation factors, etc.

Then a solar panel, type of battery, and power conditioner is selected, being DC-DC low power up to big panel with many watts.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Didn't the OP mention that birds might be a problem with solar panels? I have never played around seriously with them, but how big a problem does birds incur on solar powered devices? How is it handled "in real life"? Are there people going out there cleaning the panels up? If so, (in this case,) changing batteries every now and then sounds easier and cleaner...
I guess you could mount "bird deterrent" spikes, so they don't sit down on your panel and do their business, but it won't affect the random fly-by droppings.

I'm also curious as to what this device will be used for... From the info given, it seems like remote logging in an inaccessible location, of some kind. You can't give us a crumb and not share the cookie! :D

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

TrainzStoffe wrote:
... how big a problem does birds incur on solar powered devices? How is it handled "in real life"?
Yes birds can be a very serious problem, so much so that I have seen installations with the panels facing down towards a reflective surface, eg sand just so to avoid the bird "dressing". Obviously there were many more panels needed compared with a direct sunlight path.

Ross McKenzie ValuSoft Melbourne Australia

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

Also have your PCB coated with a conformal coating to protect from moisture.

Consider these for your caps...

http://www.paktron.com/pdf/Advantages.pdf

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

TrainzStoffe wrote:
how big a problem does birds incur on solar powered devices? How is it handled "in real life"?
That's why rain was invented. It helps if it's not on the equator so the panels will slope.

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

alwelch wrote:
Also have your PCB coated with a conformal coating to protect from moisture.

Consider these for your caps...

http://www.paktron.com/pdf/Advantages.pdf

Thanks, that's a great suggestion!

  • 1
  • 2
  • 3
  • 4
  • 5
Total votes: 0

steve17 wrote:
TrainzStoffe wrote:
how big a problem does birds incur on solar powered devices? How is it handled "in real life"?
That's why rain was invented. It helps if it's not on the equator so the panels will slope.

rain doesn't wash stuff well, if the rain is small, it will put moisture on the solar panel, and it'll collect dust and form a dust layer that is semipermanent if there is not a proper cleaning.