Extending the live of a 3v 1632- Other options?

Hey guys- I am pretty novice when it comes to frequencies and electrical issues but my company is having a problem with our CR1632 batteries not lasting very long

currently, we are using a "module" that houses a PCB board, accelerometer, Bluetooth low energy, and a battery holder all in one device. The problem we are currently having is we need real time interaction rates (accelerometer) with this module and for our module to talk to our other device WITHOUT a delay(BLE), we need to be sure the module does not go to "sleep." hence the reason the battery dies extremely fast (within a month). We need a potential fix for the battery. We were thinking of tweaking the firmare to utilize a larger battery, though that seems like a pretty extreme cost to do with every module that we develop. I haven't found any solution to rechargeable 1632 batteries and we don't have an option to plug them in at the moment. WE did look at maplin 1632 which states "For use where very low quiescent currents need to be maintained for very long periods with only occasional demands for high current drain" which in theory sounds great, but that might also be every 1632 option.

our device is paired with a tablet the entire time, which of course kills the battery as well.

any tips, tricks, insight? I am not trying to be vague, I hope that this is adequate information to go off of

There is no magic to increase the existing battery capacity I'm afraid but you might be able to reduce the load on the existing one by putting everything to 'sleep' (powered down) except of course the parts needed to wake it up again. In view of your present one month life I suspect you are already doing that. I'm thinking of the trick used by many devices that wake up by a timer maybe once per second check "is there anything to do" before springing back to full operation or sleeping again. I'm not sure how that would work with your constant pairing though. Maybe you could reduce clock speeds so current consumption is lower then go back to full speed when necessary, many microcontrollers have dual or switchable clock speeds these days.

The only other solution would be a bigger battery, you might be able to find a rechargeable one, bearing in mind you could dispense with the 1632 holder to buy back some extra space.

Brian.

Suppose you were to try a wireless power transfer?

* Light powered. (Similar to solar-powered calculators.) Install PV cells on your remote device. Shine a narrow beam of light (or laser?) at them.

* Magnetic (similar to charging hearing aids, electric toothbrushes). Send AC through one coil of wire. A coil nearby picks it up.

betwixt said:

There is no magic to increase the existing battery capacity I'm afraid but you might be able to reduce the load on the existing one by putting everything to 'sleep' (powered down) except of course the parts needed to wake it up again. In view of your present one month life I suspect you are already doing that. I'm thinking of the trick used by many devices that wake up by a timer maybe once per second check "is there anything to do" before springing back to full operation or sleeping again. I'm not sure how that would work with your constant pairing though. Maybe you could reduce clock speeds so current consumption is lower then go back to full speed when necessary, many microcontrollers have dual or switchable clock speeds these days.

The only other solution would be a bigger battery, you might be able to find a rechargeable one, bearing in mind you could dispense with the 1632 holder to buy back some extra space.

Brian.

Click to expand...

Thanks for the quick reply! I am no developer by any means. I am guessing we're essentially shutting down all operations until the accelerometer picks up a signal and says "wake up". We have an option to build out our software and put "operation times" so it goes to sleep after hours, but that is a minute fix. Ideally, the module would be asleep u til it is interacted with, but we really are trying to stay away from the 3-5 second delay from doing that. We really need an instantaneous signal as soon as it is interacted with.

The timer option sounds great, im guessing that is a firmware tweak but yes it does need to stay constantly paired.

- - - Updated - - -

BradtheRad said:

Suppose you were to try a wireless power transfer?

* Light powered. (Similar to solar-powered calculators.) Install PV cells on your remote device. Shine a narrow beam of light (or laser?) at them.

* Magnetic (similar to charging hearing aids, electric toothbrushes). Send AC through one coil of wire. A coil nearby picks it up.

Click to expand...

The only problem I see with these options is size. Our module is currently about the size of a thumbnail and needs to be pretty small. I believe the dimensions are (in MM) 22x25x12.

Do the PV cells need an concentrated light beam or would it work off say, fluorescent bulbs inside of a home? That could be a possibility as we haven't quite finished the casing design.

JeremyWeber said:

Our module is currently about the size of a thumbnail and needs to be pretty small. I believe the dimensions are (in MM) 22x25x12.

Do the PV cells need an concentrated light beam or would it work off say, fluorescent bulbs inside of a home? That could be a possibility as we haven't quite finished the casing design.

Click to expand...

Look into solar powered watches. Their PV panels may be a suitable size and power capacity.

My solar-powered calculator works in room lighting. If lights get dim, the display fades. The four PV cells measure 26mm x 8mm overall.

BradtheRad said:

Look into solar powered watches. Their PV panels may be a suitable size and power capacity.

My solar-powered calculator works in room lighting. If lights get dim, the display fades. The four PV cells measure 26mm x 8mm overall.

Click to expand...

awesome information. Any suggestions as to where we can acquire these panels? This may be an option

- - - Updated - - -

We also have the option of buying our modules without the 1632 battery holder and installing our own. Would the cost that is associated with this be worth the trouble? If there are any good options in California who would be able to assist us with this, that would be awesome.

Does anyone have an idea of the life or suggestion of a higher/longer lasting battery if we do opt to keep our modules awake during operational hours? size could be a bit bigger than the 1632 but we would definitely like it similar in size and comparable in price.

Thank you all for your help! I really appreciate it

betwixt said:

There is no magic to increase the existing battery capacity I'm afraid but you might be able to reduce the load on the existing one by putting everything to 'sleep' (powered down) except of course the parts needed to wake it up again. In view of your present one month life I suspect you are already doing that. I'm thinking of the trick used by many devices that wake up by a timer maybe once per second check "is there anything to do" before springing back to full operation or sleeping again. I'm not sure how that would work with your constant pairing though. Maybe you could reduce clock speeds so current consumption is lower then go back to full speed when necessary, many microcontrollers have dual or switchable clock speeds these days.

The only other solution would be a bigger battery, you might be able to find a rechargeable one, bearing in mind you could dispense with the 1632 holder to buy back some extra space.

Brian.

Click to expand...

This is definitely an intriguing option. Like I said, we can definitely grab the modules without the holder and try a bigger battery. I have been reading a lot on this "blinking" issue basically like the timer you suggested. Is that a software tweak? I am not sure how this would effect the delay in waking up as well. Right now, we are keeping the module awake at all times. The other option was using the timer to turn it completely on every 10 minutes. Does this blink option allow a lower voltage throughout the day and spring into full power rather quickly when the accelerometer is triggered?

I think the answer to periodically waking up and re-hibernating would depend on how long the timeout was on your pairing link. Usually there is a short period between signal loss before the link is declared 'dead' so you might be able to utilize that. The clock speed option would be worth investigating if your processor has the ability to switch it's own speed. The current drawn by the processor will be proportional to the speed it runs at so if you can slow it down until it has work to do, you could reduce overall consumption. For example in a project I worked on, the clock speed was normally 8MHz and the processor consumed about 1.5mA but by switching to a 31KHz 'standby' clock it dropped to just 25uA while still running fast enough to detect when it should be speeded up again. The processor I used had both speeds available internally.

Also, look for rechargable Lithium batteries, you can get them in thin rectangular packaging (like a thick postage stamp) which may allow larger size than a coin cell while still fitting within your board footprint.

Brian.

JeremyWeber said:

awesome information. Any suggestions as to where we can acquire these panels?

Click to expand...

Check electronics and scientific supply websites. Edmund Scientific carries a few small panels. Matching the correct panel to your module will be a combination of carefully measuring Volts and Amps, plus luck.

I once purchased a grab bag of broken solar cells from Edmund Scientific. They are not easy to work with. Brittle. Wiring them in series, I got no success soldering them. I dabbed on silvery liquid from a "conductive pen", which was excellent for that purpose.

We need to update the battery as it's not lasting long enough. Is installing a new battery holder a little more than just soldering? What position/title would we reach out to to find who builds out these systems? Embedded systems? We essentially need someone to install battery holders on an existing PCB board

I fee like the problem is going to be in the firmware. There has to be something working too hard in there for long periods of time. Right now, our battery is losing 40% of its life in 14 hours

The 1632 is a small battery, it's intended use is in very low current consumption applications with occasional higher current bursts. Your Bluetooth module alone is probably drawing more than it can sustain for long.

I appreciate there may be reasons why you might not want to disclose more about the design but if you can give use some clues about the devices it uses and where the finished product will be employed, it will help us assess the best solutions for you. Knowing what type of processor it uses will also help. From what you describe, a new battery holder isn't the best solution, you really need to eliminate the holder altogether and utilize the recovered space to hold a bigger battery.

Brian.

The module will be paired to a device at all times. Here are the specifications that I have available.

OEM Bluetooth Low Energy Platform Module 425 with internal antenna, JST connector, LEDs, temperature sensor, accelerometer and CR1632 battery holder

CC debugger adapter board with configuration jumper. The CC debugger adapter board and JST to JST crossover cable connects the cB-OLP425i-26 module to the CC debugger for programming and debugging.


The module is a UBLOX OLP-425i

- - - Updated - - -

This may be a better answer. Sorry for being a bit cryptic earlier I was In a Rush. The module we are utilizing is the OLP425i. The intended use is inside of a single story office with an open floorplan. The OLP425i-16 gives us the option to purchase the module without a battery, which like you stated gives us the option to build our own battery solution. Is scratching the cr1632 (preassembled on the module) our best bet? Or could it be the firmware factory settings on the OLP425i which is draining the battery so quickly? We really only need the accelerometer, so we could (if possible) turn off all other sensors.

We are looking to build a temporary AAA battery pack on top of our OLP425i. We can get the OLP425i with or without a CR1632 battery holder. The problem is we don't know how or where to solder the AAA battery pack to the module. How complicated is this to do? We are using a 1.5v battery, but don't know if there are any specific restrictions beyond voltage and don't know where to solder these units together. Would anyone be able to show us how to make the connection or point us in the direction of somewhere we could get help

The coin holder has solder tabs. They go to the supply traces on the pcb.

The circumference of the coin is '+' polarity.

- - - Updated - - -

Your photograph of a battery holder appears to have no red and black wires. I guess you'll need to solder on wires.

Use thin flexible stranded type. If you use solid wires, they break after repeated flexing at the solder joint.

Thank you guys so much for the help, we really appreciate all of the information. A couple of questions,

- will the AAA voltage will be okay to use with our module?
- should we use it as an add on to the 1632 or order a board without a battery holder and solder the point straight on to the board?
- I know that you guys are obviously much more advanced at this, -- basic grounding... Would anyone be about to point out The + and - points to solder to?
- The only other worry is possible shorting out and what precautions we need to take in order to be successful attaching the external battery holder?

Thank you again guys. I am doing my due diligence on the side and reading up on some electrical engineering as well.

JeremyWeber said:

- will the AAA voltage will be okay to use with our module?

Click to expand...

The 1632 coin battery is 3V, therefore your circuit must run okay on 3V. So two AAA in series should be okay too, since that makes 3V.

- should we use it as an add on to the 1632 or order a board without a battery holder and solder the point straight on to the board?

Click to expand...

This is risky because volt levels might not match. One battery will try to charge or be charged. A non-rechargeable is liable to start oozing unpleasant stuff if you try to charge it.

However if you can be certain volt levels are equal at the start, then it might be all right as long as all batteries are good. But if one battery starts to underperform, all batteries will lost charge.

- I know that you guys are obviously much more advanced at this, -- basic grounding... Would anyone be about to point out The + and - points to solder to?
- The only other worry is possible shorting out and what precautions we need to take in order to be successful attaching the external battery holder?

Click to expand...

Usually 0V ground occupies the most copper on the pcb. This is not a rule, however. You can't always rely on it.

Can you match the + and - power supply to the battery holder? This is the only reliable method, if you are not familiar with the circuit.

Do you know which pins receive the power supply for the IC? Follow the traces at those pins.

Labels on the pcb. Positive, negative, etc. It's a clue but it may apply to components only, not the power supply.

- - - Updated - - -

Your two AAA batteries will add up to 3V if they are alkaline.

Rechargeable nicad/nimh are lower voltage. Two in series total 2.4-2.7 V. Hence they are not compatible with a 3V coin cell.

BradtheRad said:

The 1632 coin battery is 3V, therefore your circuit must run okay on 3V. So two AAA in series should be okay too, since that makes 3V.
Your two AAA batteries will add up to 3V if they are alkaline.

Click to expand...

Only if they are BRAND NEW!
The CR1632 is lithium and holds up its voltage pretty well as it discharges but Alkaline batteries start at 1.6V each then quickly fade to about 1.2V each which might not be enough to run the circuit.
But two AAA alkaline cells are much bigger and have much more capacity than a tiny CR1632 battery then they will take a fairly long time to discharge to about 1.4V each.