Battery Consumption Calculation

Well these are usually designed with a comfortable safety margin...not all batteries are created equal...I would suggest field testing the battery under accelerated use... Will these batteries be recharged with solar panel?...

https://www.mdpi.com/2076-3417/9/3/559/pdf

https://www.dfrsolutions.com/blog/how-to-select-the-right-battery-for-your-application-part-1-battery-metric-considerations

..."Do Not fully discharge Lithium Polymer batteries. The low voltage limit for discharge is 3.0 volts per cell. Allowing the cell to drop below this voltage will irreversibly damage its internal chemistry."...

..." Lithium Polymer cells experience a self-discharge rate of approximately 5% per month"...

http://www.manoonpong.com/Other/main_page=page_2.pdf

https://en.wikipedia.org/wiki/Self-discharge

https://www.eastbayrc.org/index.php/tim-s-tips/78-lipos-and-cold-temperature-operation

In the line (99.83% * 83uA) + (0.17% * 247mA) = 0.42mAh, the arithmetic is a bit off;
it should be 99.83*0.083mA + 0.17*247mA ≈ 0.503mA.

The rounded-off %'s cause a slight overestimate; it's really closer to 0.495mA; call it 0.5mA.

6600mAh/0.5mA = 13200 hr. (Then 13200/8766 ≈ 1.5 yr). A 56aH battery would be almost 3x what you need.

You MUST get data from the specific battery manufacturer, it should show various curves for operation under differing loads and loads of other useful information.
You can speed up your testing by a making it transmit more frequently, say every 5 minutes, and maybe add an extra load to increase the quiescent current proportionally.
It won't be quite the same, but it's better than nothing.
Alternatively, or, in addition to the above, check on the construction of the battery. You may be able to test using a smaller batter and interpolate the results if the two batteries just scaled versions of each other.
If you can't make the equipment perform how you want for testing, just make a dummy load that can be switched in with a simple timer. The dummy load can be scaled to give say 50 times the load... don't scale it up too much else the characteristics will be too different.
Once you have an estimated battery life... half it, as the performance of batteries is very variable ! (trust me)
Del
PS. I've been in this situation and by getting good data and being very certain of my estimated 2 year battery life (we quoted 1 year) any problems of short life could be confidently attributed to faults (usually 1 cell in a pack of 4 being bad)

Thanks for the response del.. this is what I was going to do, so you’re saying halve the measured time to account for the self discharging aspect and variability between batteries?

Also I take it you can parallel LiPO batteries without any risk..

Thanks for the replies. No solar panel. I think I’ll just account for the self discharge rate in the initial sizing.

If the discharge rate is say 5% per month, is the following calculation correct?

% Of hour transmitting = (12s / 3600s)*0.5 = 0.17%

% Of hour idle = 100% - 0.17% = 99.83%

Self discharge = (.05/30/24) x 6600 = .45mA

Total amp consumption per hour = (99.83% * 83uA) + (0.17% * 247mA) + 0.45mA ~= 1mAh

On a 6600mAh battery, this would imply the device would last 6600 hours or around 275 days?

Don't forget to make allowances for temperature....this is a variable, so go with worst case scenario...

Yup, that's another one!
It's all these little things that add up and we all know that factors always conspire to make any situation worse! Hence my advice to half the calculated battery life.
Del

I would use dry cells.

Yup, that's what I did 4 alkaline C cells packs a lot of mAh ... some others, just meh
Del

Thanks for the responses.

I have re-completed my battery calculations with the above mentioned considerations and I require 18000mAh for my applications.

Just to add some context, I am going to be distributing this device and using it in the industrial world, so I need to implement a battery supply which is robust and most importantly safe.

My understanding of batteries is that having them wired up in series is much safer than wiring up in parallel:

- A shorted cell in parallel will lead to potential fire risk as the others will discharge high currents. An open cell in parallel will just mean loss of capacity.

- A shorted cell in series just means a lower voltage output. An open circuit in series will mean zero output.

Hence, the only risk I am aware of is wiring in parallel and experiencing a shorted cell. Am I correct?

I am reluctant to use parallel combinations of batteries without installing a whole bunch of fuses to protect from high currents with shorted cells. So we are left with installing batteries in series.

I have found some D-sized alkaline batteries at 18A/h capacity (output voltage =1.5V). I can see that typically the alkaline batteries supply this level of current up until about output voltage = 1V.

My device has an onboard voltage regulator and let's assume it can regulate voltages between 3-6V (required operating voltage = 3.7V). Is my safest best bet to install 4 of these 18A/h capacity batteries in series and let the voltage regulator compensate for the voltage fluctuation down the end of life voltage of 1V? This should ensure the device will until the batteries are dead.

Yup, that sounds like pretty much what I did (I used 4 C cells in series.) If you use a low drop out regulator you'll be able to get good battery life.
I think I used a Maxim part, they have some good power supply chips.
I liaised with a battery supplier to get the 4 packs built up with a connector and thermal trip installed.
Del

Why wouldn't you just go with 6 volt lantern batteries...?

Connectors, cost, lack of thermal trip.

After much more research, I've concluded the best battery for my device would be a Lithium Thionyl battery, due to temperature robustness, cost and capacity.

Tardian make a single D type 19A/h which has a continuous operating current of 360mA.

https://tadiranbatteries.de/pdf/lithium-thionyl-chloride-batteries/SL-2780.pdf

It seems your voltage will run low during the winter temps...

Yes I am hoping the voltage regulator will take care of that though :)

Actually, that won’t work either. The voltage regulator becomes inefficient and wasteful when converting voltages up or down.

Those batteries have high internal resistance and hence are subject to varying voltage swings at different temperatures.

Guess the only option is lipo’s in parallel with some kind of protection circuitry.

Does anyone know of any manufacturers who make 19Ah+ single cell lipo batteries? Or anyone who manufactures battery holders with thermal protection against paralleling lipo batteries?

If you are only looking at a life of 1-2 years, with self-discharge equal to load amp-hour or more, do not discount the common "duracell" primary battery [lower fire risk] or lithium primary cells [which can have very low self-discharge, even at 40-50 Celsius].

"Duracell/Alkaline" have a shelf life of 10 years [so self-discharge must be low], size/weight compared to rechargeable cells is favourable; replacements are easy for customer to obtain anywhere in world. When it comes to re-cycling/disposal they can be disposed of at a supermarket rather than specialist hazard handler.

After 2 years on self-discharge, few secondary cells will have a reliable capacity/life after recharge [ask manufacturer].

67model

https://www.google.com/search?q=impact+of+pulsed+discharge+on+lifepo

https://www.google.com/search?q=electrodes+of+pulsed+discharge+on+li%2Bfe%2Bpo → ?? http://jes.ecsdl.org/content/164/11/E3040.abstract ??