Member Discussion

Battery Power An Axis Camera

One of my customers needs to deploy some covert cameras (Axis P1224) in his large warehouse sized refrigerator. The temps in one cooler can be -12F and the main cooler is around 20F. Not sure if he is going to deploy in the -12F cooler, if so I might have to look at different options.

I need to power these remotely with a battery power supply. Ideally, we would like to get as much recording time as possible. I will be powering these with the secondary power input - I have 8-28VDC 4.5W max to work with here.

Are there existing battery power supplies available already for this application? I'd like it to be somewhat modular and easy to change and charge batteries. The entire deployment will be moved around quite often.

Suggestions for this application? I don't know if I need to be looking at SLA or Li-ion...?

Thanks in advance.

Battery packs can be sized that give days of recording, but they'll take up space. If covertness is key, you're not going to get much recording time.

You mention portability, but how much space do you have? How is light inside the freezer?

I also assume the batteries will need to be located in the freezer too - or can you run a cable from an outside power source to an inside camera/P12 imager box.

Options like Veracity's PointSource are out there, but they may not be a good fit without some other details.

I need between 8-16 hours of continuous recording. Covertness is critical in this scenario. We like the P12's because I can remote my recording box along with whatever battery source I can come up with.

Here is a pic of the back-side of the shelving system we would be installing the covert camera. I would like to fabricate a box that we can stick on with magnets. Installation has to be very quick and discrete. Lighting is "meh."

Running the cable outside of the fridge is not an option.

Here is a pic of the front of the rack. We are going to use the peg-holes for the sensor.

This is the sub-zero area. Same concept.

Yikes. This is tough. One question:

Would it be possible to place the camera/power sources in an empty food box or storage container?

Not only do we have an issue with capacity, we have a problem with the cold.

The customer would prefer not to have it stored in a box.

That Veracity PointSource looks rad. I think that might address my modular power needs. Looks like it is rated for -4F which might not be quite what we need for the sub-zero area. He doesn't have as much shrinkage in the freezer as he does in the fridge.

I like the portable PoE option vs. the aux 12V. Makes the deployment plug and play.

So two ways to do it:

If you end up making a smallish box, like in your photo, then if you insulate it properly, (think amazon freezer bags, or aerogel) the self-heat from the batteries and the base of the camera might do the trick. If it still gets too cold you can include a simple resistive heating element.

If you end up using the point source and a bigger box, then just use a camera enclosure w heater and run it the camera off the poe and the heater off the aux of the point source.

I have some aerogel insulation lying around as well as a couple li-on 12v supplies, tho I dont have exactly that camera, I'll do a test and let you know...

You have a power and heating requirement. I tried the following Examples with 10hr and 20hr. If using lead acid batteries the enclosure size for 10hr will be 1 cu.ft. for 20hr will be 2 cu.ft. Used insulation factor of R=1 and R=2. Assumed the 20degF freezer, tried target heating to 32deg and 50deg.

3 examples

With 20hr requirement, won't be too portable. But gives some idea of the challenges. Don't let the batteries discharge and freeze, will crack and you will have to start over. Li-Ion will likely improve portability but more costly. If you had two boxes and were substituting, you will need a heavy duty charger to have time to recharge before the other box was drained and frozen. Didn't try the -12F, but I would look at how to get a power chord to it first.

Nice work. A couple questions:

Why so big on the box?

Why so low on the R values?

Have you checked the size of the batteries? 40AH batteries are not small and obviously 150AH would be much larger. 40AH Lead-Acid Battery - L:7.76 (197) W:6.50 (165) H:6.69 (170) ~$100. A 140AH battery from the same source is 13.5" x 6.75" x 11.15". And battery boxes for lead-acid batteries need ventilation.

Li-ion batteries would be pricey. 40AH Li-ion Battery - Length (inches): 7.75" Width (inches): 6.5" Height (inches): 6.87" Weight: 13.5 lbs $580 from Amazon. 150AH Li-ion Battery - Length: 19.30" Height: 9.00" Width: 10.50" $1900

You don't need to ventilate a lead acid battery unless it is the flooded type. Use a sealed type Gel or AGM instead.

You could mount the 150AH battery on a wheeled cart just below the auto-starting gas generator.

The box needs to be big enough for the battery and have room for the insulation??

The R value past 1 gives diminishing returns, the volume for insulation will require a bigger box, hence greater surface area for heat loss. Assume 1 cuft, ambient is 20deg F, enclosure target is 32degF.

R0:37W heating, R0.5:25W, R1: 20W, R2:16W, R3:14W, R4:12W, R6:11W, R10:10W.

Depending on the insulation material chosen will have different R value for unit thickness. R10 will be 10 times the thickness of R1.

Here's something that just might work, the oft-maligned Everfocus Sidekick

Even tho I mocked its thrown together construction, in this case it comes in handy since you can detach the main unit from the router and bag. Its way smaller than the pointsource and its lithium, with over 10000 mah capacity. Its POE will run the camera and you can even use the wifi for live viewing. I have one and will try putting one in an otterbox with insulation and seeing how cold it gets, tho i'm out of liquid nitrogen so I can only go down so low...

You do know that 10,000mAH is only 10AH???

This is the point I think both Robert and I are making. The components need a lot more power than you would think, especially if you also have to heat the battery enclosure. While a 10AH battery should run 4.7W(VA), (at 12VDC approximately .14A), for around 70 hours at room temperature, you do have to de-rate the battery to allow for the low temperature. Below a certain point, you will have to use a substantial portion of the battery to heat it or risk failure.

Robert, my thinking is that the battery should be heated and the heater would be far more efficient if insulation was added to the box.


As long as the battery box temperature is maintained above 32deg F with heating, you don't have to worry about the battery freezing as it drains. The freezing point of a battery rises as the state of charge drops. I found this document on the net:

VRLA Battery Freeze Protection

Cool. Did you see anything similar about Lithium type batteries? I wholeheartedly agree that if lead-acid batteries are used, because the power density is low, the surface area heat loss becomes too great and must be compensated with a dedicated heating element...

You do know that 10,000mAH is only 10AH???

That's the rumor going around. Looking at the back plate its actually 11,600. The ambient temp. has to be > 32F. Do you have a rule of thumb on de-rating the capacity of the battery per degree?

The components need a lot more power than you would think, especially if you also have to heat the battery enclosure.

To start I am not including any other loads besides the camera and the internal load of the battery. Those are my heaters too.

Below a certain point, you will have to use a substantial portion of the battery to heat it or risk failure.

Here's where we differ, I intend to use MOST of the battery to heat the enclosure. The only part unavailable is the electricity that which goes to the sensor and is lost as heat outside.:(

I'm not sure how much is lost to the sensor, but even if it loses half, that's still a 2w heater plus whatever the battery emits. Do you have a guess what that might be?

So, for me, the most elegant solution to the problem is more one centered around containment of a small heat source(camera body, battery) using a high isolation insulation surface, than making a big box with big battery and a big heater. But you may be right, maybe thats the only way...

So what's the wager, casino man? :)

"Cold temperature increases the internal resistance and diminishes the capacity. Batteries that would provide 100 percent capacity at 27°C (80°F) will typically deliver only 50 percent at –18°C (0°F). The capacity decrease is linear with temperature. The capacity decrease is momentary and the level of decline depends on the battery chemistry."

"The performance of all battery chemistries drops drastically at low temperatures. At –20°C (–4°F) most nickel-, lead- and lithium-based batteries stop functioning. Although NiCd can go down to –40°C (-40°F), the permissible discharge is only 0.2C (5-hour rate). Specially built Li- ion brings the operating temperature down to –40°C, but only on discharge and at a reduced discharge."

Battery University - Discharging at High and Low Temperatures


"Cold temperature increases the internal resistance and diminishes the capacity. Batteries that would provide 100 percent capacity at 27°C (80°F) will typically deliver only 50 percent at –18°C (0°F). The capacity decrease is linear with temperature. The capacity decrease is momentary and the level of decline depends on the battery chemistry."

Interestingly, when they say capacity, they are referring to power that the battery can provide at that temperature, in that moment. The power is reduced because the battery is actually draining slower. Of course, at some point the battery cannot produce enough current to satisfy the load requirement.

I am preparing the battery now to go in the freezer, 26F to start, while powering a camera outside of the freezer. I have a temp probe inside the enclosure, lets see how cold it gets!

(Photo after 8 hr)

Ok, so this is the update of the kitchen test.

Here's the ingredients: 

1. Amanda side by side freezer, which according to its word was to 23F -5C

2. 1 Everfocus Sidekick in original pouch, with wifi router disconnected, flat ethernet cable on poe port

3. One Axis M3007 wide angle 3MP, recording 1fs mjpeg to micro SD card

4. One Taylor probe "turkey-style" thermometer with probe taped against outside of side kick metal chassis

5. One POE power panel poe tester inserted between sidekick and camera

What I did

Setup camera for edge recording using 8 gb micro sd card

Placed sidekick in a freezer bag (to reduce moisture), NO other insulation used.

Turned on power to sidekick, Placed side kick in freezer on box of jimmy dean sausage which was itself on a bed of ice, ran flat ethernet and temp probe out to top of coffee maker workbench, closed freezer

watched temperature go from 70F to 42F in one hour, camera still running

went to sleep

7 hours later, 8 hours total, went and retrieved sidekick camera still running, temp was at 32F or lower, photos below.

one blue light still lit on battery power indicator of sidekick

This test was performed without even the heat of the camera body or any extra insulation, (besides freezer bag).

If anyone cares I can redo the test under more stringent conditions and using better equipment, e.g., use a thermometer that goes below 32F, doh!

 (8hr stalagmite formation)



 (8 hr Warming up)

 (8hr, Out of bag, warming up, one battery blue still on...

Rukmini thanks for doing this, I like real testing validating theories. I am finding it difficult to follow with the pictures. Could you show a diagram of what is in the fridge and the interconnections.. I didnt see the battery, can I assume you are powering it remotely from a chord to a powersupply outside thw freezer? What are the inside dimensions of the insulated bag and, thickness of insulation of the bag? What is the power being fed into the devices in the bag? You seem to be inferring that you cant measure accurately the temp inside freezer, but relying on the manual? Rob

Sure, I'm putting together a better timeline/explanation now...

Just to quickly answer a couple things:

All the pictures are from 8 hrs+ after the start. They are in chronological order and all were taken within 5 minutes of eachother

the first one taken is at the 8hr mark before opening the freezer door, showing the outside components, (the camera, the poe power meter, the temp display)

the next one is from inside the freezer door, of just the inside items, the freezer bag containing the sidekick battery and pouch.

the next one is taken after removing the sidekick from the freezer and placing it with the other components

the next one is after opening the bag to show the panel of sidekick

the temp inside the freezer said -5c according to its own led. Also disregard the temp(58F)on the right side of the taylor temp, its just an alarm temp

@Robert, @Carl in this special case, since one way or another we are only heating a battery with its own power, what do you think about using a thermistor in parallel with the load? One obviously that would lower its resistance as the temp went down, thereby increasing the current and keeping the battery temp. over the threshold. It would be more efficient than having any heating element, no? Thoughts?

P.S. Do you know the old trick of when your car doesn't start because of the cold, you put your headlights on for a min, and then you can usually start it?

I am trying to do the same thing with the same camera, although it will not be in a freezer. Does any one have any suggestions for a small battery to power the camera for 48 hours?

I wonder if this thing could be hacked to work. It's small and powerful- 61,000 mAh.

No hack necessary, just this.


2 Caveats: 1. Don't plug the output into a non-poe device

                 2. When comparing batteries, amp hours are not watt hours. i.e., 5v and 10000 mAH is less power than 12v and 10000 mAH.

With that kind of power density, plus the heat that the DC-DC converter will squander, all you need from what I've seen is a small tight insulated enclosure, to keep it warm enough.  Good find, Ari! 

Disclaimer: I've used Tycon's USB to 24V passive POE one, but not this 48v one.