Is there anyway to SAFELY heat a battery box?

[spike001]

Here's one I made up that is posted here.
It does not take into account extreme cold and heat on the batteries.
http://www.solarpaneltalk.com/showth...Battery-Design

Thanks it's going into my reference file!

[Naptown]

Love your honesty....lol....but frankly I would have been just as happy to be wrong. I am seriously considering putting together a much smaller summer version.....at least I can verify the proof of concept. Interestingly the Canadian government tendered an RFQ about a month or so go for the far Arctic for something very close to this (major difference being satellite (not cellular) and that they are willing to pay to send someone up there on schedule to maintain it.....I took a quick look at it and figured that it will run them about $1M (ballpark)once they get the thing designed, depolyed, managed and maintained. Plus it is government....so our money.

If you don't mind - I just have one last question for you. Do you use a tool to make your design calcls? or do you depend on something likes the worksheets that SEI provides?

Otherwise great working with you - I was really looking forward to working with the Concorde AGM product - perhaps my next project! spike

Here's one I made up that is posted here.
It does not take into account extreme cold and heat on the batteries.

http://www.solarpaneltalk.com/showthread.php?5334-Solar-Off-Grid-Battery-Design

[spike001]

No you got it right. You are screwed. You are looking at a 1000 watt panel wattage, 80 amp controller, and 400 AH 12 volt AGM battery.

Good news is summer will be no problem.

Love your honesty....lol....but frankly I would have been just as happy to be wrong. I am seriously considering putting together a much smaller summer version.....at least I can verify the proof of concept. Interestingly the Canadian government tendered an RFQ about a month or so go for the far Arctic for something very close to this (major difference being satellite (not cellular) and that they are willing to pay to send someone up there on schedule to maintain it.....I took a quick look at it and figured that it will run them about $1M (ballpark)once they get the thing designed, depolyed, managed and maintained. Plus it is government....so our money.

If you don't mind - I just have one last question for you. Do you use a tool to make your design calcls? or do you depend on something likes the worksheets that SEI provides?

Otherwise great working with you - I was really looking forward to working with the Concorde AGM product - perhaps my next project! spike

[Sunking]

Hi Sunking - info below as requested.

Location: Latitude 46.2 degrees North by Longitude 70.0W

The daily Watt hour consumption is 566 Watts.

The worse month of for insolation is Dec @1.15 kWh/m^2 per day

When I used the formulas provided by concorde and factored in 10 days of automony a their design factor based on temp (which I think was approximately 2.25 or thereabouts) that battery bank got quite large. I also ran the numbers throught the works sheets provided by SEI and received similar results. Thoughts? Am I missing something? Thanks Sunking, spike

No you got it right. You are screwed. You are looking at a 1000 watt panel wattage, 80 amp controller, and 400 AH 12 volt AGM battery.

Good news is summer will be no problem.

[spike001]

Spike what is the:

Location
Daily watt hour consumption

Hi Sunking - info below as requested.

Location: Latitude 46.2 degrees North by Longitude 70.0W

The daily Watt hour consumption is 566 Watts.

The worse month of for insolation is Dec @1.15 kWh/m^2 per day

When I used the formulas provided by concorde and factored in 10 days of automony a their design factor based on temp (which I think was approximately 2.25 or thereabouts) that battery bank got quite large. I also ran the numbers throught the works sheets provided by SEI and received similar results. Thoughts? Am I missing something? Thanks Sunking, spike

[Sunking]

Spike what is the:

Location
Daily watt hour consumption

[spike001]

Some questions for you to ponder.

While AGM's can handle low temperatures better than flooded, their chemistry still fails and your amp hour capacity degrades as temperature goes low.

So what is the plan for the 11th cloudy day? Will you have a generator to save the batteries ? Even AGM batteries will begin to sulfate when below 70% of full.

Massively parallel banks are not good. Consider upping the system voltage to 24, or even 48V to avoid parallel strings.

Extreme cold will temperature cycle/de-torque the fasteners, and you may need SS spring washers to keep tight terminal connections. I've not had experience with lead terminals in extreme cold. DO make sure your controllers can handle the cold environment.
And maybe redundant PV arrays would be called for, as the cold will really be stressing the tab connections in the panels. Again, no experience here, but work with spacecraft makes me think of all the things that happen in cold environments.

As for thermal control, the freeze point of water, requires/releases a lot of BTU's to cycle from 31 - 33F, and there are lots of phase change thermal materials used in the 60-90F range, you may want to research if there is anything usable in your critical range. Remember, once the electrolyte in a battery freezes, the battery is unusable till the whole thing thaws. If you pump power into it, only small areas will thaw, then boil and vent, depleting your electrolyte reserves at best, and destroying the battery at worse.

I wonder if super caps will work at these low temps ? What about other battery chemistry ? Any primary batteries (zinc air) ?

Lots of good items to take into consideration. My immediate problem is the impact days of autonomy and temperature are having on the size, and hence, cost of a quality battery bank. At this point I may just have to roll back the project objectives and go with a less costly set of goals as a proof of concept....in warm weather.

[Mike90250]

Some questions for you to ponder.

While AGM's can handle low temperatures better than flooded, their chemistry still fails and your amp hour capacity degrades as temperature goes low.

So what is the plan for the 11th cloudy day? Will you have a generator to save the batteries ? Even AGM batteries will begin to sulfate when below 70% of full.

Massively parallel banks are not good. Consider upping the system voltage to 24, or even 48V to avoid parallel strings.

Extreme cold will temperature cycle/de-torque the fasteners, and you may need SS spring washers to keep tight terminal connections. I've not had experience with lead terminals in extreme cold. DO make sure your controllers can handle the cold environment.
And maybe redundant PV arrays would be called for, as the cold will really be stressing the tab connections in the panels. Again, no experience here, but work with spacecraft makes me think of all the things that happen in cold environments.

As for thermal control, the freeze point of water, requires/releases a lot of BTU's to cycle from 31 - 33F, and there are lots of phase change thermal materials used in the 60-90F range, you may want to research if there is anything usable in your critical range. Remember, once the electrolyte in a battery freezes, the battery is unusable till the whole thing thaws. If you pump power into it, only small areas will thaw, then boil and vent, depleting your electrolyte reserves at best, and destroying the battery at worse.

I wonder if super caps will work at these low temps ? What about other battery chemistry ? Any primary batteries (zinc air) ?

[inetdog]

But you also bring up a good point. I have two sets of CCs at this point and need to investigate all of the functions. One has an RJ45 jack which I assume....I mean hope that I can plug directly into my hub so that I can pull data from it directly.

Chances are very good that the RJ-45 jack is a serial bus interface with differential send and receive drivers (such as Modbus) rather than Ethernet. But some CCs and inverters do have the option for an Ethernet interface card to be installed instead. Or you can just get a USB to Modbus adapter and talk directly using the appropriate software. Some of the members have discussed Open Source software that they are working on for this purpose.

[spike001]

Since you will have a data transmission link, I assume that you have already planned for the battery voltage and temperature to be recorded and transmitted too? To give you a good handle on the state of the system, an indication of the panel current would be very nice to have. The panel voltage will not change much as a function of light level, but the current will. The voltage will change as a function of temperature and the Vmp chosen by the CC will vary considerably if there is partial shading of a series string.
If possible the data transmission should be done toward the end of the hours the panels are producing so that the extra power drain can be provided directly by the panels rather than cycling the batteries. (Assuming that data transmission will take significant power.)

Sorry if I am hitting things you have already planned for, I am just free-associating here.

No problem on the free associating.....I'd rather have that than an "oops" moment. But you also bring up a good point. I have two sets of CCs at this point and need to investigate all of the functions. One has an RJ45 jack which I assume....I mean hope that I can plug directly into my hub so that I can pull data from it directly. I plan on doing most of my "communicating" with the set up mid-day so that I can take advantage of the fact that the panels are producing some power. Since the sun sets at around 4pm I'll most likely let the system idle/sleep unless I have a necessity to wake it up. Surprisingly the power required to run the cellular hub and transmits is approximately 18W....so it isn't terrible. But I'll be able to validate all of these factors once I get the system connected end-to-end. Thanks for the free association.

[inetdog]

AAlthough the surveillance software is quite sophisticated the cameras will be managed from Toronto - I've allocated 3 hours per day for transmission.

Since you will have a data transmission link, I assume that you have already planned for the battery voltage and temperature to be recorded and transmitted too? To give you a good handle on the state of the system, an indication of the panel current would be very nice to have. The panel voltage will not change much as a function of light level, but the current will. The voltage will change as a function of temperature and the Vmp chosen by the CC will vary considerably if there is partial shading of a series string.
If possible the data transmission should be done toward the end of the hours the panels are producing so that the extra power drain can be provided directly by the panels rather than cycling the batteries. (Assuming that data transmission will take significant power.)

Sorry if I am hitting things you have already planned for, I am just free-associating here.

[Naptown]

You do have one thing going for you. You stated that you are in the woods in front of a lake and clear from east to west. Being in the great white north by setting your panels to a very steep angle (maybe not vertical but close to it) you could take advantage of both snow shedding and the albido effect of reflection off of snow or ice.

[spike001]

Since by its nature the system is mission critical to you may want to consider what to do when the battery voltage (or temperature) indicates that you are getting into trouble. Rather than run things into the ground and maybe never come back up again, consider a fallback (survival mode) in which you run the surveillance only 5 minutes out of 15 or some such ratio (choosing the time and the duty cycle based on how long it takes to come back online again after switching off. If you can find a dependable motion detection circuit for the camera, you can have it switch to continuous recording if it detects motion during one of its on periods.
Get creative with your load management too, and you may find that you can spend less on your battery bank and get better survivability.
(Of course if you think that the added complexity is more likely to cause problems than solve them, that is a decision you have to make.)

Actually you couldn't be more correct. At this point I've divided up all of the functions of the devices I am using and calculated the wattage required for a specific amount of time. (Basically opened the dc supply line and inserted an ammeter to measure current draw and voltage.) Although the surveillance software is quite sophisticated the cameras will be managed from Toronto - I've allocated 3 hours per day for transmission. I've also allocated time for the computer to sleep and time to "prepare to sleep" etc. The two devices combined have 6 different power states that I've measured and input to my load analysis. Once I nail down the power piece I'll be able to put all of the components together and start my end to end test. Thanks for your thoughts and input.....mike

[inetdog]

Thanks for the recommendations, I'll incorporate them into my calcs. Thx also for the battery tip and I'll take the luck also as you just never know! cheers, spike

Since by its nature the system is mission critical to you may want to consider what to do when the battery voltage (or temperature) indicates that you are getting into trouble. Rather than run things into the ground and maybe never come back up again, consider a fallback (survival mode) in which you run the surveillance only 5 minutes out of 15 or some such ratio (choosing the time and the duty cycle based on how long it takes to come back online again after switching off. If you can find a dependable motion detection circuit for the camera, you can have it switch to continuous recording if it detects motion during one of its on periods.
Get creative with your load management too, and you may find that you can spend less on your battery bank and get better survivability.
(Of course if you think that the added complexity is more likely to cause problems than solve them, that is a decision you have to make.)

[spike001]

OK Spike now that I know what you are doing and is MISSION CRITICAL here as my recommendations.

Minimum Autonomy = 10 days

Solar Panel wattage = 2 X what is required for winter. That way if you do go deeply discharged say 3 days without sun around Xmas just one day of sun will bring it back up full. Overkill I know, but you cannot afford to have it go down. In summer you are on cloud 9 with abundant power to spare.

So make damn sure you get your daily watt hours down to an art.

Good Luck

SK

Thanks for the recommendations, I'll incorporate them into my calcs. Thx also for the battery tip and I'll take the luck also as you just never know! cheers, spike