12v solar and chest freezer to fridge conversion - what went wrong?

[max2k]

Seriously? It's a refrigerator (converted). I'm not making ice. See my original post, or subsequent posts, or even the post that you replied to that said "I'm not making ice."

Calm down- it depends where you set your thermostat and we have no idea where. Cooling 1 gal of water just by 1 deg F takes about 2 Wh. Cooling it by 40 deg F will take 80Wh. If you set your thermostat too low that gal will freeze in there taking with it 80Wh to cool + 160Wh to freeze so as you can see these things add up. Maintaining low temperatures after things settled takes way less energy so if you measured your numbers on empty freezer and then loaded it with stuff at room temperature there's no wonder it depleted your battery.

[paulgareau]

You battery is junk. A sulphated boat anchor. You probably cannot be convinced of that until you use the Kill-A-Watt meter some more. You should be watching for battery sales while you re-measure your load.

Really, brand new from the store? And the other batteries that had been working successfully? I'm done with the project now, it was for 6 weeks as I've said a few times.

[paulgareau]

JPM brings a good point- your freezer energy demands will greatly depend where you set your thermostat - the colder you set the more energy it will require to maintain that temp. If you set it below freezing point on top of just cooling things inside you'll be making ice as well and that sharply takes way more energy. Running wattage makes sense only for a given temperature and freezer load. Say you put it below freezing and then placed some watery drink inside- freezer will be forced to freeze it and that process will take a lot of extra energy.

Seriously? It's a refrigerator (converted). I'm not making ice. See my original post, or subsequent posts, or even the post that you replied to that said "I'm not making ice."

[AzRoute66]

Also surprising was that the batteries could be floating by noon, which should be impossible too.

You battery is junk. A sulphated boat anchor. You probably cannot be convinced of that until you use the Kill-A-Watt meter some more. You should be watching for battery sales while you re-measure your load.

[max2k]

OK, I'm fairly new here so maybe I didn't ask my question the right way. BUT. This is about a chest freezer to refrigerator conversion using a johnson controls thermostat (see orig question and follow ups). I'm not making ice. This is probably good info for someone but has nothing to do with my situation or question.

Also have provided the startup and running wattage....

JPM brings a good point- your freezer energy demands will greatly depend where you set your thermostat - the colder you set the more energy it will require to maintain that temp. If you set it below freezing point on top of just cooling things inside you'll be making ice as well and that sharply takes way more energy. Running wattage makes sense only for a given temperature and freezer load. Say you put it below freezing and then placed some watery drink inside- freezer will be forced to freeze it and that process will take a lot of extra energy.

[paulgareau]

On the freezer operation alone, that is, getting the load figured out before talking about batteries: On the energy draw, that 80W is "steady state", that is, after compressor startup transients have calmed down. Startup will be more. How much will vary some.

Once running and the cooling demand is met, the unit will draw a few Watts or so when the compressor is not running. The unit will cycle on/off to maintain the thermostat set point. If the unit is self defrosting, depending on method, it'll probably also draw some power during defrost cycle.

As for the daily power consumption required to keep things cool, or make ice (the two are slightly different duties), that will depend on the load the unit needs to meet. If you fill the freezer with H2O and the thermostat set point is 0 < C., the unit will run on a fairly regular cycle, and probably a large part of the time, maybe constantly, until the H2O is frozen. After that, the freezer will cool the ice further, down to the thermostat set point.

After that, with no other additions/withdrawls, the unit will simply run to maintain the freezer temp., removing any heat gain.

If ice is withdrawn and replaced with water, as sounds like what the application is, the freezer thermostat will sense the warmer temps. and the compressor will run and make ice.

If you use more ice than, say, 300 W-h per day of input will produce and still meet the heat gain demand of the unit, the compressor will run longer to meet the increased demand for more ice and energy use will be > 300 W-h/day.

FWIW, it takes about 80 as much energy to create 1 kg. of ice as it does to reduce the temp. of 1 kg. of water by 1 deg. C.

OK, I'm fairly new here so maybe I didn't ask my question the right way. BUT. This is about a chest freezer to refrigerator conversion using a johnson controls thermostat (see orig question and follow ups). I'm not making ice. This is probably good info for someone but has nothing to do with my situation or question.

Also have provided the startup and running wattage....

[max2k]

I had the first battery tested and it did better than it was rated for - at least for mca. I later pulled it out of the bank and got pretty much the same results.

Would a cranking test diagnose a battery that charges and discharges more quickly than it should?

since you're seeing some unexpected results I'd start from checking your assumptions and the math, not out of curiosity but out of necessity. You're using your battery for deep cycles, it's cranking performance is very remotely relevant. Estimating its deep cycle performance can be done using 12V 120W headlight bulb and timer- bring it up to float and then measure time it will take the bulb to discharge it to 50%. The actual capacity would be 120W / 12V x Nhrs x 2. You might discover it is much less than you assumed it is. 100Ah should provide energy for the bulb roughly for 5hrs (50% SOC).

You stated your freezer consumes up to 500Wh over night and 80W running. The 'night' here is time without sunlight strong enough to charge your battery and pick up freezer load at the same time. It is commonly assumed to be 5 hrs so your 'night' would be 24-5 = 19h the battery would need to provide energy to run your freezer. Have you measured your freezer energy consumption over that 19 h period? Assuming 50% duty cycle 80 x 19 x 0.5 = 760Wh, the required battery capacity should be at least 5 times that if you want your freezer working during cloud spells: 760Wh / 12V x 5 = 300Ah; Just for 1 'night' it should be 760 / 12 x 2 = 120 Ah as you don't want your battery to go below 50% and this is not accounting for at least 20% losses. You are short on battery capacity for even 1 'night'.

Your 250W panel would produce 250 x 5 = 1250Wh energy but in reality could be less depending on orientation. This is also only possible with MPPT CC in sunny CA or similar and in the summer. You freezer consumes 80 x 24 x 0.5 = 960 Wh during 24 hr period again assuming 50% duty cycle so from just PV energy point of view it seems 1 250W panel should be enough for summer. During winter PV production would drop at least 40% but hopefully your freezer won't run that much.

[paulgareau]

Cranking test? are you using automotive CCA or dual purpose batteries? Yes mixing old and new batteries is a recipe for destroying your new batteries. Stop buying junk until you have done your math.

Sorry man, but I told you what batteries I was using. I also said I got them because I needed a quick and dirty system. You're still not answering the question. And are you saying it's OK to buy junk after I've done math? What math should I do? Really confused by your posts...

[J.P.M.]

I was off grid for about 6 weeks over the summer and got tired of buying ice, so I picked up a 7 cubic foot chest freezer and ordered a johnson controls thermostat. I heard similar setups draw about 300 wh per day, and I measured mine between that and 500 wh per day. The startup draw of the fridge was 800w and it ran at 80w. I started with one 100 ah battery, which i figured would be OK as long as there was plenty of sun and I could use a generator on the cloudy days. Also used a 250w panel. The battery was ending up very dead by morning, so I added a second battery and then a third (taken from another small solar system that I was using for something else). I was still seeing very surprising results like being below 50% by morning, meaning I lost over 150 ah over night. Also surprising was that the batteries could be floating by noon, which should be impossible too. Some troubleshooting showed my sunforce 1000w inverter was drawing 1.5 amps at idle, so that had to go. I replaced it with a xantrex 1000w, which could only start my fridge when the batteries were fully charged. I also upgraded to a MPPT charge controller because by that point I was committed and why not throw a few hundred more bucks at it. I think the weak point in the system was the batteries - this was meant to be a quick and dirty system so I used Walmart 24DC marine, deep cycle batteries. Since everything else /should have/ worked, if the batteries were the problem can someone explain why and why there seemed to be a disparity between the amp hours used/remaining and the voltage of the battery bank? Thanks!

On the freezer operation alone, that is, getting the load figured out before talking about batteries: On the energy draw, that 80W is "steady state", that is, after compressor startup transients have calmed down. Startup will be more. How much will vary some.

Once running and the cooling demand is met, the unit will draw a few Watts or so when the compressor is not running. The unit will cycle on/off to maintain the thermostat set point. If the unit is self defrosting, depending on method, it'll probably also draw some power during defrost cycle.

As for the daily power consumption required to keep things cool, or make ice (the two are slightly different duties), that will depend on the load the unit needs to meet. If you fill the freezer with H2O and the thermostat set point is 0 < C., the unit will run on a fairly regular cycle, and probably a large part of the time, maybe constantly, until the H2O is frozen. After that, the freezer will cool the ice further, down to the thermostat set point.

After that, with no other additions/withdrawls, the unit will simply run to maintain the freezer temp., removing any heat gain.

If ice is withdrawn and replaced with water, as sounds like what the application is, the freezer thermostat will sense the warmer temps. and the compressor will run and make ice.

If you use more ice than, say, 300 W-h per day of input will produce and still meet the heat gain demand of the unit, the compressor will run longer to meet the increased demand for more ice and energy use will be > 300 W-h/day.

FWIW, it takes about 80 as much energy to create 1 kg. of ice as it does to reduce the temp. of 1 kg. of water by 1 deg. C.

[paulgareau]

You are wasting time, food and money. If you are serious, you should start with a native 24v chest style refrigeration unit. like Sun Frost, dometic and others. They run directly from 2X 12v batteries of the appropriate Amp Hours. no need for an inverter drawing unnecessary watt's during the off cycle. If you can still return the 110v chest freezer, do it now. spend $1100 and you will save thousands compared trying to make an inverter refrigeration system.

Thanks, I think? But you didn't answer my question (see subject of post) and you missed that this was only for 6 weeks over this past summer (see first sentence of post).

[Logan5]

Cranking test? are you using automotive CCA or dual purpose batteries? Yes mixing old and new batteries is a recipe for destroying your new batteries. Stop buying junk until you have done your math.

[Logan5]

You are wasting time, food and money. If you are serious, you should start with a native 24v chest style refrigeration unit. like Sun Frost, dometic and others. They run directly from 2X 12v batteries of the appropriate Amp Hours. no need for an inverter drawing unnecessary watt's during the off cycle. If you can still return the 110v chest freezer, do it now. spend $1100 and you will save thousands compared trying to make an inverter refrigeration system.

[paulgareau]

Once you suspected the first battery was bad, you should have removed it from the system. Putting unbalanced AGM batteries in parallel won't give anything close to the Ah capacity you are thinking.

I had the first battery tested and it did better than it was rated for - at least for mca. I later pulled it out of the bank and got pretty much the same results.

Would a cranking test diagnose a battery that charges and discharges more quickly than it should?

[sensij]

Once you suspected the first battery was bad, you should have removed it from the system. Putting unbalanced AGM batteries in parallel won't give anything close to the Ah capacity you are thinking.

[paulgareau]

"it monitors the temp inside the freezer and cuts power when it's at a certain temperature". Lets call the "No Power" state resting, and the "Running Power" active. If your box is full of ice bags in tubs, the box will be in "No Power" until all the Ice Melts and the temp rises to your Johnson setting. If your box is full of hot rocks in a tubs, the "Running Power" will stay on till the rocks have cooled to the Johnson setting. How much Power does your freezer draw when it is in the "Running Power" state (I mean measured). Your usage should equal the measured requirement (volts and amps) x the time you are in the Running State. This should be straight forward.

The variable should be caused by what is in the box and what status it is in. Also what Johnson Setting is chosen (or what is built in as a temp setting). If you know the voltage the device uses when in the Runnning State, then I would sit down and time the running state length and frequency of the runs for a couple of hours to see what the load really is.

Just my 2 cents.

thanks rch

OK, if I'm following, that's the info in my original post. 300-500wh per day according to Killowatt. Around 80w running and over 800w during startup for a few seconds, also from killowatt.