Solar Math - calculating the needs of a DC Fridge (or What's Wrong with my CR65?)

[Gifu]

My 2001 Granby had a broken 3-way fridge. I replaced it with a Waeco Fridge powered by a 185-watt solar. And it doesn't work. Let's review:

Fridge: uses 576 watt hours per day
I installed a Waeco CR65 (after reading a number of threads).
The CR65 claims to use about 1.9amp-hours/hour on average (at 80F). Let's call it 2ah/h.
2ah*12volt=24 watt-hours per hour.
24 wh/h * 24 hours in a day = 576 watt-hours per day.

Solar: produces ~641 watt-hours per day (est)
I happened to have a Sharp 185-watt Solar Module (I used to work in the solar industry).
185 watts (rated) * 77% (derate for real world conditions etc) * 4.5 sun hours (ave. light intensity in CA) = 641 watt hours per day.
Now in the summer this will be much higher, as high as 7 sun-hours per day giving me... 185*77%*7= 997 watt-hours per day.
(note: sun-hours per day are a measurement of sunlight intensity and time... it's not how many hours the sun is up).

the solar runs through a Morningstar high-voltage charge controller (Sunsaver MPPT will handle the high voltage of this particular solar module at 40volts, and charge a 12-volt battery) and charges my battery bank:

Battery Bank 660 watt-hours
so for storage I have two Blue Top Optima Batteries. They are each rated at 55-amp hours. at 50% max degree of discharge... that's 55ah*12-volts*50%*2 batteries =660 watt-hours

to recap:
fridge uses 576 watt-hours per day
Solar produces 640-997 watt-hours per day
batteries can hold 660 watt-hours.

Why am I writing this all out? Because it doesn't work!!! With full sunlight the battery bank goes dead in about three days. Arrrggghhh!!!

Either the fridge is pulling way too much, or I have a bum battery.
I tested the batteries by running a large 12-volt air compressor of each one separately.
Voltage before test: 13.1 volts
Voltage during load (lasted 5 minutes, probably 500 watts) 12 volts
Voltage after load test: 12.7

Still with me? Yeah, this was a long nerdy post... any ideas would be helpful.

 

[MarkBC]

Do the "sun-hours" include -- among other factors -- the effect of angle of the sun? That is, the fact that the solar flux (and so, the power output) is reduced by a factor equal to the cosine of the angle that the sun is off from vertical?
"ave. light intensity in CA"...but Yreka is going to be different than San Diego, because of the difference in latitude (42 vs 33). Where is your system when this is running?
Of course, if you used to work in the solar industry then you must know all this...but other readers might not.

So...unfortunately, I don't really have the answer to your question, but one way to get closer to an answer -- though you may have to spend more money -- is to actually measure the solar-produced energy going into the battery AND measure the energy feeding the fridge. I use Watts-Up meters, which can measure cumulative watt-hr (since last reset) as well as other parameters. I like knowing how much energy my solar system is actually putting into the battery. I also have used one to measure power being fed to the fridge. They're really handy -- and fun for a enginerd like me.
You may decide that it's not worth buying one of these just to answer this question, but maybe you could find one -- or a similar one -- to borrow...?

What is the ambient temperature -- in the camper -- when you're seeing this failure to keep up? That is, is the fridge having to work harder than assumed?
When you say "the battery bank goes dead in 3 days", what voltage is it reading when "dead"?
Have you measured the battery voltage at the end of the solar-day (do the batteries ever seem "full"?) and again just before the sun hits the panels in the morning?

 

[Kispiox]

Math overview looks good. Refer peak draw and solar peak production are in sync , not off-set so no money going back in the bank. I'll show my age to quote an old low tech He-Haw saying,.."Tryin to haul two ton a fertilizer wit a one ton truck". Do you need more peak solar capacity? My three way refer if on DC is the only thing that my panel will not support. Not even close.

 

[pvstoy]

Math on paper is one thing, but sounds like you will have to baby sit this to find out the real numbers of what your system is producing / drawing out in real time.

Check the amp draw the Waeco is taking for startup and run time when it cycles. How many time does it cycle? Is it empty or full? Water jugs inside will help keep temp steady and reduce the run time.

Like what was said, battery condition before solar kicks in and during and at the end of the day.

Maybe you want to turn off the Waeco unit and just check the solar / battery system then add the Waeco or any load until you are convienced the system is behaving properley and as advertised. Then look at the Waeco unit for any faults.

 

[Hittheroad]

I do not see where the problem is. But I am pretty sure there is one. We have run a 2 amp fridge for many days with less solar power than you have. The watts up meter may be your best bet.

 

[pvstoy]

Might add that you seem like you know what you are doing around wiring, but double check your wire size and run lengths. Double check all fittings for tightness and make sure you have good clean grounds.

It is the simple things that cause you the most grief............

 

[Gifu]

the Sunsaver charge controller has built in meters to measure voltage, current, temp, etc.
It does not measure watt-hours created (from the solar array).
It _could_ measure watt-hours consumed (ie: Fridge) if I wired the fridge to the "load" on the controller instead of to the battery. I may try that. The controller can take 15A of load... so the Fridge is doable there.

When battery is "dead", i mean that the voltage drops as low as 11.3 or so. The controller gives all sorts of warnings and pretty lights and such.

Wire size is all over sized.
Ave. sun hours was...well... an averager because I don't stay in one place.
It's about 5.5 sun hours in the Bay Area where I live. But that means as low as 4 in the winter and 7 in the summer. I experienced problems when I was in Oregon last summer (read: sunny, a bit warm, but not horrible) and at Burning Man (read: VERY sunny, and mild weather this year... 80s to low 90s).

My theory is that either the Fridge is more than 2A
or
There are small gnomes sneaking in at night and stealing my watt-hours.

I'll try to be more scientific and record the voltages over time as the Fridge runs down.
And I will probably move the Fridge load direct to the charge controller. But, what will happen is the charge controller will turn the Fridge off at about 11.8V.

QUESTION: I need a way to have "shore power", in other words a battery charger for when I'm parked a long time. is there a recomended unit? I'll be parked in Baja for a few weeks.

 

[photohc]

You might want to try looking into a Tri-Metric. A little pricy but basically can monitor what's going into the battery and what is being taken out. Using the voltage of the battery to try and determine where you are in terms of remain battery power is only at best a guess.
With the Tri-Metric, you will install a low resistance shunt to measure what is the net current going in or out of the battery bank. You will tell it the capacity of your batteries and it will keep track of what is being consumed by the loads (refer/lights/pump/etc) and what has been put back in from the solar panels. By night fall, you will have a % remaining in the battery bank and by morning, you can see how much more the battery bank has been drawn down. As the sun comes up, you can monitor the net current being pumped back in and see the % of the battery bank go back up.
By turning off the refer, and noting the current draw, then turning back on, you will see the actual draw of the refer by noting the difference in current.
Hope this helps. I have the same MPPT controller but with two panels (100w + 150w) in series. I found the 100 watt not able to keep up with the Waeco CR80L . Now I have a net current input to the battery bank most of the day. The draw of the Waeco CR80L is about 4amps.

 

[craig333]

Keep in mind the refer doesn't run continually. How old is your battery? Mine was losing voltage faster than it should. Yet it would charge up to 13.7 just fine. My smart charger would take it up to the trickle stage just fine. Finally I took it out and load tested it. It failed. Been fine with the new battery.

 

[camelracer]

I am also using the Sunsaver controller and monitor panel. I have the fridge connected to the load terminals so I can monitor its electrical usage. The Sunsaver can be set to turn off at either 11.0 or 11.5 battery volts. The voltage shutoff on your fridge should also be adjustable.

I was also having problems with my 100W solar keeping up with the fridge. I improved the air flow around the condenser coil and added some thermal mass (water bottles). This seems to have helped but I haven't had any long duration camps yet. I'll be giving it the long duration Baja test in March.

 

[williak]

I can confirm the 4 amp draw of the Waeco 65, and 2 amps/hour usage. During a recent trip I kept an eye on a trimetric 2025 battery meter, and an ear on the fridge compressor. Over 2 hours, the compressor cycled 3 times for a total of 63 minutes. Outside temp was in the low 60s. Inside temp slightly more.

Can't really comment on the draw in higher temps, but have to assume the compressor will need to work harder. There is a post somewhere about adding insulation to the fridge to address that situation. There is a lot of airflow through the factory installed fridge vent through to the camper interior, I guess to help move heat off the compressor. That airflow becomes a problem in cold weather, but that's another topic.

 

[Gifu]

Guys this has been super helpful. Thank you.
One note, a huge off grid solar nerd told me that in off grid situations the 77% derate factor should be 60%. Yikes!
I'm off to Baja tomorrow as is. Guess if all else fails, I'll drink warm Tecate.
Had I know this thing was such a hog I would have spent the energy to rebuild my cabinets and install an ARB chest fridge. The one I use in my land cruiser only needs about 40watts to run indefinitely.

 

[NorCalSteve]

I purchased a multimeter which has a clamp on DC current meter. I keep this in the camper. You can clamp it on a wire and measure the DC current. There are a number available and I am sure others here own one and can make recomendations. I bought THIS one from Amazon but I remember it being around $100 at the time. I bought it because it looked like a B&K meter but was cheaper. That does not seem to be the case now. Anyways, it measures AC/DC current, voltage and resistance. There are probably better choices now and I often don't do as much research as I should.

Wish I had some great idea but you have done everything I have thought of.

Steve