diy solar

diy solar

Adding storage to my Enphase system

If you just need to power a laptop, what about using a 12vdc to 110vac off the buck converter you already have?
I once bought a device for a toshiba laptop that plugged into a 12 v socket. I have since used small 100-200 Watt inverters since laptops don't use much power and aren't fussy about quality.
 
Yes, that was the CPU cooler out of an old Pentium 4 Shuttle PC. It was a tiny form factor about the size of a toaster, and made almost as much heat. That was the first time I had used heat pipes. I got it in about 2004 or so. It was the only reason that box worked at all. It carried the heat from the CPU up to the rear panel above the i/o shield. It had a variable speed fan that exhausted the case through the radiator, and if felt like a hair dryer blowing out the back.

That 1,200 watt Victron inverter looks pretty good, but I just can't justify over $400 for an inverter I will rarely use. They also have an 800 watt version which in reality will probably put out more power than the KRXNY 1,000 watt one, and I found that listed for $275 That might be doable. I have powered up my cheapo 700 watt modified sine off of the 12 volt buck converter, so I know that works. The rack cabinet for the battery bank will be here Wednesday. When I get everything stuffed in there, I will see what room I have left. I got a couple 15 amp switch/breakers, for the power to the buck converters, and a handful of XT-60 connectors for the outputs. And I just got a 25 foot pack of this #10 zip cord to wire that all up.


The 12 volt converter can put out 25 amps, so I wanted to be covered. They also have that wire in #8 awg. It says it is pure copper, then says aluminum though?? Hmmm
 
I just plugged my fridge into the Kill-O-Watt and I will see how bad it is after a day. Compressor is running now and it is puling 240 watts, 270 VA which is not too bad, assuming it ever cycles.
 
It has been exactly 24 hours. The refrigerator used 4.38 KWH in 24 hours. So that is averaging 182.5 watts. While the compressor is running right now it is pulling 200 watts. So.... it is running 91.25% of the time. Not good.4.38 x 365 = 1,598.7 KWH's per year. OUCH! I knew it was a pig. I will let it go 2 more days to see if the average holds. A modern fridge is about 600 KWH's per year. So I should have no problem cutting the draw in half, but even then, it will take 5+ years to pay off a new fridge.
 
Looks to me like it could be five years at $0.20/kWh utility grid rates to break even on a new $1000 fridge (and 20% will have broken by then)
But I think it's 20 years at $0.05/kWh PV + inverter costs.

How does your accounting work out if you add more PV instead?
Also check out used (but new to you) refrigerators. I picked up a nice one for $50 and it's working great. Haven't measured its power draw yet.
Save the old one for n+1 redundancy, either to hold stuff while defrosting another, or in case the new one breaks.
 
My cheapest rate is $0.25 and my peak rate jumps to $0.42 but still, 5 years is a fair estimate. My current fridge came with this house. I have been using it for 18 years, and no idea how old it was then, could be another 5 on there. So I can't really complain too much. I actually still have the original white Energy Guide sticker from this old fridge, so I think it was fairly new when I bought the house. That sticker estimated that it would take 1022 KWH's per year. By this one day only estimate, it is using 56% more energy than that. All the new refrigerators have a Yellow Energy Star sticker, and the testing is different, so they say to not directly compare them. A new similar fridge claims 709 KWH. I found a real nice Frigidaire that is a bit more money at $1,498 at Lowes that claims just 646 KWH per year. Now that is getting down where it can make a difference. To get under 600 KWH/year you can't do a side by side with through the door ice.
 
Check the reviews on refrigerators and cross your fingers


At first I assumed the 5% to 20% bad reviews were due to squeaky wheel, But one repair guy posted failure rates for almost all brands in that range.
It may be that after infant mortality, the remaining ones are good. (or not, depending on what obsolescence was designed in.) I picked up a Kitchen Aid side-by-side, relatively new model, to match the decor (used, $50) and converted the old Frost King into a side-by-side freezer.
 
... 4.38 KWH in 24 hours ... it is running 91.25% of the time.
If it's running that much and that hard it's probably about ready to die, in which case don't think of it as a payback period to replace it.

... To get under 600 KWH/year you can't do a side by side with through the door ice...
They're out there, close anyway. First on the energy list I saw was a Kenmore available from Sears. 610 kWh/y... although the ice in the door really jacks the price up. Here's the reference link again, easier than walking around Lowes: energy star gov site data ;)
Kenmore 51764
25 cu. ft.
prod_1618213012
 
Last edited:
That is about what I have been able to find. A side by side with a through door ice maker currently is going to take more than 600 KWH/year to run. I found some much more efficient refrigerators, down to under 300 KWH/year, but they are freezer on top, and no through door ice. And about 4 cubic feet smaller inside as well. Funny thing is, they are even cheaper to buy.
 
It has now been a few days, so I figured I would take a look back at the Kill-O-Watt on my refrigerator.

At this moment, the power is running at 215 watts, 230 VA for a 0.93 PF and it says my power is at 59.9 Hz, the Schneider XW says 60.0 though, pretty close for sure. It has been connected for 89.3 hours or 3.72 days, and has consumed 15.82 KWH's. That works out to 4.25 KWH per day. A little better than that first day, but still in the area. By these numbers, the compressor was still running about 82% of the time. So much for coasting, ever.

So, I am not in a huge hurry, but I will be shopping for a more efficient refrigerator in the near future. I am going to move the Kill-O-Watt around to a few different devices and see what else is being an energy hog around here.
 
I know PF = KW / KVa ... but not really sure why PF is important or what it might be saying about your fridge. Anyone know?

PF can be a current draw that isn't a sine wave (clipping peaks with a rectifier for typical power supply or VFD input).

PF can also be an out of phase sine wave, like inductive or capacitive load. i.e., opposite polarity current from voltage part of the time.
That is shoving power back into the source. Not such a big deal on the grid, just acts like a generator and delivers power some other load will consume.
But for an inverter, that current has to go somewhere. A 4-quadrant circuit might be able to capture that power and store it in capacitors. A 2-quadrant circuit might burn it off as watts in its transistors.
 
Since it is an inductive motor load, I was actually surprised the PF was that high. 0.93 is not that far off from a 1.0 resistive load, so while running, it is not too bad of a load for the inverter. I didn't check the start surge.

I don't think a power factor requirement has been set for consumer electronic gear yet in the US, but Energy Star is trying to require new equipment to be higher than .90 and my old fridge does make that. Many cheap power supplies with a switching regulator are much worse. Some have power factors as bad as 0.45 Early PC power supplies were like that. Like the clipped peak Hedges just mentioned.
 
I got 2 nights of work done on my setup. The new 19 inch rack cabinet came in. I see why it was only rated to 130 pounds of gear. The main rails are pretty thin steel. With two pairs of adapter plates, I was able to mount in my old thick steel rails, and I am using the ones that cam with the rack to help hold up the back of the batteries and also pushed the BMS and safety switch back about 4 inches. This will put my BMS heat sink very close to one of the rack fans in the top panel. I also built up another 3 RU panel with four 15 amp carling DC breakers for auxiliary loads. One feeds a 12 volt buck converter that will power the Gateway, 2 fans, and the Network switch. The next breaker feeds a 24 volt buck converter that will power the PLC and maybe a powerline monitor. The third breaker will just feed a direct battery output that I can use for testing or running the adjustable buck converter for any odd voltage or current I may need. And I can also plug in my 600 watt charger to charge up the bank if the XW is unable to power up for any reason. The forth breaker will most likely be a stand alone 48 volt input inverter. I think I will end up getting the 800 watt Victron. That way I will have a good 120 volt sine wave AC source even if I have to shut down the XW during a blackout. Everything is after the BMS, so it is all low volt protected. The 4 breakers are being fed from a 30 amp fuse on the buss bar before the large XW inverter disconnect switch, so these will all work when the inverter is shut down.

I also made a nice mounting rail for the Class T fuse, and the 300 amp Anderson connector. Those 2 are not wired in just yet. I have to shut down the XW to do that, which will cut power to the essential loads panel. So I will also wire in the bypass switch for the "EL" panel at the same time. But I ran into a small problem there. The connectors on the switch have plates intended for clamping directly on bare wire. My crimp lugs are too think and it can't hold them. I may replace the screws to make it work, or just let it clamp the wires without lugs.

IMG_8544.JPG IMG_8541.JPG IMG_8545.JPG

Installing the bottom panel will be fun. We are going to have to lift it up and bolt it on. I have a plan, but it will take 3 people.
 
I have a plan, but it will take 3 people
I am driving to LA on Sunday Oct 11th and often stop at the Santa Clarita Super charger if you need a hand from and old guy? I am not much good for anything over 75 lbs. but your project has piqued my interest if that is a convenient day. Most of that following week I will be in Hermosa Beach working on selling an investment there so not available any other time. Send me a message if that would work out?
 
I don't have my work schedule that far out yet, but due to Covid, there is a fair chance I will be around on the 10th. It would be fun to meet up. I'll reach out in a private message later this week.
 
A few small updates.

I had not heard anything back from Schneider in about 2 weeks. Figured I would write back later this week, but yesterday, I get an email they are closing my case as I have not written them. What? I ripped off a pretty angry response that included a WTF. Well, they replied to that pretty quick. They didn't appreciate my tone. Yes, I admit, my e-mail was pretty nasty. He sent me a screen shot of the mail he said he sent me the day after my last e-mail. I searched all of the folders on my gmail account and I can't find it anywhere. Not deleted, spam, archive, nothing. So I don't know what happened.

So I wrote a nice email back, and addressed his 3 points. Two were wrong assumptions about my setup, and the 3rd was asking about my odd setup voltages, that I did explain in my last e-mail about how I am forcing it to charge and then grid sell. SO no real new info yet, but at least they are now talking to me.

So last night, I put in "normal" settings.
Recharge Volts = 52 volts
Grid Support Volts = 56 volts

This morning, the battery was down to 51.5 volts, so it started charging as expected, and it did complete, going all the way up to 57.6 as set for the absorb voltage. I had planned on just leaving it alone to do it's thing, but it would have been close to get the battery topped up, so I did up the charge current a few steps to make sure it would get to full voltage. And at just 2:13 pm the battery did reach 57.6 and it went into absorb mode for about 20 minutes. Then at 4 pm with no human intervention, it did start inverting and doing grid sell as it should. At this point it all looks good. The question is how far down will it pull the battery, AND will it be able to charge again in the morning? I am planning to take screen shots of this to send back to Schneider. Hopefully they will add a software fix.
 
Spent a few hours working on my system last night.
When solar production was dropping off, I tried another power fail test. This time I had a roughly 600 watt load on the output of the XW-Pro along with the Enphase solar, which was still producing, but only about 230 watts total from all 16 panels. This was about 6 pm with the sun low in the sky. I did certainly notice a glitch in the power. It sure seemed more than 8 ms to get stable power back on the loads. The Schneider Dashboard showed the battery was now supplying 400 watts and the solar about 200 watts, but it is rounded to 0.1 KW so those are easily 50 watts off. If I did lose solar, I wanted to do it where I would not lose a ton of production. From what I could tell this time, 2 of the iQ7's dropped out due to "AC Frequency out of range, Grid Instability", while the other 14 kept making power. I turned off 500 watts of load, and the XW-Pro seamlessly transitioned from supplying 400 watts to then absorbing the extra 100 watts and feeding it to the battery bank. Once "off the grid" it all seems to work just fine. I know this was a very low power test. I would like to try it again with more output load as well as more solar coming in. But the switchover glitch still caused 2 micros to shut down. But then I turned the AC grid feed back on. I expected that to be completely smooth, and this time I did not notice a glitch in the loads, but at least one more micro inverter shut off with the same grid instability error. Too bad the sun was now going too far down, and one by one inverters were shutting off to "DC Power Too Low" which is completely expected as there is no more sun hitting the panels. This morning, all 16 were working and I hit a very typical peak power of 3,200 watts.

I have also finished assembling the rack for the battery bank. The positive battery cable is now going through the Anderson connector and the Class T fuse. I will do the negative cable tonight after it is done exporting power at 9 pm. Here are a couple pics of the progress.

IMG_8563.JPG IMG_8568.JPG IMG_8567.JPG IMG_8566.JPG

The blue heat sink is on the front side of the BMS. The heat pipe cooler will go back on the back side in the rack under the left top fan. The side panels and front door are mesh for air flow. The rear panel facing the wall is solid steel sheet, but not installed yet. I still have to take the top off a few more times and the rear panel bolts to it. I may add metal panels in the sides to block a direct path of fire. The front still has metal panels between all the cells and the front, so I am not worried about that direction.
 
I have now sent 3 more e-mails to Schneider Tech support with no answer, but I will give them a few more days. I sent them screen shots of what is going on. I have it all down to the grid support, selling power to my breaker panel, is shutting down when the battery bank discharges to 0.5 volts above the "Recharge Volts" setting. Since the battery is still ABOVE the recharge volts... IT WON"T CHARGE. DUH

This is an old relic in the code for a system with a DC solar charge controller. It shuts off the inverter to give the solar a chance to charge it back up. If something bad happens and the batteries do fall another 0.5 volts, then it will use grid power to charge up so the batteries don't die. That makes sense, but not in an AC coupled system. When the time tells it to charge, it just needs to go into charge mode, but it won't.

The other little thing I found out. The "Grid Support volts" also effects the charging. When I set Grid Suport volts well above the actual battery volts, it will go into Absorb mode and nicely top off the batteries. But if the battery goes above the grid support voltage setting, it will keep Bulk charging, but once it hits the absorb voltage, it stops it from going into Absorb mode.
 
Back
Top