My Portable Solar System: Poorly designed, over-built and under-powered

[NinjaDeathMonk]

Has about 825 Wh of usable power and about 650W of continuous AC output. The door on the left has wires for the solar panels, and the solar input is displayed on the left screen. The right screen displays battery capacity. The right door has wires for an external AC charger.

Main components:

• 36V 158.4Wh 10s2p Hoverboard Batteries x 6 - $180
• Reliable Power 800W Inverter - $125
• MPPT Boost Solar Charge Controller (CTKEV300) - $40
• AiLi Battery Monitor 100A - $38.50
• DROK Digital Multimeter with 100A Current Shunt - $17
• Bay Marine BusBar - $30
• Hyper Tough 19-Inch Toolbox - $12
• 42V 10A Lithium Battery Charger For 10S 36V - $58

Easily went over $600 with tools, wiring, connectors, and assorted parts. Way more if you include the prior failed attempts.

The inspiration and basis was Jehu's solar generator made from a recycled UPS and hoverboard batteries. I was going to build the UPS into a case with better airflow but I had two of them fail on me. So I decided to reinvent it with what I had and use a better quality inverter. Since I was working with 36V batteries, options were limited. The Reliable Power inverter I went with was also a dissapointment, because it doesn't perform as advertised due to a fuse limitation. Which means it can't start my fridge in case of a outage.

So I ended up with a poorly designed, over-built and under-powered solar system. At least I learned a lot in the process, including:

1. I have developed a deep and powerful disdain for cutting plastic.
2. I was pleasantly surprised that the solar charger performed as advertised and the batteries have been solid.
3. Will was absolutely right in his video about DIY modular systems not being cost-effective these days. A lot changed in 10 months. I'll probably buy the Bluetti to power the fridge in an outage and keep this abomination around for lighting and stuff.

Thanks a lot to Will and the forum members that have been helpful and patient with me, even at my most embarrassing.

 

[erik.calco]

I love it. So much to say and ask. But, I have to start out by saying I love the ebay description for your battery pack:

UFO U18650 36V 158.4Wh Battery 10s2p for Hoverboard Balance Board Vape

I really really want to see someone Vape with that battery pack.

Batteries. That ebay price puts the 18650s at under $2 each. Did you open any of the packs up and test them for capacity and internal resistance? I got some good new ones on ebay for about $2 each. Less than advertised, but still averaged about 2100mA on first run, and only a few high rollers on iR. I'm guessing yours came with a BMS, but I don't see a description in your ebay listing for protections or balancing. The bottom line is building your own with $2 cells, BMS, and the parts to connect it all might add up to 30% to your battery cost. If you can get a good deal a 12V drop-in, you can carry around over 1KWH.

Case. I default to the battery boxes, but I like your choice, albeit less ventilated and lacking a shoulder strap. Yours looks a lot better than a battery box. How waterproof is it? The battery boxes have downward facing vents, but if I recall in a test I did a few years back, I think water got in when I left it out in the rain, but can't remember for sure. Need to retest that.

While I haven't built a full blown completed one like yours, I have been playing with this for at least 5-6 years long before seeing Jehu or Will, starting with just putting two of those 6Vs you can buy anywhere in series and seeing if I could charge my phone and plug 12V lights into them, using a camera case to house it... then using my big 12V boat battery (thus the battery box), then AGMs, then lithium.

Thanks for sharing your learnings and the linked parts list.

Despite Will discovering that it is not necessarily as cost effective to do DIY, there are advantages to it:

- you learn a lot, as you said you did
- you have far more options to swap things out. pick up some cheap batteries somewhere and you can put them in your own unit.
- you can build it how you want it, to your specifications.
- complete quality control. if you want to be sure you can handle higher loads, you can be sure the wiring and connectors are all able to handle it.
- ability to re-use parts from prior builds and re-purpose parts such as a faulty UPS (usually only one of its functions fails.)
- create any form factor.
- add things that aren't in current units, such as an Android remote control and monitoring system and intelligence. (e.g., Google has a sale on the Moto G7 for $99, no contracts, unlocked, and can pay $4.13/mo at 0%. Wouldn't that make a nice 6" touch screen interface to your portable with the ability to text you when the batteries are low, and play the funeral song right before they die?).

 

[NinjaDeathMonk]

I love it.

Thanks, man.

I really really want to see someone Vape with that battery pack.

Apparently, some vape pens use 18650s. But maybe we can make a 36V hookah that could use these battery packs.

Did you open any of the packs up and test them for capacity and internal resistance?

One of the packs I got had a voltage of about 12V. I'm guessing the BMS was bad. The seller immediately sent a replacement and didn't ask for the original back. So I opened up the bad pack and it was made of Samsung ICR18650-22P. I trickled charged the cells back to a healthy voltage and used a LiitoKala to charge the rest of the way and test the capacity. Every single cell had a capacity just above 2200mAh and the resistance was around 50 or less.

Case. I default to the battery boxes, but I like your choice, albeit less ventilated and lacking a shoulder strap. Yours looks a lot better than a battery box. How waterproof is it?

Until just now, I didn't know battery boxes were a thing. That would've been a decent option. My toolbox lid is not rubber sealed, so it was probably not very waterproof to being with. After all the drilling for the fan vent, inverter panel and screws, it's less resistant now. I used this one because it was cheap, seems rugged, and meant for an earlier built. But it just fit everything perfectly. Alternatively, I could've aligned the batteries in the bottom and put the rest on top. It would've balanced the weight better but ventilation and fit would be tricky.

e.g., Google has a sale on the Moto G7 for $99, no contracts, unlocked, and can pay $4.13/mo at 0%. Wouldn't that make a nice 6" touch screen interface to your portable with the ability to text you when the batteries are low, and play the funeral song right before they die?).

HAHA. I think this project is already "over-built" as it is. But it's something to think about. Thanks for all your comments.

 

[erik.calco]

used a LiitoKala to charge the rest of the way and test the capacity.

That's what I have after watching Jehu's review. I love it! I use it every day.

Every single cell had a capacity just above 2200mAh and the resistance was around 50 or less.

That is good for new and really good for used. Regardless of what the printed rating is, I view over 2000mAh as good, and 40-50 is the threshold for separating pack candidates from vaping cells.

Until just now, I didn't know battery boxes were a thing.

Battery boxes are in the auto department where all the 12Vs are. The ventilation is probably overkill as they were designed for LA. But, with the inverter in there, you can't have too much cooling in the sun on a hot day.

HAHA. I think this project is already "over-built" as it is. But it's something to think about.

You're ready to build your new 2KWH portable with voice commands!

 

[atatistcheff]

The Reliable Power inverter I went with was also a dissapointment, because it doesn't perform as advertised due to a fuse limitation. Which means it can't start my fridge in case of a outage.

I have the same problem with a 1500W pure sine wave inverter. Works for running the freezer in the garage but our refrigerator pulls too much current when the compressor starts so the UPS shuts off. It's about time for a new fridge. I'd like to know how to determine the compressor design on the various models to try and predict whether they might work with an inverter.

 

[erik.calco]

NEWS: DIY Clan Leads to Tennessee Walmart Blackout

Walmart's power was out for 20 minutes, creating panic in the store. One witness said right before the power went out they saw a couple people with "DIY Solar" shirts kneeling behind the store model of a fridge. "Sparks flew out from behind the fridge, then it just went dark."

They seemed prepared as they ran out with flashlights. One was heard shouting, "I don't think that fridge is going to work!"

 

[NinjaDeathMonk]

I'd like to know how to determine the compressor design on the various models to try and predict whether they might work with an inverter.

Usually, inside the fridge there is a label with "max current" or "full load current" or "full load amps". Use that to calculate the maximum watts.

 

[atatistcheff]

Usually, inside the fridge there is a label with "max current" or "full load current" or "full load amps". Use that to calculate the maximum watts.

Thanks, but this is a spike of current which is not represented by that rating. My current fridge is rated well below 1500W yet the compressor start spike still trips it.

I'll be heading to Walmart to do some testing, better bring my flashlight.

 

[Supervstech]

I just tonight connected my 1500watt 48v inverter to my toolbox battery... I might go test it with my 37cf fridge... not sure how wise that is though... aims psw inverter... should do it.

 

[smokerx]

To possibly get the price down, or just an option, is to consider buck convert the 36volts down to 12 volts. Since 12 volt inverters are more ubiquitous than 36volts. Pros are you can now tap in for a 12 volt supply line. Cons are the buck converter can bottle neck based on the amperage rating. Bottle necking you out of any peak wattage rating on the inverter, depending on the price = amperage rating of the buck converter.

Similarly, another option for the MPPT boost solar charge controller is replace it with a 12 volt panel to 12 volt controller to 12 volt battery to small 12 volt inverter to an AC powered 36volt battery charger mole. That might be advantageous, since hypothetically, lets say you want to just wire that to your car and use your car as a docking station for charging your bank. I assume theres some way to wire that in with your cars alternator to charge from both solar and gas powered generator. You might be trickle charging your cars battery from the solar, and you can plug small ac devices into your cars solar system. But at that point why not put in a secondary small 12 volt to run isolated from the cars starter battery, beyond charging off the alternator. So you dont accidentally drain the cars starting battery? but would still trickle charge the cars battery from the solar panel. I assume there is someway to do that with diodes of something. But believe that ac to dc 36volt charger depends on a batteries internal BMS.

 

[smokerx]

"AiLi Battery Monitor 100A - $38.50" wew, yeah nah big dog. Ill run them blind and depend on the BMS. Another option is the LY6W battery capacity tester. Which will give you only the charge state in percent. I think an issue with this one is when programming it. theres a certain range of cells you tell it is in the system to make it smart. Like 1 to 15 cells. However its possible that it programs based on batteries in Series not parallel. or vice versa, and not solely limited to 15 or 17 cells or whatever it claims is topped out setting. At least that is what ive attempted to interpret from explanations on how to make this thing work with bigger packs trying to guess at it across the language barrier.

 

[Supervstech]

To possibly get the price down, or just an option, is to consider buck convert the 36volts down to 12 volts. Since 12 volt inverters are more ubiquitous than 36volts. Pros are you can now tap in for a 12 volt supply line. Cons are the buck converter can bottle neck based on the amperage rating. Bottle necking you out of any peak wattage rating on the inverter, depending on the price = amperage rating of the buck converter.

Similarly, another option for the MPPT boost solar charge controller is replace it with a 12 volt panel to 12 volt controller to 12 volt battery to small 12 volt inverter to an AC powered 36volt battery charger mole. That might be advantageous, since hypothetically, lets say you want to just wire that to your car and use your car as a docking station for charging your bank. I assume theres some way to wire that in with your cars alternator to charge from both solar and gas powered generator. You might be trickle charging your cars battery from the solar, and you can plug small ac devices into your cars solar system. But at that point why not put in a secondary small 12 volt to run isolated from the cars starter battery, beyond charging off the alternator. So you dont accidentally drain the cars starting battery? but would still trickle charge the cars battery from the solar panel. I assume there is someway to do that with diodes of something. But believe that ac to dc 36volt charger depends on a batteries internal BMS.

Well… since those batteries went out of stock 3 years ago, the need for a solution other than what was done seems silly…

 

[smokerx]

Well… since those batteries went out of stock 3 years ago, the need for a solution other than what was done seems silly…

I think it applies to any 36v system. Not those specific batteries, its just the voltage. I guess the CTKEV300 is like "well wrap it but boys. We wont ever sell another unit. That hoverboard battery is out of stock." actually a 3.6v battery wired in 10 series seems pretty ubiquitous. and you cant tap a 12 volt. Also the LY6W looks like it would work. The C setting is cell but actually should be or is the S setting for series so a 10S5P would be set to the 10S or 10C setting and it should work. About half of the people seem to be able to get a LY6W working

 

[smokerx]

I think that shaves 135 bucks off buck converting and using the LY6W and spending a reasonable amount for a charger and you can tap 12. so $569 vs $369 and theres no way I have $369 into a 1KW pack.