A DIY Power Wall ???

[NEOH]

One of many DIY Power Walls being built ...

Is this "SAFE"? ...

http://diypowerwalls.com/t-Glubux-s-Powerwall

If it is not safe, then specifically WHY not?

Q1) He used a fine wire on each battery to individually "fuse" each and every battery - If sized correctly, is that enough? Or add real DC fuses?

Q2) He assembled with used batteries - Can that cause early failure? How?

Q3) It is installed in an separate, distant "out-building" - Why? Because it is a potential fire hazard?

Q4) Does the core temperature of each "parallel pack" need to be monitored?

Q5) Must each "parallel bank" be individually balanced?

Is there any way to make this SAFE?
If yes, then how exactly?

[Mike90250]

Fine wire is not a fuse, Better with a real fuse and holder (where do those bits of molten metal fall ?)

Matching used cells for Todays capacity is fine, What about in a month when they age some more ?

Used cells - where do they come from ? old battery packs? fleabay vendors? Are they all the SAME chemistry ? Used for 3 years, used for 6 years, nobody knows.

The final kicker - soldering the fine wire to the cell terminal. Guaranteed to ruin the seal and let the cell dry out. Factory packs are spot welded nickel wire

And yes, an outbuilding, nothing critical there to burn up.

[SunEagle]

Maybe another question to ask, is the time and money spent to build this system worth it if it fails violently?

You can't get your money back since their is no warranty but only blame for a DIY.

[NEOH]

Fine wire is not a fuse, Better with a real fuse and holder (where do those bits of molten metal fall ?)

Does a Tesla PowerWall use a fuse on each and every battery?

[karrak]

One of many DIY Power Walls being built ...

Is this "SAFE"? ...

http://diypowerwalls.com/t-Glubux-s-Powerwall

If it is not safe, then specifically WHY not?

I don't think it is safe for anyone who hasn't got a good understanding and practical experience of electronics to undertake a project like this.

I think it is safe if each block of cells is individually monitored and action taken to make sure that the battery as a whole remains balanced and no individual cells go outside their safe voltage operating range. I would also restrict the charge current to less than C/4 and discharge current to less than C/2.

Q1) He used a fine wire on each battery to individually "fuse" each and every battery - If sized correctly, is that enough? Or add real DC fuses?

The fuse arrangement is pretty well the same as used by Tesla in the Model S battery pack. It is one extra level of protection that you do not get using the large prismatic cells like the ones I have used. You still need a main battery level fuse to protect the wiring from the battery to the external equipment.

Q2) He assembled with used batteries - Can that cause early failure? How?

A good start is to capacity and resistance test each individual cell and only assemble the battery out of good cells. It is more than likely that cells made by different manufacturers or different cell types by the same manufacturer will last different amounts of time so either make the battery out of the same types of cells from the same manufacturer of roughly the same age or make sure you have the same ratio of different types of cells in each block of cells.

Q3) It is installed in an separate, distant "out-building" - Why? Because it is a potential fire hazard?

Yes, probably an essential precaution if you are not going to have any BMS!

Q4) Does the core temperature of each "parallel pack" need to be monitored?

Shouldn't be necessary if you keep the charge and discharge currents low and the voltage within the safe operating range, but good extra insurance and not difficult to do. Monitoring the temperature is important if you are running the battery at temperatures below 0^oC. You should not charge the battery below 0^oC.

Q5) Must each "parallel bank" be individually balanced?

This is very important!!!

Simon

Off grid 24V system, 6x190W Solar Panels, 32x90ah Winston LiFeYPO4 batteries installed April 2013
BMS - Homemade Battery logger github.com/simat/BatteryMonitor
Latronics 4kW Inverter, homemade MPPT controller

[PNjunction]

NEOH - this is just the modern alternative to this:

"I just found 35 lead acid batteries at the dump, and now I want to make it into my off-grid power system."

or

"I just picked up 200 7ah agm batteries from 1998 that were inside some old UPS systems. How do I wire up 200 of these old used batteries safely?"

After the safety warning, comes the LMAO. Get real.

For some reason, changing chemistry from lead-acid to lithium seems to be ok with trash-collectors. There is a large sales demographic for this kind of trash, and/or possibly "howto" videos for unsuspecting types that don't protect their wallet.

In the early days, and even today, you blew the tops off or severely gassed out the trash lead-acids. Today, THIS TYPE of lithium, the nominal 3.7v burns down your house.

HOW? When trash lithium batteries are left in a very discharged state for too long, chemical damage from secondary reactions occur. The real kicker is that the recharge process turns these chemically damaged cells into very poor performers, or as small roadway flares. Balance is all over the place. Each trash cell has different capacity and internal resistance

Basically, there are a lot of "one off" howtos and videos for trash sellers, and trash-buyers. Don't play with trash. Heh, unless you like using somebody else's toothbrush or underwear, you are better off knowing what you are doing with NEW / PROVEN cells.

[SunEagle]

Does a Tesla PowerWall use a fuse on each and every battery?

I would say that the inside workings of a Tesla Powerwall is still a closely held secret.

What it does have that a DIY system won't is a cooling system that makes the charging / discharging less likely to over heat and fail.

[jflorey2]

If it is not safe, then specifically WHY not?

Because 99.99% of the people in the US do not have the experience or knowledge to do this safely.

Q1) He used a fine wire on each battery to individually "fuse" each and every battery - If sized correctly, is that enough? Or add real DC fuses?

IF the wire is carefully characterized and tested, and attached correctly - yes, it can be enough.

Q2) He assembled with used batteries - Can that cause early failure? How?

By having some batteries with dendrites already in place due to abuse; only a few more charges and they can puncture the separator and lead to a fire. And once one cell goes up . . .

Q3) It is installed in an separate, distant "out-building" - Why? Because it is a potential fire hazard?

Yes.

Q4) Does the core temperature of each "parallel pack" need to be monitored?

It's a very good idea, although once you see a rapid temperature rise it's generally too late.

Q5) Must each "parallel bank" be individually balanced?

Yes.

Is there any way to make this SAFE?

Completely safe? No.

Relatively safe? Yes. If you:
- have a documented trail on each cell (i.e. where they were used, how they were cycled, if they were ever overcharged etc)
- test each cell for capacity and ESR and throw out ones that do not meet your specs
- have a BMS that will ensure balance at all times
- mount them in a material that will prevent one burning cell from taking out the entire pack
- have individual disconnects/fuses for each cell
- do long term trend monitoring

then it can be _relatively_ safe.

The question is - why would you do that? There are plenty of batteries out there that are a lot cheaper than the above would cost.

[NEOH]

With 40 used cells in each parallel block, how do you determine which cell is weak or has failed ?
If the "fine wire" fuse is properly sized then it will blow when that cell fails in a short circuit mode - this can be seen visually.
But if one or more cells in a (soldered) parallel pack has become weak, then how can anybody determine which cell(s) of 40 are weak?

He now has 1,120 used cells @ 48 Volts total in the shed.
If each "used cell" lasts 3 years or not, then that means on average one failed cell per day, every day, forever and ever.
Is this a maintenance nightmare?
Is the (never-ending) huge number of man-hours involved for purchasing, testing, sorting, assembling and maintaining worth the final product?

"DIY Power Wall" now has nearly 500,000 google hits.
Where are all of the catastrophic DIY Powerwall failure videos?

"Powerwall" is a trademark of Tesla and yet he displays the Trademark on his "DIY Power Wall" in the you-tube video.

[jflorey2]

With 40 used cells in each parallel block, how do you determine which cell is weak or has failed ?

That's an excellent question.

If the "fine wire" fuse is properly sized then it will blow when that cell fails in a short circuit mode - this can be seen visually.

But that's not really the problem. The problem is that the two most common causes of "short circuit mode" - physical damage and dendrite growth - will cause the cell to catch on fire. And once one cell goes, it will heat the cell next to it hot enough to cause it to start burning. At that point, your only hope is that the building it is in has concrete/masonry walls, and that the contents of the building can be easily replaced.

So seeing the blown fuse really doesn't buy you much.

But if one or more cells in a (soldered) parallel pack has become weak, then how can anybody determine which cell(s) of 40 are weak?

Again an excellent question. Per-cell monitoring will do that, but that's expensive and complex.

He now has 1,120 used cells @ 48 Volts total in the shed. . . .Is this a maintenance nightmare? Is the (never-ending) huge number of man-hours involved for purchasing, testing, sorting, assembling and maintaining worth the final product?

Another good question.

(BTW kudos for asking all these questions. Once people start asking these questions, they are in a much better position to evaluate claims made by battery manufacturers.)

"DIY Power Wall" now has nearly 500,000 google hits.
Where are all of the catastrophic DIY Powerwall failure videos?

Perhaps he is disincentivized from posting such videos?

"Powerwall" is a trademark of Tesla and yet he displays the Trademark on his "DIY Power Wall" in the you-tube video.

He may argue that "power wall" is generic and distinct from "Powerwall" the trade name. If it does go to court, lawyers will make a lot of money.

[Mike90250]

...He may argue that "power wall" is generic and distinct from "Powerwall" the trade name. If it does go to court, lawyers will make a lot of money.

No, the lawyer gets a burnt out shell of a battery shed !

[Mike90250]

Here's a illustration of a couple less then perfect Li-Ion setups ( which look to be commercially sold in a fancy case)
Li-IonFires.png

[max2k]

With 40 used cells in each parallel block, how do you determine which cell is weak or has failed ?

Weak cell will lead to overall less capacity. Failed (shorted) ones will blow their individual fuses. I'd use normal fuses instead.

...
He now has 1,120 used cells @ 48 Volts total in the shed.
If each "used cell" lasts 3 years or not, then that means on average one failed cell per day, every day, forever and ever.
Is this a maintenance nightmare?

it is

Is the (never-ending) huge number of man-hours involved for purchasing, testing, sorting, assembling and maintaining worth the final product?

IMO- no

this whole setup is not worth effort/time/risk, especially when individual 100Ah-200Ah LFP cells are available. The cells from the site are Li-ion which are much less suitable for the application.

[NEOH]

RE: The "fire" in message #12 ( two posts above )

What is the "fire code" for the Power Wall?
Must it be installed on a non-flammable wall - like brick?
What if it is mounted outside on a vinyl sided house with a wood substrate = major fire hazard?
The Insurance companies will have a problem with this increase risk.
They hate anything that increases their risk of lost profits, and rightly so.

Are these PowerWalls ever installed inside the home - in the living quarters? in the basement?
It look like a significant amount of toxic smoke & fumes were released.

If each cell is individually fused, then how did the first cell catch fire, as seen in message #12 ?
If the individual fuse did not blow then one or more the cell(s) overheated while operating at "normal" amps.
So then the Temperature Sensor, if there is one, failed too?
At some high temperature, the circuit should have opened and stopped all charging amps or discharging amps, long before the ignition point?
Is it possible for a cell to continue in thermal run-away, even after the circuit is opened?

You can't (cheaply) monitor the temperature of each and every cell, so this seems to be very risky in hot weather.

Does it make more sense, from a safety point of view (increased cost), for each cell to fully protect itself from volts, amps & heat ?
Then the manufacturer can add another layer of safety to the battery pack as whole.

If this passed inspection then it must be UL Approved".
I wonder how UL tested these battery packs?

[adoublee]

UL 1973 is for batteries used in a stationary application. More recently, UL 9540 is for the whole system.