Bob's Powerwall Project

I just use a standard glue gun. I actually go a small one (used for small arts/crafts; got it in that section @ Wally World) as I didn't want to deal with a large stick of glue and handling a large gun.
I think a glue gun is a glue gun in this instance. I doubt it makes much of a difference like the soldering irons do
 
Korishan said:
I just use a standard glue gun. I actually go a small one (used for small arts/crafts; got it in that section @ Wally World) as I didn't want to deal with a large stick of glue and handling a large gun.
I think a glue gun is a glue gun in this instance. I doubt it makes much of a difference like the soldering irons do
Click to expand...

Thank Korishan, I am alway prone to buying the wrong crap thus the reason I ask. I agree, it's not like the soldering iron. I will likely get something from Wally World.
 
RobertBaumer said:
owitte said:
mike said:
I think you are wasting time and electricity doing a refresh test (3 iterations). If the cell is bad, it's going to show poor performance on the first test, in terms of heat and/or low capacity.
Click to expand...

yes, that's right. Me too, I'm just doing one charge/discharge/charge cycle. But only the discharge part is done by my OPUSses, both charge parts are done by TP4056s. I made two banks of them, one for pre-charge and the other one for charging after the test, so I don't get confused which cells are already processes and which are not...
Click to expand...

I just ordered the parts to build 40 of those TP4056 and since my computer PS can handle up to 120amp, I should be good to go.....

Here is the video I watched:

I bought 4 packs of 10 of these:http://www.ebay.com/itm/27258452672...49&var=571700784834&ssPageName=STRK:MEBIDX:IT

and I bought 5 sets of these: http://www.ebay.com/itm/172546327816?_trksid=p2057872.m2749.l2649&ssPageName=STRK:MEBIDX:IT

I'd like to find that wire stripper he uses in the video.

-Bob
Click to expand...

Hi Bob,

here's the wire stripper, I have one of those: Pressure adjustable, works fine from very thin wire to thick solar cables.

And here's a picture of one of my TP4056 chargers:

image_olzdaa.jpg

As you can see, I also use it to charger other types of Li-Ion cells :)

Have sun!
Oliver
 
Korishan said:
I just use a standard glue gun. I actually go a small one (used for small arts/crafts; got it in that section @ Wally World) as I didn't want to deal with a large stick of glue and handling a large gun.
I think a glue gun is a glue gun in this instance. I doubt it makes much of a difference like the soldering irons do
Click to expand...

I just bought a Ryobi Lithium and it was worth the money. I used it today and it works like a charm and not having a cord is awesome. I am sure it runs on 18650's!
 
LOL, Awesome! And when the controller BMS goes dead, you have some more lithiums to grab ;)
 
Woo hoo, I have almost enough cells for my second pack. I am debating on waiting until I have all 14 packs worth of batteries and then start sorting or just try to stay with a 2500 mAh average. My first pack came in at 302,619 mAh at 120 cells. I can soon start posting pics! Big thanks to Mike and Korishan for the OPUS advice, now I do charge/test rather than discharge/refresh. Now I just need my TP4056's to come in, they are on one of those slow boats from China.... Oh yeh, and those 3 OPUS chargers that I believe were totally fake. I give them till June 28th to get me the product, if not, follow through with getting my money back from eBay
 
Help please.....

I have an old Dell 42u closed cabinet that I plan to reuse for a battery cabinet. Being that it is only 19" wide or 20" if I strip out the posts, I can on fit maximize capital by putting in two banks of totaling 14s450p. Does anyone see a problem with this? I have been looking on eBay for shelves that will supply the weight and I plan to use a .25 inch acrylic to insulate the bottom and another .25 acrylic as a guides. The shelve can only hold 500# o weight each.

Please let me know if this is doable.

Bob
 
Note that 14s450 is 300kg of batteries :)

If you start ripping out the posts that in most of the rack cases do carry the load it can be a bit flimsy.
Im not sure you should squize it to much. Some air between the packs is healthy for them :) On the other hand someone need to test.
 
Crap, that means the shelves I am looking at will not hold the weight, it looks like I will need some metal reinforcements, oh wait, that 14s450p would be distributed across 3 shelves and do to size I could only fit 6 shelves. I definitely am going to be a guinea pig for this one. If all goes will I will have two banks of 14s45Op capable of delivering almost 100 KwH of service.

I have another idea, has anyone toyed around with a RAIB50, Redundant Set of Inexpensive Batteries? In my configuration I will have a 15th battery set that will not be usable, but was looking for use it for an automatic spare. I guess this is just a dream or perhaps a battery management solution can do this.
 
2x14s450 is 600kg... Not many modern raid cabinets that like that weight :D

RAIB5+0 ?

I have spare pack. But its manual labour. No use at all to have one spare that is automatically gets inserted if you got 2 or more strings. I run 4 main strings and 3 secondary so i can easily just swap out 1 string until i have fixed the faulty pack.
 
While I like the idea of a redundant pack to be able to put in the switching needed to be able to put it where it is needed automatically would be nuts. I'd rather have it sitting partially charged for when a cell has a problem and then it can be swapped out to keep the rest of the batteries operational.
 
I agree, I guess I was having a maxi-zoom-dweebie moment and try to figure out how to automate something that does not need it..... thanks for the advice.... I suspect I will just keep am extra pack available i needed.
 
There is however somethink you could use a spare pack for:
Active balancing. I have been think about it, and perfect (but slow) active balancing could be achieved using a spare "flying" pack and some relays (14S-pack would need 14*1+5 relays to be precise).

There are different algorithms for this, but I would use:
Step0: Log all cell-levels
Step1: Select Higest and Lowest.
Step2: Connect flying to higest for x minutes
Step3: Disconnect for x ms
Step4: Connect flying to lowest (from step1) for x minutes
Step5: Disconnect for x ms
REpeat, and keep doing this as long as system is operational. Under charge, under load, allways.

Has it been done before? yes. Its called flying capacitor balancing - but this uses solid-state and caps and can therefor not overcome a minimum voltagedrop through the FET-switches for balancing - I would use mechanical relays, with a specific loadconnecting relay (protected with freewheeling diodes), and the rest as normal. 12V auto-realys cost nothing and are good for millions of swithes under no load, and balancing would converge to 0.00V difference.

Try a google for "flying capacitor balancing"

Main advantage - active balancing "for Free" since we allready have some spare batteries :)


image_umucnb.jpg
 
Greetings All, after dealing with some major health issues I am back, not sure how back yet, but back. Anyhow, in the next 3 weeks I expect that Tesla will be out to install 9.1 kw of solar panels, I refused their power wall for reasons you all are too aware of.... I have begun emptying my Dell rack in my basement, need to sell my blade centers, this project is more fun than the 25 servers I have in my rack. The joke in my neighborhood was that when I turned all my systems on the lights would dim. Needed to water cool my electric meter, well now I hope to get it to spin the other way and I placed my computer lab on my 24 core 128gb MAC. For years I have been sucked into over clocking while heating the house. Now I have found you guys with Pete being the king pin and now I have a new hobby.

Today I finished my first PAK, finished as in all the batteries have been tested over and over. I re skinned them, I know it does nothing to the function, but it is like OCD or something. I have decided to buy 6 rack shelves for my 42 Dell rack shelves which are 19" x 39" and place 7 packs on each shelf. My packs are 240p, so I will be 14s 240p with the goal of having 14s720p x 2500ma x 3.7v. Since Tesla will not allow me to directly access the PV system they will be installing, I am going to have to go the AC Coupling route. So I am watching all the videos and reading all the white papers I can. So right now I am trying to determine the best bus bar design to use. After watching Darrel's video on the 906a spot welder, I think I am going to buy one to use because the Fast Axial Glass fuses are so small and their is not a lot of real-estate. I also am using Mike's suggestion of 6 awg copper wire.


image_traoln.jpg

image_vegats.jpg

image_mwkvac.jpg


Perhaps if I am successful with the AC Coupling I will start a thread of my build..... for now, looking for advice on my pack deign.

-Bob

Just got off the phone with Tesla aka Solar City and they scheduled my install Sept 9. Guess I am going to have to hide all my DIY stuff :) Thus far I like the Magna Sine 48 volt inverter.... hoping to get some input on that from someone who has experience.... Also, the electric company has to install a Net Meter and they say that can be 6 to 8 weeks..... ouch....
 
Looking awesome Bob
 
Huge packs! Good work m8 :)
 
Amazing man! Those packs are gigantic! What do you expect your capacity to be?
 
I am just finishing up on testing to full a second, I have an idea of a bus bar design, just waiting for the spot welder because I want to use the Glass Fast Axial fuse. Each pack will be 240P with average of 2.5 giving me like 2.22 kWh each
 
You must log in or register to reply here.
Back
Top