Wolf's PowIRwall

[Wayne's World]

Extra capacity is just a matter of another 14 packs tied in with the existing packs. Not sure what you mean by that.

How are you going to connect them. - Will each new 1s pack be in parallel with its original, and so make new cables from one pack to the other, or a completely separate 14s system, and parallel the entire system at the overall + and - ?

I'm thinking of using either XT or Anderson plugs for each pack, and making Y adapters to add extra 1s packs to each. Once I have at least 2 packs in parallel, this would allow easy removal of a pack for servicing and not interrupt the power supply, albeit that remaining pack will go out of balance if left by its lonesome too long. Hence the benefits of having a spare pack or 2!

I see you went with the Batrium 4(?) that uses Longmons. Any reason why that and not the later versions. I've watched a couple of their vids but not yet across the application for each.

I saw in one of your photos that you had a bunch of different cell testers. Did you find varying results and how did that effect your database and cell selection criteria? I'm presuming you've got a thread or videos dedicated to this?

 

[Wolf]

How are you going to connect them. - Will each new 1s pack be in parallel with its original, and so make new cables from one pack to the other, or a completely separate 14s system, and parallel the entire system at the overall + and - ?

I see you went with the Batrium 4(?) that uses Longmons. Any reason why that and not the later versions. I've watched a couple of their vids but not yet across the application for each.

I saw in one of your photos that you had a bunch of different cell testers. Did you find varying results and how did that effect your database and cell selection criteria? I'm presuming you've got a thread or videos dedicated to this?

I believe I will parallel the entire system at the overall. Unless someone can give me a good reason not to. My whole 14s battery is designed with many factors in mind. As you know I am not just throwing arbitrary cells together into a pack. I try to match them as best as I can with capacity and IR.
My next pack I am going to be much luckier at because I found a boatload of LGEBM261865 2600mAh cells, 1600 to be exact with an average IR of ~38mΩ to ~40mΩ. Testing at 95% to 100% of capacity with my SkyRc MC 3000. The balancing of this one should require much less work.

I chose the Watchmon 4 because it has the capability of a higher balance current. Besides the later versions are designed more for EV batteries not 18650s.

Oh yea I have and had Opus bt-c3100 v2.2, Liitokala lii-500, Zanflare c4, Xtar dragon vp4, Skyrc mc3000, and Foxnovo F-4S as far as the commercial units are concerned.
The Liitokala lii-500, and Zanflare c4s I had I gave away recently here on SLS. Their mAh results where a tad to high for my taste ~7% to 10%.

I do have a thread on the comparison of chargers LOL https://secondlifestorage.com/showthread.php?tid=6428&pid=41085#pid41085

In my spreadsheet I recorded what charger/tester I used on each cell and it did influence me whether I chose the cell or not.
I.e. if a cell had marginal mAh and SOH including IR within my criteria for the pack and it was tested with a Zanflare it was a serious strike against the cell and if I felt that it was not worth putting it into the pack I didn't. I also re-tested some of these cells in my Skyrc mc3000 ( The best one out there but expensive and a relative steeper learning curve on how to operate it.) that I had doubts on and was proven right with lackluster performance.

If I were to do it all over again. Stick with one charger/tester brand! If I had know what I know now it would be the Foxnovo for the cheap and the SKYRC for the pricey.
Wolf

 

[Wayne's World]

I've opted for the cheap option for chargers. I'll have 6 Lii-500 and a 500s once they all arrive. At least they'll all measure similar. I'm getting pretty solid results, but having good cells to start helps this. I've just converted a DELL 500W PSU to drive these and a iCharger 4010Dual that I'm waiting on stock.

I'm getting heaps of tool batteries and the most common cells are 25R and LGABHE, followed by 30Q then 20QR. Should get some decent capacity packs out of those. I'm hoping I'll get enough 30Q so that'll keep the cell count requirement down.

In years to come 21700 will become available as these are starting to be put into high performance packs by tool manufacturers. I've already got some out of a Dewalt and a Milwaukee, and I know Bosch and a couple others use them.

 

[Wolf]

I'm getting heaps of tool batteries and the most common cells are 25R ........................................

That's great.
If you stick with just tool batteries you will be working with high drain chemistry. Most have very low IR in the 13mΩ to 35mΩ range.
INR and IMR chemistry is optimal for high discharge and low temp if you can go exclusively with those that is super.
Also most of the time I have noticed that the internal BMS and or fuse goes bad rendering the pack useless but the cells are still in very good shape.

Best of luck
Wolf

 

[Wolf]

Update,

So I have been playing around with the Batrium and I like it.
It is very granular in all the settings and can be confusing at first but if you understand Aussie logic it starts to make sense.
Experience with arduino and ESP8266 and ESP32 programing helps as it is based on that.
My 14 Packs which where discharged/charged for storage voltage got hooked up as you know and I have them on a ~1.5A trickle charge to get them up to snuff and ready for production. Can't wait to finally get my PowIRwall into the Projects on SLS as a working example.

So from ~3.6V per pack over the last week or so they have gone to ~4V. So far no balancing has happened as they haven't reach the threshold yet.
My SoC% Low Cutout is at 85.0 and I'm only at 58.0 as you can see by the pic.
All in all I am very pleased with the performance of the pack so far. A 0.09V deviation between the packs is well within the acceptable norm I think and takes into account the variation of the longmons which I believe is + - 0.05V.
As all of you know me, I will be comparing the longmon V with my fluke DMM and report the discrepancies but I doubt that they will be more than 0.03V either way. We will see.

I also got the telemetry posting to Thinkspeak and hope to find a way to get this info into my influxdb at home. I do not like to send all this information into an unseen cloud. Call me old school............

So hopefully this weekend the 6 additional lexan sheets will be done and I can then move the packs into the shed and start using it.




That's it for now
Wolf

 

[Wolf]

Update,

This weekend was the very productive.
The glass guy didn't come through with the lexan sides so be it. I have to take the packs apart anyway once I get the box built to install them into their permanent enclosure. So that is when I can add those lexan sides.

The packs where moved to the shed and are now in a testing and staging state.
The wiring is all preliminary and will all be cleaned up and organized so for the OCD's like me have no fear.



Reset my charge controllers to Li-ion settings at 58V and hooked everything together. 4.14V per pack.
I will probably adjust that to 4.1V or possibly 4.05V as I see how the pack is working and what the SOC is as Batrium does it's magical calculations.
The Sun did its job as requested and finished charging the battery pack.
Pack balance does not seem to be a problem at all as all packs have maintained a ~0.06V average difference between packs during charging and discharging.
I have not needed to engage any balancing burnoff as of yet.
Saturday evening at 8:30 pm



Sunday morning at 6:30 am



Monday morning at 6:00 am




So far I have been running on solar and storage 24/7. Nice
You can see by my history that at night the Pb batteries would fade out rather quickly.
I did have the Pb cutoff voltage set very conservative as not to damage the batteries.
But the Li packs I can discharge way further without damage and it shows!! As it keeps my house from dipping into my wallet to pay for grid power at night.




Lets see how these carefully designed packs perform in the log run.

So now I can get onto the projects board right Mike?

Oh and on another note the next pack is already in the beginning stages of testing.
I managed to score 1600 LGGBM261865 cells that look really good. Rated at 2600mAh
I'll keep you up to date on that.
Box of liberated ones 2 deep.





Wolf

 

[floydR]

I believe you will be pleased with both the capacity and ir of the LGGBM261865 cells have tested 1200 LGGBM261865 cells and only 10 were below 2550 mahs or damaged in some way. Have another 1200 to remove from 36v batteries. most have tested in the 2650- 2750 range with an IR of 38-44 mohms.

1190 LGGBM261865 cells tested<2600+>, 1200 LGGBM261865 cells to test, 300 modem cells lgabb41865 tested<2500+> Plus 40 4ah 36v hoverboard batteries, 15 dismantled hoverboard batteries.
later floyd

 

[Stigagno]

Hello, some post ago you write that you want to build a box outside house for the packs and for the temperature management put on the top one flap-vent and two on the side. This is also my goal for safety and space reasons. Im planning to build a box with insulated panels 4 cm thickness and a substructure of steel square tube 40x40x3mm. For the flaps, vents I have doubts about the choice of ventilations to avoid the entry of water and humidity especially when there is wind together with rain. Have you found ventilation grills suitable for outdoor use and well tolerating bad weather? Thank you Luca

 

[Wolf]

Hello, some post ago you write that you want to build a box outside house for the packs and for the temperature management put on the top one flap-vent and two on the side. .....................................Thank you Luca

Stigagno,

Actually the 'Battery Box" will be in a shed. so the box will have a protected environment other than the cold.

I am mainly building it to maintain a >2° C temperature in the winter as it gets quite cold here at times.
The cooling fans and vent flaps are just in case the batteries and the batrium longmons get hotter than normal ambient temp which I do not anticipate.



Wolf

 

[Wolf]

Update,

It has been a stellar week and my poweIRwall so far has exceeded my wildest expectations.
My average kWh consumption per day is ~15kW and in the evening and at night is when I would use 6 to 8 kW.
My Pb based system would not accommodate that as I was very conservative with the batteries and never letting them get below 12V.
With li-Ion I could now change that discharge curve if needed all the way down to 3.2V and still maintain a margin of reserve.

Adjusted the charge controllers to Li-Ion set the max to 56.7V adjusted the Grid ties to shut down at 44.8V and all is good.
The only issue I had was with the Batrium system which was resolved as described in this tread. https://secondlifestorage.com/showthread.php?tid=9173&pid=62888#pid62888
As of Saturday the 16th I have been running on Solar and Lithium. I still need to make some small adjustments on my grid ties but other than that my Grid consumption has gone to next to nothing. 1kW a day!! In my neck of the woods that is 13 cents.




Battery box is half done and I am continuing the testing of cells for the second pack.

Goodby my old friend




Hello new




Wolf

 

[Wolf]

Update

So I finally got the battery box built and wired. Battery box environmental control system is functional.
Still have to work on installing the vent flaps,servos and heaters. But at this time of the year I doubt I will need heaters although at 8C this AM on June the 1st I'm starting to wonder.
I apologize for the foggy pictures I must have been so excited to get this far this weekend that I was extruding steam
I will update later with clearer pictures.
Some wire management is still in order but all in all it is functional.




ABB Shunt trip breaker installed but still have to wire the trip to the Batrium.



Batrium at its final resting place.



Battery box metrics are sent to influx and visualized with grafana.




I then wanted to get some of the batrium data over to my influx database so I ordered a RasPi and downloaded Daromer's image and played around with it. The image loaded just fine and worked but for me it was a bit too much. I probably will use it in the future if/when I get a decent charge controller but for now my MakeSkyBlues are working just fine.




I just needed a simple UDP listener to port the batrium metrics to influx. Found a sketch for a udp listener that published to thingspeak,
a little code revision here and there got rid of thingspeak and now have the metrics going to influx with a $10 ESP32.
I will massage the data and metrics ( still have to figure out some of the readings collected) but for now the essentials are being recorded.




That's it for now

Wolf

 

[Wolf]

Small update:



The battery box environmental control systems are working as programed
At least so far, winter will tell if the heaters are enough.





Wolf

 

[Wolf]

Update;

What to do when you have too much time on your hands?
Play with Node-Red of course.
So I saw this post https://secondlifestorage.com/showthread.php?tid=8645&pid=59143#pid59143
that Bloater did on Node-Red picking up the Batrium UDP and creating a little dashboard with the cell metrics.

I got his Flow and imported it and checked it out. I must say Bloater did a lot of heavy lifting!!
Now how to get this data and a bunch more into influx..............
A node here a node there, a little code here and a little code there and wella all done.
Now what am I going to do with my spare time Oh yea build another Pack.




Nice day and completely charged!



Wolf

 

[Wolf]

Update;

So I have been running on my powirwall for a month now and looking good.
CDC cycles are looking good Batrium is working good and my electric bill is looking much better.

Grid import chart for 60 days
Can you tell when I switched over to my PowIRwall?
Did have a couple of cloudy days and had to have the grid back me up.
Cant wait to get my second pack done.




Metrics for this AM. Making coffee took .07 of a volt for about 15 min.
Nice to see the packs staying very close together even at a discharge.
Not bad for a bunch of used 18650 batteries.
6hr chart




24hr chart




Wolf

 

[Wolf]

Small update

The environmental system is up and running quite nicely.
Pretty warm in the plant "Shed" today.
Wolf

 

[Wolf]

Alright its been awhile since the last update.
I got involved in DC IR and other stuff but in the background I was still working on my 2nd 14s80p pack.

I managed to procure 1600 of LGGBM261865 and LGEBM261865 commonly referred to as the LG M26 (2600mAh) cells.
Date codes varied from 27-Mar-2017 to 30-Nov-2018.
After thorough testing I filtered out the best 1120 cells needed for the pack.
Lowest V drop after 30 day rest, Lowest IR, highest SOH, and newest cells by date code.
I whittled the results down to 1124 cells coming up with the best of the best.
At that point I could have just thrown all the cells into holders at random as they were all really close to each other.
But no I wanted to punish myself and to keep track of all cells and which pack they go in so I ran them through Nemos Repackr.
Numbers look good with estimated capacity at 216.68Ah
The Average cell mAh outcome was2705mAh and average IR (AC)was 35.11mΩ with a low of 32.84mΩ and a high of 36.95mΩ.
If Nominal capacity is to be calculated as by the spec sheet at 2600mAh then the tested average results of 2708mAh is a 4.15% elevation due to charger inaccuracies. Not bad all in all.




1120 Cells all sorted and ready for spot welding




I do love the KWeld it just works great!






5 packs spot welded 9 to go.




Testing the Ah capacity of the first pack.
Procedure is to charge the pack to 4.2V with a 100mA cut off and discharge through a resistor to 3.0V (Cell min is 2.75V)
I will be doing 2 packs this way and from then on I will be using regenerative discharge.




Great results this AM, 207.42Ah which gives me an average of 2593mAh per cell that's 99.73% of SOH with a discharge to 3.0V.




Wolf

 

[gauss163]

^^^ Hmm,2705mAh is 9.5% higher than HKJ's testat 1A (2470). What rate did you discharge them at? Perhaps Opus is to blame for that (maybe older version units)?

But 2593mAh looks more accurate if it was at low rate (best HKJ got was 2551 at 0.2A)

Congrat's on the build.

 

[OffGridInTheCity]

When these are ready... do you have 14 packs and this will take you to 28 packs? Will you be having 2 x 14s batteries or paralleling these new packs in with existing packs.
Its so much fun to get 1,000(s) of cells hooked up ready to charge and discharge!

 

[Wolf]

Well I don't care about HKJ's tests other than a reference. His cell was dated 10-Dec-2018. Maybe it was a bit cooler than my environment.
Cells where C/D/C at 1A where applicable and possible OPUS, SKYRC, and 1A/0.5A for all others.
Opus C/D/C units are 2 years old and BT-C3100 firmware 2.2 all ordered fall of 2018.
All chargers that were used in pic below.




The OPUS albeit the worst was not the only offender in the whole scheme of things Foxnovo and Vapecell close seconds. And if we are to take HKJ's tests as gospel then the XTAR had it all over them including the SKYRC.

Here is a chart of individual slots on all testers and how they averaged out sorted highest to lowest.
OPUS 1 slot 4 was constantly high. So yea OPUS is to blame . You see there is no perhaps something is to blame, I know what is to blame, as I record things very meticulous and have all my facts in a personal format from my experiences.




The #1 out of 1480p pack discharge was done at ~20A translating to ~250mA per cell.

Congratulations?
Thank you but a bit premature 9 more packs to spot weld, 26 more buss bars to build and -.
13 more packs to test to be sure Ah are close. All the connector cables to cut, crimp, and heatshrink.
Lexan covers to install along with the batrium longmons.
The list goes on before this pack is ready for production. But if you stay tuned the progression will be posted here as new data and info comes available.
Wolf

 

[daromer]

Wolf: Perhaps you said this before but what do you use to check Humidity and temperature and what is used for the PWM to the fans ?