Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
First build questions
(02-13-2019, 10:52 PM)Dallski Wrote:
(02-13-2019, 02:15 PM)rearden Wrote: Thank you for your replies!

Enphase:  It is being considered as the first inverter because it is what I have on hand.  I can get 215w/h out of it or approx. 5kwh/day off my base load per inverter.  Constant ~10A draw at 24v.  I do not live in a Net Metering area.  I am still testing with some lead acid batteries.  I have an EPEVER tracer 40A solar charge controller which I also use for low voltage cut off and load control.  Start up current is a problem.  I assume the enphase MPPT does a sweep and since it is expecting a solar panel to come up gradually it looks like a short to the current source at the beginning.  At the moment I am using a 2.5D-20 NCT as a current inrush limiter.  I want a better solution that would swap to a shorted connection after a period of time.  I am hoping that maybe a sound system inrush limiter may have a drop in circuit I could use but I have not found one yet.  I have found ones for AC, but I don't know if they will work with DC.  Suggestions?

I have 16 Enphase inverters connected to the grid.  What I really want is to divert all exportable power to battery charging for use later.  Next would be to also have it available when disconnected from the grid like a UPS.  And a pony.

BMS:  So one vote for Electrodacus and one vote for Battrium.  I don't need the MPPT solar controller necessarily, I would rather divert exportable  gridtie power to the batteries and keep AC house loads as first priority, but I don't know how to do that yet.  I am looking at a power diverter from one of the guys at Open Energy Monitor but getting that variable source as an input in to a charger is something I don't know how to do and probably needs to be moved to a different part of the forum.
In your view, what useful features does the Battrium offer compared to Electrodacus?

Cells:  So NiCd is out.  I live in the US.  I can look around for more old laptop batteries.  Cost is an important factor.  It would seem to be fairly easy to run up the cost so that the DIY is as expensive as a retail Powerwall.  What should I look for in purchased cells if cheap Chinese are not recommended?


Were you successful in getting the m215 to power up with the inrush current limiter? I experienced the same "short" with a battery connected and thought I fried the microinverter. BTW, it's not that the microinverters have a gradual startup, it is just that they are not expecting too much more current than they are rated for. For example, if you have to do maintenance on a solar panel midday, you cover the panel with something opaque, and when you are done, you remove the opaque covering. The microinverter is instantly exposed to the full power of the solar panel and doesn't get damaged. Could even be like 350 watts from a 320 watt panel depending on a lot of factors. But that's still a lot less than what a battery can provide. 

Based on your motivation, it seems like you should look at this post: It talks about using grid tie inverters like your m215s and a second compatible inverter to run backwards to charge batteries when there is excess power. You can then have a contactor to switch from off-grid to on-grid when there is no solar and your batteries run too low. 

Electrodacus is making a SBMS0 which does not do MPPT, it is strictly a BMS. It was going for $130 shipped in his last Kickstarter. Batrium has a lot more customization and functionality and 5x the balancing rate. Each system has its pros and cons. Only you can decide what is right for you and your powerwall.

So far it has worked.  I have only used it for a few days because of the rain and testing time, but it seems to work.  The load is reporting about 240w in and I get around 220w out.  I ordered some 1ohm versions and I will test those, but I think they will allow too much current.  I am still looking for one which will switch to a shorted relay after a period of time and take the NCT out of the circuit.  If could get one which would take it out of the circuit, I would use an even higher resistor. I think the 2.5 ohm version will limit the current to less than 20A initially.

I will check out the inverters link and look at the non mppt SBMS0 version.
Generic likes this post
I have cleaned several hundred 26650 LiFePo 9.6wh 3.2V cells, more to come, obtained some 4x4 26650 holders, 100 5.3A fuses, a Weller 90W soldering iron, some 100g of 1.5mm lead free rosin core solder and a small tub of flux.

I have started soldering the first set of fuses.

Now I am deciding upon how to organize the cells and what to use as cell pack bus bars.  currently 8S 24v.  I could reconfigure to 48v 16S easily.

I want the flexibility to add to my battery bank as I get new batteries and not have to wait until I get several hundred batteries before i can add to the capacity.

I will start with an 8S16P or perhaps an 8S8P.  To maximize parallel buss bar space.  I will have the parallel batteries in a row of 2 or perhaps 4.  If I use the 4 columns that I would have to collect 8*(4x4) batteries for each expansion which is a lot.

I have not come up with a good idea on how to handle the expansion packs.  I don't have a good hot swap/add system.
It seems like extending the parallel buss bars, dropping in a new set of 4x4 and then soldering in to the pos an neg bus is just not that elegant or easy.  and I would have these hot #6 wires extending and hanging around.  I guess I could sheath them with some pvc.

From my calculations: in a 24v 8S16P = 128 @3.0V batteries, at 1000VA I would draw 2.6a from each battery and 41.67A from each 16P buss and at 6000VA 250A.  which is a heck of a lot and I would need more batteries (at least 512, 64P) to stay below the fuse rating.  That would require 4-6 awg to solder the fuses to.   And probably .2 x 8mm nickel strips across each row of 4 batteries to the buss. And each BMS lead will go to a separate P pack.

Comments?  Better ideas?


Attached Files Thumbnail(s)
Some of those solder joints appear a little 'dry'. Also, be careful of stray blobs of solder, as they could drop inside the vents, and inhibit the function of the CID. Or slip over the plastic ring, and short the cell out. There is only a mm or so between the positive tip, and the negative outer ring.

Who read this thread?
93 User(s) read this thread:
hbpowerwall, daromer, Sean, mike, owitte, Korishan, PAF, jdeadman, wim, watts-on, Dan Lim, Stefaan De Ridder, BlueSwordM, Geek, mr_hypno, Grumplestiltskin, Oz18650, djuro, maunz, Lindsaymac1, Chablis_m, rodagaster, rogales141, Redpacket, Ibiza, Bubba, Riplash, 100fuegos,, Mikethezipper, chuckp, MGPTWEGB, emuland-metroman, Chass90, Truth, kremvax, Shawndoe, deniscroombs, Wolf, Lekr, camthecam, Generic, Speicher, PaulDane, richie241, ajw22, BmoreThor, RikH, lloulou11, TimBelgium, mart1nn, OffGridInTheCity, Maniac_Powerwall, LEDSchlucker, zoookii, 1Lukas1, tsmurray, kurt13r, Richard, Overmind, TPranger, mskip, Church1182, bournevexed, beserker786, intra, Lachi, zeme, Edster, Lucan, strat24v, rearden, mileoca, Randy.skidmore, BarryG, ironzap2, planepilot, navmanl200, CU17, Boliver, Ebarnett, 100kwh-hunter, GeneralDJ, ben teker, PhilipW, billvon, yocor103086, Saiya, krobertson, stok, Cody, kpeters000, Slayer1

Forum Jump:

Users browsing this thread: 1 Guest(s)