diy solar

diy solar

NMC Bank, Bus Bars and Batrium Love

This hasn't been a "set it and forget it" situation. I've been changing the balancing voltage to ensure all but #3 are getting burned off as much as possible. I adjust it 5+ times/day as they move up and down. There's usually a point at night when all cells drop below the balance voltage, and I have to adjust it when I wake up. Then during charging, I have to adjust it up to keep from accidentally burning off #3 too. I've missed it a few times as you can see #3 has burned off > 1Ah in the post above.
Did you know there is an auto balance function?
It is called auto level. I haven't tested it to the depths that you would, but in theory it's only going to burn voltage off the highest cell(s).


I'm 100% with you on wishing there was a real manual. Not just hundreds of unlisted FAQ pages (support articles)and videos. It might work better if there was a table of contents (for the FAQ) at a minimum.


Edit: I just wanted to add that the build looks great. Love it
 
Did you know there is an auto balance function?
It is called auto level. I haven't tested it to the depths that you would, but in theory it's only going to burn voltage off the highest cell(s).


Very cool Thanks for that. It's probably too late to be of much use, but I'll give it a shot.

I'm 100% with you on wishing there was a real manual. Not just hundreds of unlisted FAQ pages (support articles)and videos. It might work better if there was a table of contents (for the FAQ) at a minimum.

Yeah... it's funking mental. I got their agreement to work with me to update the Victron instructions. I'll run it past the community folks for input before I submit. Gotta re-learn and actually understand what I did before I can do that... :)

Edit: I just wanted to add that the build looks great. Love it

Thanks!
 
Did you know there is an auto balance function?
It is called auto level. I haven't tested it to the depths that you would, but in theory it's only going to burn voltage off the highest cell(s).


FRICKEN SWEET!

I have no idea if I configured it correctly as I had to guess at the parameters, and there is no documentation, but it's automatically burning off any cell 0.01V or higher than the lowest cell.

Found this after the fact... Video:

 
Got to within .01 last night automagically. Decided against dwelling at 3.92. I'm pushing it to ~4.15V/cell to do a capacity test. The auto balance works nicely on the way up keeping deviation to 0.02V.
 
Got to within .01 last night automagically. Decided against dwelling at 3.92. I'm pushing it to ~4.15V/cell to do a capacity test. The auto balance works nicely on the way up keeping deviation to 0.02V.
I’m impressed….your quest for detail is rarely a human trait .. I suspect your actually from Vulcan…possibly Romulus ..but I’m leaning heavily toward Vulcan.?
 
Got to within .01 last night automagically. Decided against dwelling at 3.92. I'm pushing it to ~4.15V/cell to do a capacity test. The auto balance works nicely on the way up keeping deviation to 0.02V.
Nice, imagine how much quicker you'd have been there I'd the Batrium documentation was logically set up and available ?
 
I’m impressed….your quest for detail is rarely a human trait .. I suspect your actually from Vulcan…possibly Romulus ..but I’m leaning heavily toward Vulcan.?

I'm a train driver.

Nice, imagine how much quicker you'd have been there I'd the Batrium documentation was logically set up and available ?

About 3 minutes after I fixed the comm cable... ?

Had I not installed ESS, it would have been live the moment I sorted the cable since I already used DVCC for charge current limiting on the old lead acids. DVCC is actually what's enabling BMS control of the inverter/MPPT, not ESS. There isn't even a mention of DVCC in the installation. Next time I'm on-site, I'm going to remove ESS and confirm inverter control.

I suspect most using Batrium with Victron are EU folks using grid interaction for solar + peak shaving, and they already have ESS/DVCC properly installed, thus the instructions work fine for them.

With ESS running, the voltages for the inverter are wonky due to the defaults, and I'm concerned the inverter is going to shut down at around 51V.

Thanks again for the auto balance info. I slept sounder last night! :)
 
Depends on your needs and what level of communication you need.

Looks like that's a 13S NMC battery. The Batrium has no built-in cut-off like a FET BMS. You can certainly add relays as you like, or you can potentially interface with your hardware.

Right now, my only safeties are the BMS terminating charging by turning off the MPPT and discharge by turning off the inverter. There are no physical safeties
 
Right now, my only safeties are the BMS terminating charging by turning off the MPPT and discharge by turning off the inverter. There are no physical safeties
That seems risky!
Batrium is designed for an external contactor or shunt trip breaker.

I'm using this breaker:
 
That seems risky!
Batrium is designed for an external contactor or shunt trip breaker.

Yep. My usage is a small fraction of the battery capacity, and I'm keeping the cells between 3.50-3.92. I'm okay with this on a temporary basis.

I'm using this breaker:

Looks good. Thanks for that. The relay they mention on their videos is not available, so I've been looking for another. I'll probably get that one.

I also plan to implement hard shut-offs through the expansion board. In this fashion, the MPPT and inverters are shut off not via software but with their control relays.

With the breaker in place, that will be 3 levels of safety: software cut-offs, hardware cut-offs and the contactor.

As I was composing this email, Something weird has started happening. Cell #14 is jumping around. I've been discharging continuously for over 24 hours with the MPPT off conducting a discharge test, but #14 alternates between 3.73 and 3.86V:

1657484725092.png

I have had auto-level off since I wanted to test how they tracked with nothing more than the ~40-80mA of active balance. They've been extremely consistent at .01-0.2V deviation. When I saw the .14V deviation out of nowhere, I reactivated auto-level, and it immediately came back down to 3.72 from 3.86. #14 has the jumper wire on it, and I deliberately checked that all those wires were secure in the connector. I'll do it again next time I'm on site.

On the chance that the activation of bypass corrected the error, I've enabled it to activate at a 0.05V difference all the way down to 3.00V/0% SoC.

Here's what the discharge test looks like so far:
1657485332896.png

Up until I saw the voltage wonkiness, I was very pleased.
 
These were the final balance results:

1657517119491.png

Took 41-56Ah bled off from cells 1-2 and 4-14 to get all cells within 0.01V @ 3.92V (I subtracted the 2Ah I accidentally burned off #3 while manually chased the balance target before @400bird clued me in on the auto-balance feature.

Took around 4 days to make it happen.

I'm still discharging, and it's maintained .01-.02V range from the starting voltage of 4.05/4.07 to the current 3.63/3.64.

Pretty tickled.
 
Whelp, I found the lower leg...

1657606680440.png

Very pleased that the min/max stayed within 0.02V right up until that point. No balancing except for the ~40mA from the active balancers.

Current BMS:
1657606979895.png

I'll be shutting off the inverter before bed and load will drop from ~190W to about 22W. Should survive that for the night.

Shunt says I've used 370Ah. or about 19kWh.

VRM says I've used 17kWh, BUT that's only in AC consumption, so no inefficiencies or idle draw there.

I probably have a few % left down here and probably about 5-7% I didn't nab at the top by only charging to 4.05/4.07V, so the ~21kWh capacity/441Ah is probably pretty accurate. I think I'll drop it to 420Ah just for giggles and to leave a little margin.

MPPT has been turned on, so we'll start getting a little power around 6:30 in the morning. I've disabled the fridge AC circuit, so it won't start pulling until I tell it to. Once charging starts, that should end the 68 hour no-solar streak...

Inverter should turn back on once we're above 3.35V/cell and get the batteries in the RVs topped off as they'll be draining due to about 80W of lighting and the DSL modem.

All of which he says with unfounded faith.... :p
 
YAY! EVIDENCE BASED FAITH NOW!

Woot! Survived the night!

1657632863956.png

1657632784701.png

Solar kicked in a little earlier than I thought, and we're already pushing ~400W @ 6:30.

Enabled auto-level to balance on the way up.

Changed recovery voltage to 3.5 before inverter turns back on.

I've lowered the charge voltage to 3.8V/cell since our usage while off-site is very low, and I want to cycle the battery as little as possible and closer to 50% if possible. Partly cloudy today, so I might not make it back to the target voltage. Weather Underground says it's going to get cloudy with a 70% chance of rain peaking at 1pm.

Fridge happily not spoiling our food running on propane. Should have at least 12 more hours of 12V backup in the RVs to wait out the inverter, and I have a lot more confidence in the system/battery/BMS.

The initial deployment was incredibly frustrating and a little disappointing. Fortunately, I was able to get it all sorted remotely. If I weren't a stubborn SoB, I'd probably be curled up in the fetal position, sucking my thumb and weeping. MANY hours spent staring at the Batrium Toolkit, VRM and tweaking settings.

I'm going to claim a win here.
 
Last edited:
So cool to see you're system up/running!

I've been running Batrium w/longmons for a few years now - on 18650 (lithium-ion) packs. I find that once 'balanced', healthy packs don't need active balancing. After 9 months the max difference will increase by 20-30mv - at which point I use Auto-Level for a week to bring it back down.
1657636407470.png

Since I rarely reach 4.0v (max charge) but often cut-off at 3.54v (min) and because Auto-Level is more effective when the packs aren't charging/discharging - I tend to bottom-balance.

I also don't even try for 'exact balance'. The pack max difference will vary from 40mv to 70mv thru the full range (3.54v -> 4.0v) but stay the same every cycle. So I go with that as it doesn't make any operational difference. Here's this morning's snap of the state.
1657636060978.png

Ran across this recently - the limits of longmons voltage measurements.... and I wonder what the MM8 limits are. It's a bit eye opening as to the accuracy of what you see on the screen for balancing.... https://s3.amazonaws.com/helpscout..../BMon-and-LMon-LeafMon-Data-Sheet-Ver-2.0.pdf
1657637122410.png

In my trailer, I have a 14s88p bank w/Batrium. It's been over a year with modest active use (20days) and the rest just sitting. The initial balance of 10mv max difference hasn't changed at all with no balance:
1657637552081.png
 
Last edited:
Wow. That's very interesting. Hmmm.... I need to check the sense lead reported values against my Fluke...

I'd be contacting Batrium for that calibration feature!

BTW, @OffGridInTheCity was a big help to me via PMs. My struggle would have been substantially greater without that guidance.

THANK YOU!
 
Well, that spec sheet is eye opening.

I found a spec sheet for the MM8, it doesn't quite spell it out the same. Just "resolution down to 0.001 volts" nothing about accuracy.

I can't find the spec sheet for the WM5, but it appears that Batrium has revamped and sorted/organized their pages!

I'm not sure how helpful the updated information will be, but at first glance it seems like a huge improvement.
 
Well, that spec sheet is eye opening.

I found a spec sheet for the MM8, it doesn't quite spell it out the same. Just "resolution down to 0.001 volts" nothing about accuracy.

I'll be using one for the second 14S battery, so I'll have to check its accuracy as well.

I can't find the spec sheet for the WM5, but it appears that Batrium has revamped and sorted/organized their pages!

I'm not sure how helpful the updated information will be, but at first glance it seems like a huge improvement.

yeah, I just stumbled upon this while doing searches... not sure if it's actually better.
 
Back
Top