Another Deep Cycle test on BYDs

[Solarfun4jim]

I suppose if you were swapping out all 8 cells from another pack to replace the 8 weakest, within all your packs(going by the data you have collected thus far), could be fruitful, but just replacing those interconnectors with more substantial ones across every every terminal, makes this an unwelcome job. Not much 'terminal post' to work with, for making the re connections. I suppose it depends on how far out all 8 weakest cells are, as to the overall outcome/gain? Probably far more bother than it is worth. Maybe right at the start of a project, where you could justify it as 'prep' work. At this stage, yours do the job you want, so i guess if i was in your shoes, i'd be loathed to tackle that as well.

 

[Darkstar]

Cell 6 in pack 41
Cell 1 in pack 10

And here is the dismantling video (not at all an easy task)...

This fellow broke the packs into into individual cells and built a 12V battery out of the cells. He says the cells do not expand when you seperate them. The main issue I see is terminals for the cells once they are separated. There are more relevant videos on his youtube channel.

 

[batteryclearinghouse]

The cells expand slightly on the sides a little when you remove them but there are guys that have banded them back togther using the shipping straps. One actually borrowed the tool from someone at home depot.

The challenge is with the posts is that they are aluminum. tapping them to install a bolt/nut seemed like a good idea until I realised that aluminum and other metals don't always get along well over time.

As for years on these packs...there are several generations out there. You can kinda tell by the diffrences in the onboard BYD board.

 

[Picasso]

As for years on these packs...there are several generations out there. You can kinda tell by the diffrences in the onboard BYD board.

Yes my worst packs have the none balancing byd bms.

 

[jasonhc73]

I just discovered the ISDT Q8.
The specs indicate that it can balance at 1.5 amps. My Ant-BMS's only do 0.200 amp. I'm gonna get one and see what it can do.

The oddball thing, right now, I am charging to 54.4V (3.4V per cell) and the CellDiff is 0.005V, 0.003V, 0.004V and 0.007V - WITHOUT BALANCING, that's across 16cells (two packs). My MPP-Solar LV5048 is set to only 10 amps(x2), which weirdly only gives the entire 8 BYD's 16 amps.

I dunno what these freaks are doing.

 

[Maast]

The only 'cure' I can think of other than a complete teardown and rebuild I can think of is to parallel a booster cell with each of the weak cells. I've got 10 8ah headways coming in for my packs. It'd be extremely difficult to reconnect the cells after a teardown in a manner that'd allow you full amperage of the modules.

It might take one or two booster cells per weak cell to make up the difference. I figure than even if I overshoot a little no harm done - it'd just mean at the top end I'd have a cell that'd be a little low and at the bottom end I'd have a cell a little high.

 

[Darkstar]

The cells expand slightly on the sides a little when you remove them but there are guys that have banded them back togther using the shipping straps. One actually borrowed the tool from someone at home depot.

The challenge is with the posts is that they are aluminum. tapping them to install a bolt/nut seemed like a good idea until I realised that aluminum and other metals don't always get along well over time.

As for years on these packs...there are several generations out there. You can kinda tell by the diffrences in the onboard BYD board.

The bus bars are made out of what looks like aluminum foil. Many many layers of aluminum foil.

How about a CU/AL approved electrical terminal riveted to the bus bar for a cell terminal? Use aluminum rivets, attach your copper cables to the CU/AL terminal...

That and gobs of NO-Ox perhaps.

 

[BiduleOhm]

You can use stainless bolts, it should be OK with both Al and Cu.

 

[batteryclearinghouse]

I use two different types of hobby charges for my battery cycling. Not sure which I am in love with yet. They are both great units.

- EV-PEAK A9 1350W 25A High Power Lipo Charge
- ISDT T8 Battgo Lipo Battery Charger/Discharger

The EV ones feels better made, but the ISDT has been very impressive.

Both need to be supplied DC power. I have some Mean Well RSP-2000-24 Power Supplies which are very well made. I just don't trust the cheaper ones with how hard I push them.

I also have the Venom Pro Power 1350W DC Power Supplies... I like these because you can adjust the voltage on the fly. As you adjust the hobby charger amps you can see the voltage draw change on the venom unit. This lets you dial things in a little. It takes a little bit of experimenting, but I feel like I can match the demands of the charger with the voltage it's getting feed.

The hobby charges can take a wide range of voltages coming in and will convert it the charging requirements going out. Mismatched DC voltage generate inefficiencies and that means heat. I notice this most with the ISDT.

The fans on the units will go into a turbo/high mode when they start getting hot. I start of at max amps and work my way down until I hear the fans running at normal speed. The ISDT is rated to 30A on the charge, but I want to say I have them set at 26A charge when doing 24 volt charges.

One the upside with a hobby charges is that they have 1.5A balance on the cells during the charge. The cells just come out looking awesome.

Icharger is another hobby charger that looks good, but I have never owned one.

I know it's an investment... but I can't help think they would help solve a lot of drama.

 

[jasonhc73]

Now onto the other end of the spectrum, charging.

I tried to charge to 54.4V(3.4V per cell).
Now a 3rd pack is going bananas.
One cell is zooming up and out of whack.
I have my packs number 10/11, 20/21, 30/31 and 40/41, no real meaning, just a way to number them.

Cell 6 in pack 41 - dive bombs low voltage
Cell 1 in pack 10 - dive bombs low voltage
Cell 8 in pack 40 - shoots up to overvoltage

ANT-BMS saved the day every time. I have the CellHighProtect at 3.59V. As soon as that 3.59 reached the BMS disconnects and starts its active balancing. On a "220AH" cell, 0.200 amp doesn't do much of anything, unfortunately.

I have tried to seize the cells up by limiting their cell difference through the BMS by not letting them get more than 0.100V cell difference, but they shut down the entire pack when that happens. If the BMS could do a 1 amp cell shuffle it might work, but 1/5th an amp against such a large cell just isn't enough.

6 hours later I gave up on trying to get to 54.4 and have it set to float at 53.3V (3.331V). The offending cell is still at 3.347. It took about an hour to self-discharge to 53.3V on the other 3 sets.

40/41 Low High Cell Diff 3.309V 3.348V 0.039V
30/31 Low High Cell Diff 3.325V 3.334V 0.009V
20/21 Low High Cell Diff 3.329V 3.336V 0.007V
10/11 Low High Cell Diff 3.323V 3.338V 0.015V


To sum up, I now know I have 3 bummer cells robbing me of the 32~36 kWh I bought. Even limiting to 80% of their rating I'm barely getting 1/3rd of the usable capacity.

I'm going to give another shot at this with tighter setup on the BMS, and isolated 16S packs.
I have the Cell unit voltage Gap to 0.050V now on all 8 packs. We'll see how much of a mistake this is tomorrow.

 

[jasonhc73]

Here it is Tuesday evening.
Recharging the packs after a nice pretty Sun-filled Kansas day.
Cell Diffs:
40/41 0.004V blinking with 0.003V - 6.6 amps charging
30/31 0.004V blinking with 0.003V - 2.9 amps charging
20/21 0.004V blinking with 0.003V - 3.5 amps charging
10/11 0.010V blinking with 0.009V - 5.4 amps charging

I have the BMS set to cell balance with only 0.100 amp. The Balance Mosfetts did not seem to care with 0.200 or 0.100.

Total charging is set at 20 amps but the watchtower only shows 16 amps from the LV5048s.
The amps shown on the BMS are basically a guess. You set the current range in the app, and it just shows whatever it thinks the amps are. You have to have an ammeter to be 100% accurate, I matched the amps to the setting with a DC hall sensor.

After doing a deep cycle test, with the BMS stoping it because of two cells dive bombing, it has taking 4 days to get all 64 cells back in balance.
Across all 64 cells, the highest cell is 3.314(pack 10 cell 1) and the lowest 3.295 (pack 30 and cell 1,2,3,,5) (Pack 30 the lowest is 3.296 and the highest is 3.299 - 0.003V across the entire pack).

I've had the Active Balancing going for about 24 hours now, and it's made a huge difference in the cell gap. It may be resistive/passive but none the less, it is working.

 

[jasonhc73]

A week later and starting another deep dive test.
This time active balancing at 0.200 amps.
The same dunderhead cells are already acting up.

52.1 40/41 Cell Diff 0.030V
52.2 30/31 Cell Diff 0.003V
52.2 20/21 Cell Diff 0.004V
52.1 10/11 Cell Diff 0.030V

Cell 6 in pack 41 - dive bombs low voltage
Cell 1 in pack 11 - dive bombs low voltage

Now with the cell balancing, I think I should get at least some better performance.

This time, I have solar panels. This test is not exactly the same as the pure battery test of the last week.

This test is an off-grid test supplemented by 2.4 kWp of solar, with a forecast of cloudy tomorrow .

With sun aided panels today, I started at 53.1V across the board, in 100% battery mode when I turned off the grid.

Net meter / sent to house/received from the house





I did not try to charge higher than 53.2 this time to see if there are other cells that are runaway cells like last week. I know there is at least one cell in pack 40 that does this, and I suspect another, but I don't know where yet.

Like any good experiment, one thing at a time.

Right now the packs see 5.0 amps discharge. New, these were 220 AH cells, they are being subjected to 0.0057 C rate.

 

[jasonhc73]

The next morning.
Whereas last week's test ended at this time this week's test is continuing.
4 packs are shut down due to voltage spread, thanks for the BMS.
Now that BOB is up and wrenching on the voltage spread. It's clear 2 cells are crap. Now to find the top end junkers.
The sun is charging nicely.

18 kWh sent to the house from the batteries (and solar).

 

[nwillitts]

can you give links for those meters and mounts?

 

[Solarfun4jim]

The next morning.
Whereas last week's test ended at this time this week's test is continuing.
4 packs are shut down due to voltage spread, thanks for the BMS.
Now that BOB is up and wrenching on the voltage spread. It's clear 2 cells are crap. Now to find the top end junkers.
The sun is charging nicely.

18 kWh sent to the house from the batteries (and solar).
View attachment 7862View attachment 7863View attachment 7864
View attachment 7865

Percy verence, cousin of Patience Tested....lol. You will get there Jason, and it will be worth it in the end, no matter how you remedy the challenge. Love getting this feedback on real life situation.

 

[jasonhc73]

can you give links for those meters and mounts?

The meter on eBay.(link)
The Meter Socket from Missouri Wind and Solar (link)

Percy verence, cousin of Patience Tested....lol. You will get there Jason, and it will be worth it in the end, no matter how you remedy the challenge. Love getting this feedback on real life situation.

I just rearranged the 2x4 solar panels into 3x3. Through the DC cable the amps down (~24 to ~18), the volts up (~60 to ~90). And the total watts up (~1400 to 1800). Overall I added 1 panel, from 2396 kWp to 2655 kWp. Very satisfying to see it work out. And I didn't need any more wires or connects, actually it used two less Y connects.

 

[Solarfun4jim]

Jason, i saw this and thought of your pack......a tad over expensive, but 10A balancing.
Noted this though "The balance current is set independently in the range of 0.1~10A, independent of the cell voltage difference;"

2~24S 10A Bluetooth Active Equalizer Balancer 4 Li-ion Lipo Lifepo4 LTO Battery | eBay

Find many great new & used options and get the best deals for 2~24S 10A Bluetooth Active Equalizer Balancer 4 Li-ion Lipo Lifepo4 LTO Battery at the best online prices at eBay! Free shipping for many products!

www.ebay.com

Too expensive to put these on every pack....but there might be others come to market?

Even the 5A version is $363 :-(

2~24S 5A Bluetooth Active Equalizer Balancer 4 Li-ion Lipo Lifepo4 LTO Battery | eBay

Find many great new & used options and get the best deals for 2~24S 5A Bluetooth Active Equalizer Balancer 4 Li-ion Lipo Lifepo4 LTO Battery at the best online prices at eBay! Free shipping for many products!

www.ebay.com

 

[jasonhc73]

Watching the batteries get some solar.
Now they have solar to charge throughout the day. Two packs D ratings should never have been sent out.

Time to find the top-end ones. I know of one already that is junk at the top end, hopefully only one.

3 of 8 known crap packs. ?

(You might be able to comment on the screenshots if you care to)

Battery test 21-22 Feb - Google Drive

BYD packs underperforming and showing off.

drive.google.com

 

[GVSolar]

Sorry to hear that Jason. Are you getting replacements - or a credit for your capacity loss?

 

[Darkstar]

Sorry to hear that Jason. Are you getting replacements - or a credit for your capacity loss?

Sounds like so far no one has responded to Jason's inquiries.