Lishen 272AH thread?

[michael d]

I am trying to find clear info on the bus bar threads for these lishen 272Ah cells. I have searched the forum ans not found it. I was hoping it would be M6, but when I asked Xuba they responded M4. Thank you

I have Lishen 280Ah cells likely the exact same as 272Ah ones that are supposed to do 285Ah. they have M6 set screws that use a 3mm hex key to tighten into place, they came with bus bars and serrated flange nuts also 16mm x M6. I am almost positive none are M4 as M6 is only like 1/4 inch diameter anyway. they are small. the set screws are also referred to as grub screws. mine set screws were a little short for thicker lugs so I ordered some longer stainless steel 25mm and 30mm M6 grub screws.

 

[MadMax03]

I am trying to find clear info on the bus bar threads for these lishen 272Ah cells. I have searched the forum ans not found it. I was hoping it would be M6, but when I asked Xuba they responded M4. Thank you

They're all M6.

 

[stobbie]

I'm at work now, I'll try to share the STL files this evening (europian timezone)

 

[Pyke13]

I ordered 8x 272 ah cells on the 4th of December, still waiting. I’m in the uk and probably got a while to wait yet.
Getting slightly worried as now the alibaba app keeps crashing every time I try to log on to it. If I go thru there eBay add it works fine, when I try to go thru the app too check and see if anything’s changed on the shipping it just crashes and returns me to my home screen. Probably just my paranoia

 

[JohanB]

They're all M6.

Thank you! It seems like Selina at Xuba mixed it up. She now also confirmed it is M6 after me questioning the previous info.

 

[JohanB]

I have Lishen 280Ah cells likely the exact same as 272Ah ones that are supposed to do 285Ah. they have M6 set screws that use a 3mm hex key to tighten into place, they came with bus bars and serrated flange nuts also 16mm x M6. I am almost positive none are M4 as M6 is only like 1/4 inch diameter anyway. they are small. the set screws are also referred to as grub screws. mine set screws were a little short for thicker lugs so I ordered some longer stainless steel 25mm and 30mm M6 grub screws.

Thank you for the update, very useful!

 

[shorawitz]

If you use silver flashed copper, make sure the silver is polished before assembly, The silver tarnish is a lot less conductive than copper or copper Oxide. Presumably, the silver will not tarnish where it is pressed against the terminals (and/or itself), but I can't be sure of that.

I generally just use the rectangular copper bar from online metals.... It is a lot cheaper.

On the inner 'boxed' busbars, Put the serial bars down first. They are likely to be carying more of the current and therefor should be directly on the cell terminal pad. (The parallel bars will only carry enough current to keep the voltage equalized amoung the parallel cells).
View attachment 30064
If Possible, I would use heaver duty bus bars on the long 8 hole busbars. This will help prevent voltage changes between the cells. To explain why, the following diagram models the resistance of the bus-bar, the current and voltage drops along the bus bar. (This modeling is a first order approximation. A better model would seperate the copper resistance from the connection resistance to each terminal). As you can see, in this model, the voltage drop across the whole busbar is 16 times the voltage drop between terminal 1 and 2.
View attachment 30041
The question then becomes: How big does this voltage actually get?

Lets assume that with the connections to the terminal pad and the copper you have .2 mOhm (.0002 ohm) resistance between any two adjacent terminals. Now lets assume you are pulling 100 amp total from the bank. The 100 Amps battery draw would show up between the positive and negative, so in this model 4i=100A. That means i is 25A. Consequently v is .0002ohm x 25A = .005V and the total drop from one end of the bar to the other is 16x.0025=.08V. (The voltage will increase/decrease linearly with current)

Is 0.2mOhm higher than reality? I don't know but .2mOhm is certainly not a large number.
Is .08V a killer? I don't know, but it is certainly getting in the range to think about.

The bottom line is that anything you can do to reduce the resistance on the long bus-bars is important in order to keep the voltage at the cells as similar as possible. Having a single bus-bar instead of stacking 2-hole factory bus-bars is a huge improvement (Far fewer connections). Making the single bus-bar extra thick/wide will help as well.

When you set this up, it would be interesting to put a sensitive voltmeter across the length of the long busbar and see what drop you get under a heavy load.

I was looking over Will's site again, and if this is to be the preferred bus configuration, I think Will should update his diagram in the "How To Configure Cells In A Large Battery Bank". His site says to connect in Parallel first, then add the Serial connections

 

[FilterGuy]

I was looking over Will's site again, and if this is to be the preferred bus configuration, I think Will should update his diagram in the "How To Configure Cells In A Large Battery Bank". His site says to connect in Parallel first, then add the Serial connections

Interesting. Could you point me to the page you are referring to?

 

[shorawitz]

Interesting. Could you point me to the page you are referring to?

Design Your Own LiFePO4 Solar Power System

This is a complete system solution complete with: BMS Protection Features (over/under voltage/over-current/balancing) Low Temperature Cut-off Programmable Charge/Discharge Bandwidth Settings Matched...

www.mobile-solarpower.com

About 1/3 of the way down the page.

 

[FilterGuy]

I have now sanded all bus bars en battery posts, now the biggest voltage drop over the longest bus bar is 7mv at 50A
Do you think that will be low enough?

Is that on the 4 hole bar? If so, that seems pretty good. That would translate to about 3.5 (.0035m) between parallel cells. Is that with fabricated long bars or stacked factory bars?

 

[FilterGuy]

Is that on the 4 hole bar? If so, that seems pretty good. That would translate to about 3.5 (.0035m) between parallel cells. Is that with fabricated long bars or stacked factory bars?

I just noticed the reading was at 50A. At 140A (inverter maxed out) the drop between parallel cells will be around 9.8mV.

Are you still thinking about doing an 4p configuration? If so, that projects out to 58.8mv across the 4 parallel cells with the really long 8 hole bus-bars. I would definitly try to reduce that....but if your measurment was with the factory busbars.... the single peice bars should make a huge improvement.

 

[FilterGuy]

Design Your Own LiFePO4 Solar Power System

This is a complete system solution complete with: BMS Protection Features (over/under voltage/over-current/balancing) Low Temperature Cut-off Programmable Charge/Discharge Bandwidth Settings Matched...

www.mobile-solarpower.com

About 1/3 of the way down the page.

Lets see if we can summon the wizard for comment: @Will Prowse.

As I said in the earlier post, I recommend the serial bars first because they will almost certainly be taking more of the current. However, I do not know how big of a difference it will make. I figure that even if it is a small difference, why not do it?

 

[michael d]

for my 2p8s battery, I connected the 2 cells in parallel 1st as Dacian recommended. then I put each set of 2 cells (now 560 Ah each) in a series configuration. this is a 24-volt battery bank, it is working great at the moment with the Electrodacus SBMS0 And the Electrodacus DSSR20. the cell balancing is displayed in 3 decimal points. I like that a lot. no guessing what is going on. I have the wifi version so I can see it on my android phone. plan to set it up on a larger screen laptop also. I am going to build another 24-volt battery bank soon. this is hooked up to a 6000-watt low-frequency PSW split phase inverter. have not got anywhere close to using 6000 watts at once yet.
I conspicuously labeled every positive terminal on each of the 16 280Ah cells with red tape and a Dymo label positive as I accidentally arc-ed a couple of cells. scared me a bit as it acted like an arc welder and cut halfway thru one stud and ruined another stud. ordered new longer ss studs 25 and 30 mm long x M6. quickly used a plastic caulk gun tip to get it apart, put on my safety glasses and gloves after that (a little late) - should of done safety 1st!
excuse me but I have no plans of putting 4 in parallel which would be 1120 amps.
280Ah x 3.2V x 16cells = 14336 wh or a 14.336 kWh. big enough for one battery for the moment.
i was running a small 10amp load thru the 6000-watt LF PSW split-phase inverter and pulling 16 amps into the 24-volt LiFePO4 battery bank with only 2 250 watt panels hooked in parallel thru one DSSR20. while scooping some snow

 

[shorawitz]

Lets see if we can summon the wizard for comment: @Will Prowse.

As I said in the earlier post, I recommend the serial bars first because they will almost certainly be taking more of the current. However, I do not know how big of a difference it will make. I figure that even if it is a small difference, why not do it?

I thought the serial first made sense too. Would be interesting if we had empirical data to compare.

 

[Ampster]

I think serial first makes sense because that is where most of the current is flowing. Long term I am going to make my own and eliminate that question by using a single busbar with six holes to create a 3P pack. Each bar will parallel 3 negatives with 3 positives and it will stagger up that way. Since it will be a 3P16S pack I will have to turn the corner if i want a shorter pack with several rows. I still may be able to turn the corner with a single wide busbar with 3 holes on each side. I prefer not having to worry about resistance between stacked interconnects.

 

[FilterGuy]

Since it will be a 3P16S pack I will have to turn the corner if i want a shorter pack with several rows. I still may be able to turn the corner with a single wide busbar with 3 holes on each side. I prefer not having to worry about resistance between stacked interconnects.

Yup.... This is a bigger potential issue than the question about serial or parallel first on the 'boxed' connections. There is so much copper at the boxed connections, either way is going to produce a good result. However, The long bars will have an additive voltage change so there is more opportunity for an issue. I am pretty sure that if you fabricate a single heavy duty bar that spans the 6 holes, the voltage difference will be small enough to not be a concern.

I'll be working on a 2P8S pack soon and it will have a 4-hole bar that 'turns the corner'. I'll take some measurements.

There are 2 reasons to worry about any of this.

1) Overall energy loss.
Anyplace there is resistance is a place that is burning up your energy. However, I don't think this is a big enough issue to be concerned about. There are probably other energy wasting parts of the system that would yeild bigger pay-offs for the same effort.

2) Uneven wear on the cells.
If there are voltage differences between parallel cells, then the cells are charging and discharging at different rates. With the charge and discharge curve being so flat, small voltage differences can make large differences in the charge and discharge of a cell. However, with the LiFePO4 lifetime cycle count being so high, it will be a long time before there would be any noticeable difference.

Another mitigating factor is that for the vast majority of the time, our systems tend to be running fairly low current. That means the voltage drop will be very low as well. It is when we draw high current that there is a potential issue.

So.... Is it worth loosing sleep over? Absolutely not! However, I still believe in doing what you can to optimize the characteristics of the system so I'll put the serial bars down first and build heavy duty bars for the long 'turn the corner' connections.

 

[michael d]

I thought the serial first made sense too. Would be interesting if we had empirical data to compare.

Yup.... This is a bigger potential issue than the question about serial or parallel first on the 'boxed' connections. There is so much copper at the boxed connections, either way is going to produce a good result. However, The long bars will have an additive voltage change so there is more opportunity for an issue. I am pretty sure that if you fabricate a single heavy duty bar that spans the 6 holes, the voltage difference will be small enough to not be a concern.

I'll be working on a 2P8S pack soon and it will have a 4-hole bar that 'turns the corner'. I'll take some measurements.

There are 2 reasons to worry about any of this.

1) Overall energy loss.
Anyplace there is resistance is a place that is burning up your energy. However, I don't think this is a big enough issue to be concerned about. There are probably other energy wasting parts of the system that would yeild bigger pay-offs for the same effort.

2) Uneven wear on the cells.
If there are voltage differences between parallel cells, then the cells are charging and discharging at different rates. With the charge and discharge curve being so flat, small voltage differences can make large differences in the charge and discharge of a cell. However, with the LiFePO4 lifetime cycle count being so high, it will be a long time before there would be any noticeable difference.

Another mitigating factor is that for the vast majority of the time, our systems tend to be running fairly low current. That means the voltage drop will be very low as well. It is when we draw high current that there is a potential issue.

So.... Is it worth loosing sleep over? Absolutely not! However, I still believe in doing what you can to optimize the characteristics of the system so I'll put the serial bars down first and build heavy duty bars for the long 'turn the corner' connections.

I used 2 6guage cables to turn the corner on my 2p8s 24-volt battery bank 272Ah Lishen cells. 16 cells = 13,926.4Wh. will be making better busbars soon. right now using the ones provided by the vendor.

 

[michael d]

Lets see if we can summon the wizard for comment: @Will Prowse.

As I said in the earlier post, I recommend the serial bars first because they will almost certainly be taking more of the current. However, I do not know how big of a difference it will make. I figure that even if it is a small difference, why not do it?

I am using the Electrodacus SBMSo0 with the Electrodacus DSSR20 and Electrodacus DECT16. for my 2P8S 24Volt Lishen 272AhDIY battery bank. if you follow his direction (Dacian) for the Electrodacus components. he has repeated numerous times to parallel 1st. that is what I did and all is working great. I don't plan to put 3 or 4 in parallel. 2 in parallel at 272 Ah each was 544Ah and the battery has 13926.4Wh capacity. enough to start.

 

[shorawitz]

Lets see if we can summon the wizard for comment: @Will Prowse.

As I said in the earlier post, I recommend the serial bars first because they will almost certainly be taking more of the current. However, I do not know how big of a difference it will make. I figure that even if it is a small difference, why not do it?

Here's another reference for the "serial first argument" I tend to rely upon - Adam Welch video (4:51):

He basically uses the same reasoning.

 

[FilterGuy]

Here's another reference for the "serial first argument" I tend to rely upon - Adam Welch video (4:51):

He basically uses the same reasoning.

He uses the exact same argument I use.

I also note that he does not fully 'box' the bus bars. He only does one parallel bar. I suspect this is fine, but I go ahead and put the parallel on both sides just to be sure any differences in voltage drop accross the serial bars is 'corrected' on the other side..... yup..its overkill .