My Offgrid House & My Programable Solar BMS for any type of Lithium battery.

[electrodacus]

Not many wish to live in that style!

Additionally, where I live electricity is the cheapest source for heat also.

I will also use electricity for heat but I do not now.
It will be cheep from solar panels directly with no battery involved but you need a large thermal mass like I have.
Else I think electricity is not that cheep I guess there is no natural gas in the area also guess Kwh is over 10 cent.
I do not need more power for anything there are just me and my wife both using computers more than 8 h/day sometimes it runs over night to render videos or 3D graphic.
LED for light cooking with electricity only that is the largest use and maybe I can have a larger fridge than the one I use now but it seems just enough for now.
Not sure what else I will need that uses power. My oscilloscope just 4W and the bench multimeter just 500mW

[Mike90250]

...Can you be more specific what is not safe with my controller ? ...

The terminal connections. I charge at 40 amps, and that will be 60A when I get my 2nd array up.

If I had 2 controllers, 20a each, that maxes out the terminal limit. And was mentioned earlier, wire
size is an issue, it won't take a large wire. Large wires, large terminals = low heat. Small wires, small terminals and couple years down the road, the terminals have turned black from the heat, maybe even coming loose off the PCB inside. And I'm not sure 100mA is going to be able to balance a cell when there is 20A going through the string, remember, the batteries were low, and the sun is going down, I don't want to waste any solar watts just because one cell is waiting to balance out. I'd think a 10A bypass would be needed at least. (But i'm not a Li--- user and don't know all the kinks about it) but while i have sun, I want to stuff as much as I can, into the batteries. )

~~~~~

| Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV

[electrodacus]

The terminal connections. I charge at 40 amps, and that will be 60A when I get my 2nd array up.

If I had 2 controllers, 20a each, that maxes out the terminal limit. And was mentioned earlier, wire
size is an issue, it won't take a large wire. Large wires, large terminals = low heat. Small wires, small terminals and couple years down the road, the terminals have turned black from the heat, maybe even coming loose off the PCB inside. And I'm not sure 100mA is going to be able to balance a cell when there is 20A going through the string, remember, the batteries were low, and the sun is going down, I don't want to waste any solar watts just because one cell is waiting to balance out. I'd think a 10A bypass would be needed at least. (But i'm not a Li--- user and don't know all the kinks about it) but while i have sun, I want to stuff as much as I can, into the batteries. )

~~~~~

| Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV

Those terminals that I use are a good brand with low contact resistance and used in the specified limits 20A/300V.
They can accept max AWG12 but there are usually short wires to battery and inverter and or PV disconnect switch no more than one or two feet.
Since you use 4 of this wires with 4 separate contacts for 80A to the inverter or battery and this is just positive the negative you can use any wire size you want. The fact that are 4 and not one makes for larger cooling surface area than a single equivalent conductor.
20A is the max that I can use since it needs to be soldered on the PCB. This is also a compact unit that was also one of the targets to get it as small as possible of course with no compromise.
Your problem with the black connectors and heat is probably from improper torque or not good quality connectors.
What brand where those 20A controllers?
100mA is more than adequate even for large cells few hundred Ah.
I just recently made a manual balance on my large cells after more than two months and was just a few hundred mAh difference between the highest and the lowest cell. I do manual balance until l I have my BMS working. I only have one working sample and I need that for development.
Usually cells get more imbalanced if used in high discharge rate applications not the case with solar slow charging under 0.3C and low discharge rate at most 1C in my case usually discharge is under 0.5C the largest load I have is an 1400W oven and I usually use that during the day when the largest part comes form the PV panels directly.
If one of the cells got to set voltage 3.6V or 4.2V depending on type the others can not be far at most 0.5% or so and that is not much additional capacity since your battery is already 99.5% full.
Imbalance of the cells occurs because of the small difference in internal resistance.
Also consider that 7 cells can require balance at one time that is the worst case and say they are Lithium-ion so 4.2V max 7x4.2V =29V x 100mA = 2.94W of heat dissipated that is the most I can afford in that small space without additional cooling the power resistors have thermal contact with the aluminium PCB on the bottom.
Everything is calculate to work in worst case scenario I have this bug from my last job designing airbag electronic control units.

[Sunny Solar]

Electrodacus. Im far from convinced that those terminals are a good choice. You even agree that #12 wire is about the limit that's not good for 20a even at 2 feet length.. And that would never be a practical length for most users... And not many would like the idea of using multiple small cables to connect to the battery/ies or inverter..??Its not just the quality of the screw connector. those terminal blocks have only small connector to solder at the back. I have seen them melt the solder because the wire has not made perfect contact and overheating follows... It could never be offered for sale like that. Its just to risky for the average user.

And after having LIFEPO4 experience for last 4 yrs and seeing experiments done on them every day. Where I work we have almost every type and manufacturer of those cells and . have seen all the differences between them under many conditions.. Its been unanimous here that 100ma balancing is only good for below 20ahr cells. And really only 10ahr for best results. Above that you need 2a balancing. The time factor will be just to great otherwise.. This has been proved beyond all doubt..


Like I said its one of the big reasons that small cells are used by car manufacturers as its just so hard to balance many big cells... It appears its just so much easier to manufacture close tolerance small cells. .And make a relatively simple BMS..

Another possible problem I see with your charger . its very compact and im assuming you are using 100a dc solid state relays.? Those things need heatsink if passing 80a. ?? I cant see that being possible in such a small case.??

[electrodacus]

Electrodacus. Im far from convinced that those terminals are a good choice. You even agree that #12 wire is about the limit that's not good for 20a even at 2 feet length.. And that would never be a practical length for most users... And not many would like the idea of using multiple small cables to connect to the battery/ies or inverter..??Its not just the quality of the screw connector. those terminal blocks have only small connector to solder at the back. I have seen them melt the solder because the wire has not made perfect contact and overheating follows... It could never be offered for sale like that. Its just to risky for the average user.

And after having LIFEPO4 experience for last 4 yrs and seeing experiments done on them every day. Where I work we have almost every type and manufacturer of those cells and . have seen all the differences between them under many conditions.. Its been unanimous here that 100ma balancing is only good for below 20ahr cells. And really only 10ahr for best results. Above that you need 2a balancing. The time factor will be just to great otherwise.. This has been proved beyond all doubt..


Like I said its one of the big reasons that small cells are used by car manufacturers as its just so hard to balance many big cells... It appears its just so much easier to manufacture close tolerance small cells. .And make a relatively simple BMS..

Another possible problem I see with your charger . its very compact and im assuming you are using 100a dc solid state relays.? Those things need heatsink if passing 80a. ?? I cant see that being possible in such a small case.??

First thanks for the feedback.
AWG 12 is good for 25A in open air single wire as is the case here and even as part of a 3 wire cable can do 20A
2 feet will be actually practical for any user that wants to use this at max capabilities since like I mentioned battery need to be as close or closer. The cell measurement and balancing cable also need to be short so battery can not be far.
Then inverter is also a good idea to be as close especially if is a 2000W inverter.
As for the solar PV they are further but you need to have a disconnect switch placed not more than 2 feet from the charge controller
Do you disagree with any of the above?
If wire dose not make good contact because user did not applied enough torque that is a problem for any connector no matter how big or oversized it is.
I have an advantage since I use 4 terminal chances are than the other 3 are properly connected good so the one that is not will not heat up.

As for the 100mA balancing that is good for any battery size especially that this controller is not designed for high discharge rate and high DOD where cell imbalance can be much more pronounced.
I have a large 100Ah battery used to power my house almost no imbalance in a few months of daily use and also have a 4 cell 3Ah battery used with over 10C discharge with a cordless drill that needs rebalancing every charge cycle.

I do not use solid state relay I use multiple MOSFET's and got the lowest RDSon available cost optimisation is not so important for small number of units.
I made the calculations before in this thread but here it is again. Typical RDSon for this is 1.7mohm I have 4 of them in parallel for the load that is 0.425mohm then in series with this is a 0.25mohm shunt for current measurement, overcurrent protection and short circuit protection. in total they are 0.675mohm and mounted on an aluminium PCB if you seen the design.
At 80A voltage drop will be 54mV (80A x 0.675mohm) that is a max of 4.3W that can easily be dissipated on that thick aluminium PCB on the bottom of my unit not to mention the unit is designed so that it can be mounted on an additional heatsink if required but is not necessary in normal operating condition.
There is of course also internal temperature measurement and programmable protection for over temperature as well as external over and under temperature protection for the battery.

As you say you work with LiFePO4. What is the best brand that you have seen in your tests ?

[russ]

As you say you work with LiFePO4 what is the best brand that you have seen in your tests ?

John never provides the name of companies - it is against the lab rules I believe.

[electrodacus]

John never provides the name of companies - it is against the lab rules I believe.

That is fine it makes sense. I already have my batteries but I wanted to know if others ask for recommendations I did not studied the battery marked in the last year maybe there are better batteries available.

[Sunking]

First thanks for the feedback.
AWG 12 is good for 25A in open air single wire as is the case here and even as part of a 3 wire cable can do 20A

Well that is correct but does not mean anything. On low voltage systems you design for voltage drop and NEC 310 tables go out with the trash.

Cable from panels to controller is designed at 2 to 3 % loss, and controller to battery at 1 to 2%. 12 AWG has 1.6 Oms/Kft. So with 25 amps flowing you loose 1 volt for every 12.5 feet one-way loop distance. 1 volt at:

12 volt = 8%
24 volt = 4%
48 volt = 2%

So 12 AWG is pretty much useless unless you are talking low current and short distances.

[electrodacus]

Well that is correct but does not mean anything. On low voltage systems you design for voltage drop and NEC 310 tables go out with the trash.

Cable from panels to controller is designed at 2 to 3 % loss, and controller to battery at 1 to 2%. 12 AWG has 1.6 Oms/Kft. So with 25 amps flowing you loose 1 volt for every 12.5 feet one-way loop distance. 1 volt at:

12 volt = 8%
24 volt = 4%
48 volt = 2%

So 12 AWG is pretty much useless unless you are talking low current and short distances.

You only design for voltage drop if you do power transmission this is not the case here where the longest cable is two feet 60cm this is called chassis wiring from that point like from the PV disconnect switch you need to calculate the voltage drop since you normally have much higher distance.
But let also calculate voltage drop for a 60cm (2ft) of 12 AWG wire at 20A the resistance /ft is 1.588mohm so for two feet 3.176
0.003176ohm x 20A = 0.06352V x 20A = 1.27W will be dissipated on the two feet cable (I can calculate also the delta of temperature increase above ambient but I'm sure is fine from experience) Voltage drop as you can see is not significant at 24V it represents 0.26%
The AWG 12 can handle 25A as I mentioned in free air single conductor with the worst 60C thermal insulation it can do 30A with higher quality 90C thermal insulation and as much as 40A just copper no electrical insulation. Well within the 20A max for this connectors that are UL listed and designed to support 20A continues.

[Sunking]

You just do not get it. 12 AWG has no application in battery wiring unless you are dealing with toys. Terminals need to be able to handle up to 2 AWG.

[electrodacus]

You just do not get it. 12 AWG has no application in battery wiring unless you are dealing with toys. Terminals need to be able to handle up to 2 AWG.

This dose not seems to be a constructive criticism.
Do you disagree with my calculations? are there any errors ?
I think I demonstrated that AWG 12 can more than easily handle 20A (even AWG 14 will do that at the limit)
This was designed specific for my needs and not as a product for everyone.
It has the right type of connectors for the specified current limit.
The 4x AWG 12 are equivalent with an AWG 6 and will support the total of 80A
This is maybe a "toy" for you if you require hundred of Amperes.In that case you will need a different product.

[Naptown]

This dose not seems to be a constructive criticism.
Do you disagree with my calculations? are there any errors ?
I think I demonstrated that AWG 12 can more than easily handle 20A (even AWG 14 will do that at the limit)
This was designed specific for my needs and not as a product for everyone.
It has the right type of connectors for the specified current limit.
The 4x AWG 12 are equivalent with an AWG 6 and will support the total of 80A
This is maybe a "toy" for you if you require hundred of Amperes.In that case you will need a different product.

Do you live in the US?

[electrodacus]

Do you live in the US?

No. I live in Canada. Why ?
I took my electrical engineering degree and worked in Europe just recently moved in Canada.

[Mike90250]

This dose not seems to be a constructive criticism.
Do you disagree with my calculations? are there any errors ?
I think I demonstrated that AWG 12 can more than easily handle 20A (even AWG 14 will do that at the limit)
This was designed specific for my needs and not as a product for everyone.
It has the right type of connectors for the specified current limit.
The 4x AWG 12 are equivalent with an AWG 6 and will support the total of 80A
This is maybe a "toy" for you if you require hundred of Amperes.In that case you will need a different product.

This IS constructive criticism. We are saying, that while the design works on paper, and for small batteries and for small inverters that won't pull much power, it's OK. But once a person trys to run a house or something larger, it will work for a time, but long term, it's going to haunt you. There WILL be failures and returns, because of the meager 20A terminal. Wires in parallel are a bad idea, they seldom share the current, and will sequentially fail, in the long term.
In the USA states, the lawyer fees from the first court case will bankrupt you.

Are you using 3 or 4 oz copper in the PCB to handle the current ? Even the high current paths in the PCB will turn into fuses and mess the inside of the module all up.

[Naptown]

No. I live in Canada. Why ?
I took my electrical engineering degree and worked in Europe just recently moved in Canada.

The Canadian standards are similar to the US
It would be against the NEC to parallel #12 conductors to acheive an 80 amp capacity.
This is only legal here for over 1/0 conductors. I am sure Sunking will correct me if I am wrong.