Balance at the cell low voltage limit

Dan,

In your diagram, you are not using the buddy system right? Which would mean that you have two strings at 48V in parallel. If one string has a dead cell then the other 48V string would try to charge the weak string with a huge amount of AMPs in theory right even if the converters or charger was off. That is certain death for the weak string. Is my logic correct?

I think that is how I resulted in 14 weaken cells after my last incident with 2 bad cells. In my case I had 4 strings of 2kwh in parallel. The other three string pretty much overcharged the weak string to the point where the cells were gasing popping even after I turned the system off.

 

I see it now. That is a lot of wiring...

 

Just my two cents:

Balancing at the bottom will increase the number of deep-cycles for your battery pack thus decreasing over-all lifespan (better to keep batteries topped off than bottomed out). What you really need is a reliable balancing system that works 100% of the time, but clearly you know that already...Again, as you pointed out earlier since your batteries will have different capacities (presumably the new ones will have a higher capacity than the old ones) it will be impossible to keep them in sync without effective balancing--whichever way you end up choosing to do it.

Andrew

 

Here is my take on the whole thing from what I have read and from what I heard from Jack Rickard.

You can either top balance or bottom balance its totally up to you. The whole balancing thing is to surrender to the weakest link in your chain.

If you want to top balance then you will have to monitor for your weakest cell on discharge. When your weakest cell reaches its lowest discharge point then you want to disconnect the whole pack. The disadvantage with this design is you cannot use total pack voltage for your SOC. If you do you risk killing your weak cells. Now if you have a per cell LVC then top balancing can work because when that one cell reaches the cutoff voltage the whole pack shuts down.

If you want to bottom balance then you have to stop charging when your weakest cell reaches its full state. You could almost use the total pack voltage for your SOC but you could still risk killing a cell

Now using shunts is a whole new issue. Shunts could cause a fire if the BMS is not properly designed. But there really is no way around using them in a BMS if you want to balance you cells automatically. Unless there is another way I don't know about.


One of the other problems is voltage sag. With the Enginer system I don't think it would be that much of an issue but with an EV you can draw down the voltage below 2.5v per cell during acceleration. If your BMS is designed to cut off the whole pack when one cell reaches 2.5v then your pack will be disconnected. I dont think this would be an issue with the Enginer system since it more of a constant power output.

If I did my calculation right the Enginer system would draw about 1C from the battery cells or about 40A. So I think once the voltage reaches say 2.5v the cells would be pretty much be empty at 1C.

I'm no battery expert but I'm learning a lot. Just dont know if what I'm learning is the correct stuff.

I'm getting tired. Dont even know if this post makes since or if I just smashed my keyboard.

 

I think we are on the same page now. That is what I have in my mind. The balancers may be doing more harm then good when charging and when finished charging. You have the same idea as I have. When the drift starts to show I will give it a boost charge from my 3.6V 2amp charger or I can discharge individual cells with my H7 halogens bulbs to get to about the same voltages. I know it is not an exact science but I am hoping that I will not have to tear down the system for a complete 64 cell parallel balance more than once a year.

I am on my 16th cycle since following the bottom balance and have not used the balancers for the second day in the row now. I think I may not have need the balancers after my commute on the first few days but wanted to be safe. My cells are about 9mV from highest to lowest. It is pretty amazing. I am pulling about 6kWh from the grid on each charge for about 35miles. Considering a 80% efficient charger that is about 4.8kWh is being used on my 35mile commute. I think my charger maybe turned down to 56V as a cut off since I did it myself. Will experiment with turning up the charger later after about 60 cycles. For now I have enough charge to run the 2 converters for the entire 35 commute. My last two days with 70 mile commutes resulted in a 99.9mpg daily from the MFD trip meter. The ScanGuage II is telling me I am getting about 110mpg daily right now.

 

Accordingly to your sophisticated diagram, you have two balancer leads over the same pair of batteries. Why? and what's the benefit of reading the same values with 2 different balancers?
Your batteries "buddy" installation is wrong by standards of pairing batteries for usefully load combination. By fallowing your procedures, you'r limiting the total WAh total power to the wiring sizing jumping cells across the two batteries arrays.
I have an extra spool of wire that I can donate you if you want to keep doing "jump rope".
But this is only about you and your imagination.

 

In Dan's diagram the main current flow would the + and - terminals of the 2 48V strings? The strings would be the combined load not the buddy cells. The "jump rope" as you so cleverly referred to is just to pass current between cells of two differnet strings. Or am I just not understanding the diagram??

I also run two balancer per buddy group at first to help to balancer faster...But since I don't use the balancers much anymore I use the leads for my cell monitors.

 

Correction, in this configuration you just pass VOLTAGE not current to top off buddy cells arrays.
Still, the double configuration of the balancers is wrong and defeats it's sole purpose.
But I guess that having spare electrical components and willing to experiment in your own is part of the learning process, but it is not right to present some experiments that will lead to inconclusive results that will promote early components breakdown under other's people warranty.
OK kids, the sand box is still open, time to play. :cheer2:
I will seat down now and keep watching ya over my shoulder...........

 

Dan,

Yes they are the same connectors and are wired the same. Completely plug and play. I just got time to catch up on Jack Rickard's Videos and notice he included a segment on the same cellLog 8S I am using as a viable BMS. The CellLog has a really cool difference measurement for Pack and cell voltages. That really helps for quick glances during my drive. I suggest you take the time to watch his video on how chargers work. It has really enlighten me on this Enginer kit and my up coming EV conversion of a Toyota MR2.

-Chan

 

I admit I am a novice but how does one go about passing voltage from one cell to another???? Would that not consist of certain amount current passing from cell to cell there by raising the voltage in the less charged cell while lowering the voltage in the other cell. Batteries hold charge right???

I am going to take this oportunity to put a disclaimer in this thread. Everything I discuss here is of my opinion and use any idea here at your own risk. I actually purchased extra 2kWh of batteries from Jack at Enginer as I did not want to abuse the warranty if this experiment went bad.

 

Good Sunday Dan,

Here is a link to Jack Rickard's Weekly Shows. If you scroll down you will see all the videos. When you get time watch the the 11/13, 12/18, and 1/15 episodes. They are quite long and he can talk up a storm but very informative.

Electric 1957 Porsche 356 Speedster EV

As for CellLog devices, USB function is right. You can setup multiple files on the device to log to and using the LogView software(included in the box) you can down the files after you complete your charge or drive. You can then save the files to CSV files for use in Excel. I setup two files, Discharge and Charge. I select the files for the situation and have the device log for later download. The device has like 16mb of storage which is good for like 3 days of recording.

 

Dan, the commentary I made about the balancers is as you said, redundant in this application but I never mentioned that the balancers are wrong; instead, that their double application defeated their purpose.
A concise explanation of this particular balancers operation: these do balance any cells connected to it to the variable top level of charging handled by a microprocessor, when any given cell reaches their own top value, could be 3.6v in this case, a resistor shunts (bypass) the cell passing the string charge voltage to the total string, this shunt is designed upon the designed application of the battery cell type and kind.
When, like in your application, you stack 2 balancers together in the same cell, both balancer branches fight to have control of wherever it is connected to them, in this case one battery and 2 shunts loads at the same time, the operation will work, completely out of specifications with out a defined purpose and probably doing more harm than good in a charging process.
This balancers are designed to handle certain shunt current at all times in certain amount of time, now if you “buddy†cells in a 2 or 3 or 4 configuration, meaning, paralleling their (+) and (-) terminal together, so you can equalize un-equal cells and level charge them, this Chargery balancer are not rated for the current handling that are implemented with, more likely will take 4 or 5 times longer in an unmaned mode.
These Chargery balancers do nothing on the low end discharge, so no comments about.
Taking a consideration of the basic origin of components for a lower priced introductory PHEV kit, the actual enginer system doesn’t provide a more sophisticated charging/balancing system for the same low price. In the future, may be. Is it there room for improvement? Yes.
You will have to resource yourself to different suppliers up to now.
In my case, I will opt for a more sophisticated Battery Management System in order to keep the health of the LiFePO4 batteries. A newer addition to the experimental EV/PHEV market is the MiniBMS equipment at the lower cost option, next step will be Elithion BMS.