18650 equaliser

Egam and this may be for others too. Note that dumping of energy for balancing is generally only done on initial balance. With that said if the balancing afterwards dumps that much energy in form of wasted heat that it becomes a problem you have a bigger issue on the battery bank that anyways need to be sorted :)

I have looked at how much I have balanced on my 100kWh battery bank last weeks and frankly I have not balanced more than a few Ah total over 6++ weeks. And thats only done when fully charging tha pack = i have waste energy to waste kind of...

With that said the active balancing have its advantages and you can use uneven packs since you can both bottom and top balance at same time transfering energy to extend the range.

Great progress btw. Flying balancer is actually a decent way. There are a couple of really good datasheets out there describing the main different types of balancing if anyone is interested to read more.
 
Egam Said:

The questions I have are:

Can you put two of these boards in series and use 14 of the 16 cells?

I have not tried this but it should be OK.

OR can you just use the flying cells for the load and use 8 cells as the flying sells instead?

I dont understand the question

Can you hook the 7 flying cells in series?

No!
 
Brian Drury,
I asked about the flying cells because there are seven of them in your design, and my intended application has 14 cells in series.
 
If I understand the design correctly, the basic circuit is for a single flying cell between a pair of static cells. That flying cell is always in paralle with one of the two static cells, and it gets switched betwwen them regularly. Therefore it equalizes with one, then the other.

To take this to more that 2S, you end up with a zigzag between regular and flying cells, like:
1
A
2
B
3
C
4

A switches between 1 and 2, B switches between 2 and 3, C switches between 3 and 4. So to have less than N-1 flying cells would require a vastly different design.
 
@ Egam
Yes your explanation is correct.
I wrote a short circuit description and diagram of a simplified 3 static cell arrangement which I hope will explain things even more:

S1, S2 & S3 are static cells wired in series.
F1 & F2 are flying cells.
Q1 Q8 are MOSFETs.
The MOSFETs are switched in pairs. Q1 & Q3, Q2 & Q4. (Q1 & Q5 are P type)
The top 14047 controls Q1, Q2, Q3 & Q4.
So, if Q1 & Q3 are ON then F1 is in parallel with S1. If S1 has more charge than F1 then current will flow from S1 into F1. Alternatively if F1 has more charge than S1 then current will flow from F1 into S1.
When Q2 & Q4 are ON then F1 is in parallel with S2. If S2 has more charge than F1 then current will flow from S2 into F1. Alternatively if F1 has more charge than S2 then current will flow from F1 into S2.
The bottom 14047 controls Q5, Q6, Q7 & Q8.
The basic action is as above so charge is distributed S2 S3 & F2. Also, because the two 14047 are not synchronised there will be occasions when F2 is in parallel with S2 and F1.
If we start with S1 charged and S2, S3, F1 & F2 discharged then eventually the charge from S1 will be evenly distributed between all 5 cells.

image_rhzgwn.jpg
 
I spent a while writing code for my DIY BMS to provide an automatic charge/discharge cycler so I can characterise the cell equaliser performance. Now I can get some serious data without having to monitor it all the time.

In the following graphs the charge current is 405mA and the discharge rate is 1.3A. The horizontal axis is time with a tick interval of 10 Seconds. The vertical axis is in volts. The top two traces indicate the application of charge current or load.

The BMS is set to apply a 1.3A load until the lowest cell voltage reaches 3.5V. It then turns off the load and switches on the charger. The charger stays on until one of the cells reaches 4.0V when the charge is switched off and the load is re-applied.

The first graph is without the equaliser. You can see that cell 7 has more charge than the others and cell 1 is a bit low. As expected, the performance remains the same for multiple cycles because there is no cell balancing.

image_lggfht.jpg

The second graph simply carries on where the first graph stops but this time the equaliser is switched on. The initial discharge looks similar to the first but as charging takes place the charge on cell 7 is being equalised with cell 6 which flattens the rate of rise on cell 7.

image_ncrxyi.jpg

You will also notice that without the equaliser charging terminated after 65 minutes but with the equaliser charging took 188 minutes on the second cycle.

Static cells 1 and 8 have only one flying cell to share power with therefore I would expect them to take longer to stabilise than the other static cells.

The power required to charge to 3.8V without the equaliser fitted is 13.5Wh. The power drawn is 13.4Wh to reach 3.5V

The power required to charge to 3.8V with the equaliser fitted is 29.5Wh. The power drawn is 29.3Wh to reach 3.5V

So, the equaliser provides an additional 15.9Wh or + 118.7% for + 87.5% extra cells.

My conclusion so far is that the equaliser is highly beneficial. Not only are the cells now working in harmony the energy available from the pack is increased by 118.7% with no wasted power.

I intend to carry out more tests and will be interested to hear what others think. Also, it would be great if anyone has data they can share using alternative methods.
 
Brian,
Are you ready to move on to the next stage of figuring out having the boards assembled and shipped? For my project I'd love to use your equalizer and charging board but I need to start working on it soon to be ready for use by April. I need to design custom cases for my battery packs (to be 3d printed), which I can't properly do until I have samples on hand. In the interim can you provide me the following:
-footprint of the equalizer PCB
-footprint and height of the Nano/charger PCB (excluding heatsink)
-is the Nano/charger PCB off-the-shelf or custom? if off-the-shelf, source? if custom, how hard would it be to instead have a monitoring board with the nano and use separate means of charging? If not feasible, then is it possible to implement CC/CV charging? I know you have decided against it, but it would be desirable for me.

For the issue of the rainbow ribbon cables having high resistance, that seems to be an implementation-by-implementation issue, and I have my own ideas on solving that which only apply to my use case (individual batteries similar to those for drills or other power tools). I'm going to use battery holders like https://www.amazon.com/dp/B01M8P2HH2/ with a PCBs that are sandwiched between two such holders, one PCB set up for the series cells and one PCB set up for the flying cells. That allows the resistance between cells to be low (traces as wide as possible within footprint) and minimizes the length of cable necessary to connect to the equalizer and monitoring boards.

Side note, I have decided to use your suggestion of keeping each battery as 8S1P/7F and use DC-DC converters to step down to my application's voltage.
 
I got these boards made by http://www.ourpcb.com/

There were no problems at all, the boards worked great and i didnt find any mistakes.
(I've only used the company once though)
They did post me a card for new years though, so they're the kind of company that will go out of their way to be friendly
 
Michelle01 said:
I got these boards made by http://www.ourpcb.com/

There were no problems at all, the boards worked great and i didnt find any mistakes.
(I've only used the company once though)
They did post me a card for new years though, so they're the kind of company that will go out of their way to be friendly
Click to expand...

How many did you order and what was the price? Any comments on the process? You just gave them the files for the board and the list of components?
 
Brian Drury said:
Static cells 1 and 8 have only one flying cell to share power with therefore I would expect them to take longer to stabilise than the other static cells.
Click to expand...

This is all pretty fascinating and promising. This is exactly the way balancing should be done, very clever bit is using flying cells instead of flying (super)caps, as they get expensive fast and also have to think of a way to deal with high currents if using them.

Anyways have you thought of adding a flying cell between static cells 1 and 8, then you would make topology even and balanced? So same number of flying cells as static, then the low and high cells would get the same balancing as rest?

I definitelly would be interested to get one of your boards!

Brian Drury said:
The basic action is as above so charge is distributed S2 S3 & F2. Also, because the two 14047 are not synchronised there will be occasions when F2 is in parallel with S2 and F1.
Click to expand...

Also would it be hard to synchronize the switching, to make monitoring the balancing process easier? I would think it would make sizing the fuzes and other things easier also as balancing currents would be easier to calculate but I dont know if this is important.
 
Improved topology, suggestion:

Flying cells 1-8: F1, F2....F8, divided to two groups, odd F1, F3, F5, F7 and even F2, F4, F6, F8
Static cells 1-8: S1, S2....S8

Alternate connections by switching mosfets between even (flying cells connected) high configuration and odd high:

Even (flying cells) high, F1 connected to S1 so F1-S1, F2-S2, F3-S3, F4-S4, F5-S5, F6-S6, F7-S7, F8-S8.
Odd high, F1-S3, F2-S1, F3-S5, F4-S2, F5-S7, F6-S4, F7-S8, F8-S6

Maybe easiest to draw this on a piece of paper, it is pretty simple idea really in the end. Basically same as flying cell between high and low cells of the pack, but distributed evenly to keep the potential differences smaller - reduces smoke if switching goes wrong.

This way voltage/charge in the cells is evenly circulated and balanced all around the pack, and BMS has no balancing left to do, just monitoring function left. Battery always in balance, whether charging or discharging, full or empty or in between. Top, bottom and middle balancing all at the same time.

Flying cells could be pretty substantial capacity, as capacity is not really wasted in them, but actually is in active use in parallel to fixed cells almost 100% of the time (minus the switching delay, few microseconds?). Say 10-50% of the fixed cells could be good capacity for flying cells, obviously need to run some simulations to find optimal value. Also the switching speed could be pretty low, I would think 1 Hz would be just fine. Might even work with good old DPDT-relays, could maybe lower the switching speed even more and maybe also make it adaptive depending on balance status of the pack.
 
Just a question about the capacity of the flying cells...

Am I correct in thinking that the capacity of the flying cells do not need to be the same as the 'main' cells?

E.g. I could use this concept to balance a 14s80p with 13 flying 'batteries' of say 20p each? In fact, maybe I could/would use 13 of the 15Ah 5C LiPo packs I have. Those 13 cells would add another 700+ W/hr to the system, and be able to handle balance currents of up to 75 Amps. I would assume this would be sufficient.
 
Grumplestiltskin said:
Am I correct in thinking that the capacity of the flying cells do not need to be the same as the 'main' cells?

E.g. I could use this concept to balance a 14s80p with 13 flying 'batteries' of say 20p each? In fact, maybe I could/would use 13 of the 15Ah 5C LiPo packs I have. Those 13 cells would add another 700+ W/hr to the system, and be able to handle balance currents of up to 75 Amps. I would assume this would be sufficient.
Click to expand...

Definitely flying cells do not need to be same capacity as main cells, but I think they should still be substantial percentage of main cell capacity. I am thinking something like 10% - 50% of main capacity would be good working number.

I also think for concept to work really well every main cell should be connected to a flying cell most of the time (minus switching delays), all the time, when charging and discharging. This way flying cells would form sort of integral part of the pack capacity and whole setup would stay balanced. If all staticcells expect the high and low cell have a supporting flying cell connected most of the time, and high and low only 50% of the time, I am thinking this could easily lead to imbalance by itself, as those two cells would effectively have less capacity as rest of the pack.

Also this way (having equal number of static and flying cells) switching frequency could be lowered a lot, 1 Hz or even 0.1 Hz or even lesswould work just fine. Lower switching frequency might even permit using relays instead of mosfets, maybe.

This means number of flying cells should be same as static cells, in my opinion.
 
I see no reason why the number of static and flying cells cannot be the same, but then that really doesn't mean much :)

As an extension of this entire concept, it would even be possible to have a single flying cell that gets switched across all static cells one at a time. Achieving equilibrium/balance across the packwould certainly be slower, but once achieved, I don't see any reason why it cannot be maintained. Doing it this way might make it an easier design. Or not...
 
Grumplestiltskin said:
As an extension of this entire concept, it would even be possible to have a single flying cell that gets switched across all static cells one at a time. Achieving equilibrium/balance across the packwould certainly be slower, but once achieved, I don't see any reason why it cannot be maintained. Doing it this way might make it an easier design. Or not...
Click to expand...


Of course this would work, but then you kind of loose the capacity of flying cell and also for balancing to happen in reasonable time you have to switch pretty fast, meaning higher switching losses and not being able to use relays.

Also there are some dangers in this design, you have to have a switching path between high and low cells, and if some switching element does not switch at the right time there will be smoke or at least blown fuses. Topology I suggested keeps maximum voltage potential difference in around 3 cell (around 12V?) levels whatever the pack size. This becomes even more important with higher voltage packs, if you have a HV pack manyswitching elements (mosfets, relays etc)might not have rating for the entire high to low level difference.

All in all, more flying cells makes the system simpler and better than having just a few. And as you are not really wasting anything with flying cells, I don't see that many negatives.
 
Hmm, have been thinking this a bit.

What if we went to make this balancing network even deeper? So connected to static "main" cells there would be a layer of first level flying cells, each connected to two main cells with a pair of DPDT latching relays, one in each terminal. These latching relays would have their coils connected in parallel, so every latching "up" or "down" would shuffle all flying cells at same time, switching them around so that the charge and voltage would balance/move around. Then behind this layer would be another layer of flying cells, connected same way but to the first layer of flying cells. Latching these levels would be done staggered out of phase, but with same frequency. Maybe this balancing network could be two or three flying cells deep? Maybe this balancing control could be turned on and off depending how balanced the pack is.

Result: automatically top/middle/bottom -balanced pack with no sensing electronics. No wasted capacity. Almost no power wasted balancing. Very simple control electronics.
 
I wonder if this kind of balancing (switching packs under load) is even possible with a high current setup ( + 150A draw ) ...
i mean, you have to make all connections, relays, wires, capable to carry those amps... or am i missing something here ?

Is a single flying pack not the better (easy) way to go ? and keep the static packs to carry the loads....

Also just thinking :)
 
wim said:
I wonder if this kind of balancing (switching packs under load) is even possible with a high current setup ( + 150A draw ) ...
i mean, you have to make all connections, relays, wires, capable to carry those amps... or am i missing something here ?
Click to expand...

I would think in high current setup current in balancing (flying) cells is directly proportional to their percentage of capacity. With 150A draw and 10% of thecapacity in flying cells maximum current forthem should be around 15A, right? Pretty smallrelays already take 15A easily, and automotive type relays easily much more.

There are also ways to mitigate this. First balancing couldbe disabled while draw is high. This could be controlled by BMS and/or dedicatedcurrent shunt. Also there could be a current limiting relay connected to flying cells, a relay with parallel resistor for example. This relay would open while switching to limit switching current, and then close immediately afterwards to limit losses, as balancing current should settle to lower value pretty fast after switching. Or even some more fancy current limiter, PWM even.
 
Well, the load current is devided by the strings (14s) in parallel, so i have 3 strings, so 50A/string. if i made 1 string 14 packs "flying" there would be 50A flowing when they are connected to a load.
This is not the balancing currents i have in mind, but the loads on the whole battery, flying packs included...
Yes, there are ways to control this, but it might get complicated (= expencive)real quick.

This is why i like to beleave a single (small) pack conecting to every pack in the battery, oneat a time is a easy way to get some kind of constantbalancing... and leave as much as possible static packs connected to handle the load.

But, i might be wrong... so i keep a close eye on this thread.

I am very intrigued in your way of thinking, and i willdo some testing on my system to see, and messure, what level of balancingasingleflying pack gets.
The reason i want something like this, is because my packs are a bit to big for a single longmon frombatrium.
 
You must log in or register to reply here.
Back
Top