Batteries and New Battery Technologies

[CDK]

Well, I was thinking about Bert's project, with the cells in tubes and the tubes paralleled. Mechanically a nice solution, but to avoid fireworks in his design each cell has to be monitored unless the charging current is impractically low.
He probably has to invent a tube where you can stack 12 cells and they are all parallel....

[Jeremy Harris]

Yes, in a multi cell pack using Li cells the cells have to be paralleled first, then connected in series, or else they do not load share well and a cell failure will result in that whole string failing (they fail to a high resistance or open circuit condition normally).

All the packs I build, or those built commercially, use this technique, as it not only provides better overall reliability, but it tends to balance out the inevitable small capacity differences that exist between cells, making the whole pack have a higher usable capacity (usable capacity being determined by the first cell set to drop to the cut-off voltage).

[michael pierzga]

Be careful charging those batteries.

http://www.sailmagazine.com/boatwork...too-hot-handle

[BertKu]

Originally Posted by Jeremy Harris

Yes, in a multi cell pack using Li cells the cells have to be paralleled first, then connected in series, or else they do not load share well and a cell failure will result in that whole string failing (they fail to a high resistance or open circuit condition normally).

All the packs I build, or those built commercially, use this technique, as it not only provides better overall reliability, but it tends to balance out the inevitable small capacity differences that exist between cells, making the whole pack have a higher usable capacity (usable capacity being determined by the first cell set to drop to the cut-off voltage).

Thanks Jeremy and CDK. Well I have certainly lots of hurdles to overcome. Attached my concept of the pipe system. I am able to expand, 12 per pipe, or like I have it now, 8 LiFePo4 in series, 24 Volt charging with only 900 mA. I can pack it on top of each other or, have it under a large area under the floor.
Folks I have time, the hole night, the whole week the batteries can be re-charged. It is not a motorcar, which need a re-charge in 12 seconds. I had hoped, that the solution of CDK with 8 x 5 Ampere Diodes in series, simple and like many members of this forums state : get a KISS solution, maybe also have to be abolished. I still have to try it out at 10 – 40 degrees Celsius. Here my e-mail to the manufacturer. I am now really confused. I may have to stick to Spiral deep charging SLAB’s
***************************************************************************
Happy Thanksgiving day to you and your family.

In the future days god will bring you the most beatiful things to your life. And thanks god bring you to my life, god will bless you and your family forever.

Yours,
Sincerely
Black
Dear Black,
Thank you for your well wishes, same to you
Will you do me a favour. Will you speak to the engineers.
I have your 16 watt, 5C , 3.3 Volt 90 mm / 32 mm LiFePo4 batteries bought 2 years ago.
Question.
If batteries are not deep charged , but kept charged at regular intervals.
How many years will they approximately last.
5 years.
10 years
15 years
2nd Question.
If I charge the batteries with only 300 mA.
Instead of 5 Ampere or 1C.
Will there be a long term advantage or disadvantage?
Many thanks Black, if you could find out for me.
Bert
*********************************************************************
Dear Bert
Long time didn't hearing form your,how is everything for your?
Regarding the LiFePo4 batteries question,our answer is follow:
1.If batteries are not deep charged,just kept charged at regular intervals.you can used about 2000times.ps:i can don't how many times your will use at 1years.so they approximately last it 2000tims.
2.Will be a long term advantage for the battery,if you recharge the battery under 1c. that's could be the best way. <<<<<<<<<<<<<< !!!!!!!!!!!!!!!!!!!!!!!!!! (1c = 5 Ampere)
Looking forward to hearing form yours feedback.
Best regards
Black
2011-12-16

Hi Black
What I mean is.
If I only charge a LiFePo4 battery 5 times per year.
Will the battery last for 20 years???
Or will it dry up.
And only last for 10 years??.
What is the life expectancy for your LiFePO4 batteries, I bought from you.
***************************************************************************
Hi Bert
My engineer said the LiFePo4 battery can last for 5-7years.use about 2000times.
Do you have good ways to find the customer?
Looking forward to hearing form your soon
Best regards
Black
*****************************************
Bert

[BertKu]

Originally Posted by Jeremy Harris

Bert,

Below is the 4 lithium cell charge controller that I use. The charge voltage for each cell is limited by the cell circuit on the left (there is one of these circuits for each cell sub-pack - 4 off for my 8p, 4S battery pack). The charge controller as shown is for a four cell battery pack, but it can easily be extended to as many cells as you wanted.


This is the circuit on my boat battery pack and it works well.

Hi Jeremy, After I have overcome the one nightmerry, I am encountering the next. Am I correct in saying that should I use 960 (1000) cells. i.e. 2 x 480 cells per motor, I must place 40 batteries parallel, then the next 11 times (36 Volt) or 7 x (24 Volt) of sets of 40 batteries paralllel in serial. According what I have understood, I only need then 8 or 12 of those circuits as per your diagram. CDK suggested already earlier to use a transistor. I did experiment with the LM1117 but I had only the 3.3 Volt type in my stock. I placed a diode in series with the device.
While I was in Cape Town in December with a supplier, he unfortunately did not have smaller transistors. The BDW82 did not fit between the battery and the wall as per photo in thread 1069. I have space for a long narrow printed circuitboard, I have space for diodes, but then it gets cramped, special when it needs so many other components on them.
I tested the 8 x 5 Ampere diodes in series, not suitable, the leakage current at 3.3 Volt is too high. It means I am discharging the batteries continously.
Will experiment some more with your circuit. The advantage of the square pipe batery system, it is 40% lighter than the Spiral SLAB. The disadvantage is, I need too much electronics, it need slightly more area space and the cost is getting out of hand.

Jeremy have you read my motor disaster? It never came up in my mind that after I tested one sensor, that many others were the wrong type. It took me ages to remove the glued sensors between the ancres, to have them replaced with the correct sensor.

With the second motor, I just left them and placed/glued new latched types in the motor. Thanks so much Jeremy and CDK for all the help given. I am pleased that that is now working from 1 rev/sec to 4800 Rev/min.

Bert

[Jeremy Harris]

Originally Posted by BertKu

Hi Jeremy, After I have overcome the one nightmerry, I am encountering the next. Am I correct in saying that should I use 960 (1000) cells. i.e. 2 x 480 cells per motor, I must place 40 batteries parallel, then the next 11 times (36 Volt) or 7 x (24 Volt) of sets of 40 batteries paralllel in serial. According what I have understood, I only need then 8 or 12 of those circuits as per your diagram. CDK suggested already earlier to use a transistor. I did experiment with the LM1117 but I had only the 3.3 Volt type in my stock. I placed a diode in series with the device.
While I was in Cape Town in December with a supplier, he unfortunately did not have smaller transistors. The BDW82 did not fit between the battery and the wall as per photo in thread 1069. I have space for a long narrow printed circuitboard, I have space for diodes, but then it gets cramped, special when it needs so many other components on them.
I tested the 8 x 5 Ampere diodes in series, not suitable, the leakage current at 3.3 Volt is too high. It means I am discharging the batteries continously.
Will experiment some more with your circuit. The advantage of the square pipe batery system, it is 40% lighter than the Spiral SLAB. The disadvantage is, I need too much electronics, it need slightly more area space and the cost is getting out of hand.

Jeremy have you read my motor disaster? It never came up in my mind that after I tested one sensor, that many others were the wrong type. It took me ages to remove the glued sensors between the ancres, to have them replaced with the correct sensor.

With the second motor, I just left them and placed/glued new latched types in the motor. Thanks so much Jeremy and CDK for all the help given. I am pleased that that is now working from 1 rev/sec to 4800 Rev/min.

Bert

Yes, Bert. Build parallel banks of cells and put one shunt circuit across each bank, then connect the banks of paralleled cells in series to make the pack. The diagram below (for my battery pack on the boat) shows how it's normally done. The advantages are significant, you don't need as many cell monitor circuits, a failure of a single cell (they fail open circuit) doesn't lose much capacity and the cell capacity variations tend to be evened out, giving a higher overall usable capacity.

Zener diodes won't do the job, as you are finding out. The standby leakage current needs to be small and the "knee" where the shunt starts to draw current and protect the cell needs to be sharper than a zener can give.

[BertKu]

Originally Posted by Jeremy Harris

Yes, Bert. Build parallel banks of cells and put one shunt circuit across each bank, then connect the banks of paralleled cells in series to make the pack. The diagram below (for my battery pack on the boat) shows how it's normally done. The advantages are significant, you don't need as many cell monitor circuits, a failure of a single cell (they fail open circuit) doesn't lose much capacity and the cell capacity variations tend to be evened out, giving a higher overall usable capacity.

Zener diodes won't do the job, as you are finding out. The standby leakage current needs to be small and the "knee" where the shunt starts to draw current and protect the cell needs to be sharper than a zener can give.

About the zenerdiodes, I asked the manufacturer whether they ever would be able to find a proces to produce a sharp "knee". It would be a fantastic solution, simple. However they said, it is impossible to create a very sharp knee in a production proces for zeners below 9 - 10 Volt.

I am deligthed that I don't have a second nightmarrry on my hand, although you ask me to throw my beliefs on NEVER having batteries directly connected parrallel over board. O.K., I am willing to change from religion and will make a sixt pipe parallel to those 5 already exsisting pipes and place the electronics in them. Drill holes along the 6 pipes and connect all batteries parallel. It is certainly much simpler than any other known solution.
Thanks Jeremy, I will start building it, in the way you have made the sketch and proposed.
Bert

[Landlubber]

...just a question for someone out there that understands this.
I am helping my mate wire his boat, and I explained to him about series and parallel connections. He connected his bank of 5 batteries in parallel by joining the + and the _ terminals, but he has actually completed the loop, so ALL the + and ALL the - are connected.
I have never seen this before and was wondering if there would be any reason not to leave it that way and just get my output from and of the two terminals.

[Jeremy Harris]

Originally Posted by Landlubber

...just a question for someone out there that understands this.
I am helping my mate wire his boat, and I explained to him about series and parallel connections. He connected his bank of 5 batteries in parallel by joining the + and the _ terminals, but he has actually completed the loop, so ALL the + and ALL the - are connected.
I have never seen this before and was wondering if there would be any reason not to leave it that way and just get my output from and of the two terminals.

No, that's a good strategy for balancing current taken from all the batteries, a little like the ring main used for house wiring, it shares the load more evenly.

A way to get perfect load sharing is to parallel the batteries with big cables, then take the negative feed from the battery at one end and the positive feed from the battery at the other end. That way all the batteries will contribute equally to the load, with equal cable voltage drops across the whole pack.

[Jeremy Harris]

Originally Posted by BertKu

I am deligthed that I don't have a second nightmarrry on my hand, although you ask me to throw my beliefs on NEVER having batteries directly connected parrallel over board. O.K., I am willing to change from religion and will make a sixt pipe parallel to those 5 already exsisting pipes and place the electronics in them. Drill holes along the 6 pipes and connect all batteries parallel. It is certainly much simpler than any other known solution.
Thanks Jeremy, I will start building it, in the way you have made the sketch and proposed.
Bert

Bert,

Some batteries work very well in parallel, some will die if you try it!

Lead acid cells and lithium cells, are voltage limited, so work better when parallel connected at the cell level. Nickel cadmium, nickel iron and nickel metal hydride cells are current limited and will cause big problems if you parallel them together at the cell level.

[philSweet]

landlubber,
completing the loop also allows you to isolate and remove any one battery without additional jumpering.

[BertKu]

Originally Posted by Jeremy Harris

Bert,

Some batteries work very well in parallel, some will die if you try it!

Lead acid cells and lithium cells, are voltage limited, so work better when parallel connected at the cell level. Nickel cadmium, nickel iron and nickel metal hydride cells are current limited and will cause big problems if you parallel them together at the cell level.

Well, I am never too old to learn something. With Lithium it is a learning curve for me. I am very surprised about Lead acid batteries. We may have to add one sentence for clarity for Landlubber. "Provided all batteries are new , the same kind i.e. kind of material used for the plates and fully charged". As soon one put different sizes, different plate materials batteries parallel, I am afraid, I prefer than some schottky diodes between them.

[RayThackeray]

Let me understand this, because I'm planning a very large multiple 12 volt bank with lead-acids. My plan was to buy Trojan T-105 6 volts and series to 12v, then parallel them there.

Obviously you can't parallel at cell level because commercial batteries don't offer that - or do they? Are you saying buy 2volt (presumably single-cell) batteries, and parallel a bunch of them, then series the paralleled banks to 12v?

Ray

Originally Posted by Jeremy Harris

Bert,
Lead acid cells and lithium cells, are voltage limited, so work better when parallel connected at the cell level. Nickel cadmium, nickel iron and nickel metal hydride cells are current limited and will cause big problems if you parallel them together at the cell level.

[BertKu]

Originally Posted by RayThackeray

Let me understand this, because I'm planning a very large multiple 12 volt bank with lead-acids. My plan was to buy Trojan T-105 6 volts and series to 12v, then parallel them there.

Obviously you can't parallel at cell level because commercial batteries don't offer that - or do they? Are you saying buy 2volt (presumably single-cell) batteries, and parallel a bunch of them, then series the paralleled banks to 12v?

Ray

Ray, as long the 6 Volt batteries you place in series are the same kind, preferable fully charged and then connected parallel at the 12 Volt level, it should not be a problem. Although I personally prefer schottky diodes to connect parallel switched batteries, but that is a personal preference. The problem will surface, when one battery packs up in a few years time. Then you have to make sure in what you replace, has the same kind of condition, like plate construction and material, same level of charge as the older batteries. If one uses Schottky diodes, one does not have to be carefull. However Schottky diodes has its own problems.

Because of the difficulty with Lithium batteries in monitoring each battery cell, the solution from Jeremy suits me to the bone, and I am certainly going to try it out. Bert

[RayThackeray]

OK. So for the sake of clarification, I'm looking for two 12 volt banks of 8 6volt cells (all purchased at the same time, same charge level on installation).

So if I understood this correctly, I should parallel in "groups" of four 6v batteries, then series the groups into 12v banks?

Schottky isolators are probably better than silicon because their voltage drop is about a half, I was thinking about that myself. But I'm leaning to simply paralleling without isolators and maintaining the banks carefully.

Originally Posted by BertKu

Ray, as long the 6 Volt batteries you place in series are the same kind, preferable fully charged and then connected parallel at the 12 Volt level, it should not be a problem. Although I personally prefer schottky diodes to connect parallel switched batteries, but that is a personal preference. The problem will surface, when one battery packs up in a few years time. Then you have to make sure in what you replace, has the same kind of condition, like plate construction and material, same level of charge as the older batteries. If one uses Schottky diodes, one does not have to be carefull. However Schottky diodes has its own problems.