Charge li-ion efficiently for cheap and balancing? go here

[frodus]

I think a charger would try and regulate voltage and current. You really want a supply that gives out a constant voltage output for all of the loads (DC-DC converters). Also, as the loads of all the DC-DC converters drop, the load on the Power supply will drop, and if its a charger, it might see that as time to trickle charge/turn off.

This idea is great though, and props for Doctorbass in pushing this forward and testing. Distributed charging has always been good for balancing batteries. In fact, I've got 12 Vicors and a 120/240VAC input 48V 32A output power supply. Just trying to get it working with my Sealed Lead Acid, but its very easy to scale.

 

[Doctorbass]

I think a charger would try and regulate voltage and current. You really want a supply that gives out a constant voltage output for all of the loads (DC-DC converters). Also, as the loads of all the DC-DC converters drop, the load on the Power supply will drop, and if its a charger, it might see that as time to trickle charge/turn off.

This idea is great though, and props for Doctorbass in pushing this forward and testing. Distributed charging has always been good for balancing batteries. In fact, I've got 12 Vicors and a 120/240VAC input 48V 32A output power supply. Just trying to get it working with my Sealed Lead Acid, but its very easy to scale.

Where did you got your 48V 32A power supply!!?? I've been searching alot on ebay for them.. and they are often 240V dur to the total power at output when 20A+ 48V..

I've bought a pioneer electronic 27A 48V..with power factor correction to 0.98.. and it work on 120V 19A max.. but when i received it does'nt worked.. the fan turn, but the output board seems to not operate or start... i've been playing alot with the inhibit pin and i got the connection diagram.. but i would really like to get the schematic!! it does'nt smell bad.. and the two board seems to be cosmeticly correct.. no fet or diode seems to be cracked... very stange. it ibclude some sub board vertical with epoxy and the semiconductor under that... like a pwm specialized board... and difficult to troubleshoot!

Also, the 48V 32A i've seen on ebat in the last days only work with 240V... :?

from now i will use my 0-100V 0-10A Power Ten power supply.. also got on ebay like new for 89$...great deal compare to the 48V supply i got too...

Doc

 

[frodus]

I think a charger would try and regulate voltage and current. You really want a supply that gives out a constant voltage output for all of the loads (DC-DC converters). Also, as the loads of all the DC-DC converters drop, the load on the Power supply will drop, and if its a charger, it might see that as time to trickle charge/turn off.

This idea is great though, and props for Doctorbass in pushing this forward and testing. Distributed charging has always been good for balancing batteries. In fact, I've got 12 Vicors and a 120/240VAC input 48V 32A output power supply. Just trying to get it working with my Sealed Lead Acid, but its very easy to scale.

Where did you got your 48V 32A power supply!!?? I've been searching alot on ebay for them.. and they are often 240V dur to the total power at output when 20A+ 48V..

I've bought a pioneer electronic 27A 48V..with power factor correction to 0.98.. and it work on 120V 19A max.. but when i received it does'nt worked.. the fan turn, but the output board seems to not operate or start... i've been playing alot with the inhibit pin and i got the connection diagram.. but i would really like to get the schematic!! it does'nt smell bad.. and the two board seems to be cosmeticly correct.. no fet or diode seems to be cracked... very stange. it ibclude some sub board vertical with epoxy and the semiconductor under that... like a pwm specialized board... and difficult to troubleshoot!

Also, the 48V 32A i've seen on ebat in the last days only work with 240V... :?

from now i will use my 0-100V 0-10A Power Ten power supply.. also got on ebay like new for 89$...great deal compare to the 48V supply i got too...

Doc

32A is during 240.... unfortunately

its ~16A on 120VAC..... which is fine, I can do up to 12 or 13 DC-DC converters with that. Each one draws 1.25A under full load. So I can get 12 working full power no problem.

I think I've seen one at 20A, but not sure if it was at 240 or not.

 

[jeffkay]

Frodus, Like Doc, I have been using DC-DC's and power supplies in my charging. I agree, when charging say a cell-voltage sub-pack I need to start with higher voltage like 5v and then taper for CC as I go along. The knee comes on and of course then it takes differing voltage to maintain the current. Now interestingly, if I just use say 3.75 fixed from the start, then like you say, the current is high at the start and as it tapers off, since the voltage is locked, it would take several times longer at the back end of the charge to get the full charge condition. Some tests I did show that using the CV method does not seem to make any difference as long as you are not thermally hitting the cells. This is probably why the CC is the correct spec. Now here is something interesting: When using a power supply (in this case 61.5 volts for a nominal 48 pack) I notice that the amount of series cell-packs (16 in this case) seems to even out the charge amps. I see this clearly as i am monitoring with an amp meter and charging the pack at 8 amps.

Jeff K. "Deep Cycle" project

 

[safe]

Now interestingly, if I just use say 3.75 fixed from the start, then like you say, the current is high at the start and as it tapers off, since the voltage is locked, it would take several times longer at the back end of the charge to get the full charge condition. Some tests I did show that using the CV method does not seem to make any difference as long as you are not thermally hitting the cells.

This is the same result that was being presented for extending SLA cells... that you have a lot more freedom to hit the cells harder at the beginning and get more charge faster by doing it. But the thermal question was the real difficult part... new SLA cells are best not hit that hard with voltage and current because that wears them out. Old cells need the increased activity to wake them back up and shake off the sulfation.

Temperature is what limits the charging... you are "safe" going with the lower voltage all the time, but it's slower. If you want to go fast you have to be smart about it... or you get troubles.

Doc's fixed voltage and variable current concept is a good cheap way to get the job done and all those DC to DC units make sure that you are paying attention to the cells individual needs. It looks good... a good solid concept.

 

[Doctorbass]

Another advantage is the sensing input that they have!!

Cause charging normally at let say 15A with resonnable wire gauge like 14AWG, you will loose half a volt if taking account to the connections losse, wire lenght resistance etc... so the voltage may being 3.600V at the DC-DC output, but not at the cell tap...

So the CV stage in this case is not a real CV ... .. or at least it isdirectly and only at theDC-DC output :|

Ex: 3.600V at the DC-DCoutput
-the Wire lenght (2 x 5ft) at 2.52miliohm per ft for a total of 25miliohms (according to:http://www.powerstream.com/Wire_Size.htm
would give you a losse of V=R x I = 0.025 x 15 = 0.375V !!

+ the conections losse : let say 100mV

-for a total of 475mV !! so you could have 3.600V at the Dc-DC output and only 3.125V at the cell tap... :x

that situation is normal and that will cause your cell sto charge slower than they could!

Instead of charging at 10A for ex: 1h for a 10Ah pack , that could take 50% more time and taking 1h30 !

So the use of the sensing input to avoid this situation must help!

That would need you to add 2 small wires per dc-dc that will connect to the + and - sensing input and link to the other end to the cell tab or close to the connection plug that goes close to the battery.

What will happen in this case is that your DC-DC will use that feedback voltage measured from the battery plug end and will ensure that you always have the right voltage at this precise place. to do that , it will compensate and boost the output from 3.600V to 4.075V. knowing that you have 475mV lost at 15V the resulting voltage will be 3.600V.

that's a feedback controlled output.


and... most of the DC-DC have this option!

I use this 4 wire sensing principle for my cell capacity and internal resistance acurate measurement test when discharging. I mooded my RC megapower charger to have these 2 more wires as sensing input availlable outside of the test device. see that post:http://endless-sphere.com/forums/viewtopic.php?f=14&t=3410&p=49760&hilit=4+wire+sensing#p49760

see the difference with and without real CC-CV

see below the exemple of the difference of power a CC-CV and a not CC CV cahrging process have:

Doc

 

[efreak]

what a great summery post this dc dc Doc stuff is archive worthy ty

 

[RealPeterPan]

Hey Lads,

after 2 weeks searching the net ans specially the sphere i have to ask my question.

My situation is following:

I wanna build my own lithium battery pack from a huge ammount of Li-Ion cells i had a very cheap source for.
In fact I got about 800 battery packs for digital cameras for free such as this:

Bigger version:
http://webf.mypicturetown.com/P2Pwe...sIYqgDd4m1Zna5t6DNyoA8d%3D1XWDs9WwC3JzdeN-uJa

As you can see they have 2 Cells.
The pack itself has 7.4V which is a bit weired and i couldt find anything with this voltage.
The measured voltage is even higher ( up to 7.9V).

Anyway my questions now are:

1. How can i find out what kind of Li-Ion they are and is this really needed to know because Lithium batterys are all charged the same way?

2. Could I just parallel this packs to my needed Ahours and leave the cells in the pack with the BMS and erverything and chage them with Docs method DC-DC?

3. If not and i have to open and rebuild the the cells, has anyone a source for BMSs?
I serached ebay and google in every corner of the world but probably with the wrong keywords.
BMS semms to be only used in the sphere and by battery pack dealers who dont sell them seperately.

4. if there are no ready to use BMSs I probably wont be able to do this because i have a lack of tech skills and sources for electrical stuff (ireland is still in the dark age
would it be worth to seperate the cells from the batt-packs and sell them in packs of 10 or so on ebay?

My general plan is to build a pack of 48V 30Ah for an e-scooter.
So btw does anyone know if i can power the brusless mars with li-ion or is the current to high for the cells?

If you need more pics or info let me know.

Thanks in advance
Sven

 

[Doctorbass]

Hey Lads,

after 2 weeks searching the net ans specially the sphere i have to ask my question.

My situation is following:

I wanna build my own lithium battery pack from a huge ammount of Li-Ion cells i had a very cheap source for.
In fact I got about 800 battery packs for digital cameras for free such as this:

Bigger version:
http://webf.mypicturetown.com/P2Pwe...sIYqgDd4m1Zna5t6DNyoA8d%3D1XWDs9WwC3JzdeN-uJa

As you can see they have 2 Cells.
The pack itself has 7.4V which is a bit weired and i couldt find anything with this voltage.
The measured voltage is even higher ( up to 7.9V).

Anyway my questions now are:

1. How can i find out what kind of Li-Ion they are and is this really needed to know because Lithium batterys are all charged the same way?

2. Could I just parallel this packs to my needed Ahours and leave the cells in the pack with the BMS and erverything and chage them with Docs method DC-DC?

3. If not and i have to open and rebuild the the cells, has anyone a source for BMSs?
I serached ebay and google in every corner of the world but probably with the wrong keywords.
BMS semms to be only used in the sphere and by battery pack dealers who dont sell them seperately.

4. if there are no ready to use BMSs I probably wont be able to do this because i have a lack of tech skills and sources for electrical stuff (ireland is still in the dark age
would it be worth to seperate the cells from the batt-packs and sell them in packs of 10 or so on ebay?

My general plan is to build a pack of 48V 30Ah for an e-scooter.
So btw does anyone know if i can power the brusless mars with li-ion or is the current to high for the cells?

If you need more pics or info let me know.

Thanks in advance
Sven

First, hello and welcome abord!!

i'll try to answer you the best as i can.

A1: just look on those for a part or any number that you can search on google. they dont look like the 18650 popular li-ion cell but maybe you can measure them and see if the diameter and lengh correspond to this number expressed in mm. Ex: the 18650 mean 18mm dia and 65.0mm lenght

A2: these cell are rarely high current rated like the one used in the power cordless tools. so they can not take high current and their chemistry il often li-co (lithium cobalt and they are the one that are dangerous if used or charged not correctly!! fire or explosion can occur!

Normally, power cordless tools are li-manganese wich are safe and no need a special bms.
I would not recommand to charge these li-co cells with a dc-dc for the reason that they can not be chareges at high current and nees a special temp and voltage monitoring circuit.

A3: I know that batteryspace website have many lithium cell bms and will certainly have one for your cells

A4: I would recommand only to high skill in electronic person to build pack with li-co cause they are very critical to use and you need large ecxperience and comprehension of them to work with them safely!

Battery space have many great cell that are safe that could help you in your project.
48V 30Ah is a very large pack and would need a particular attention to the heat management and protection and compartment of many sub pack desing to gat a safe pack.

I dont want to discourage you to do that, but i would prefer to recommand you to start with a smaller project to get familiar with battery desing.

Doc

 

[sharkmobil]

Hey Lads,

after 2 weeks searching the net ans specially the sphere i have to ask my question.

My situation is following:

I wanna build my own lithium battery pack from a huge ammount of Li-Ion cells i had a very cheap source for.
In fact I got about 800 battery packs for digital cameras for free such as this:

Bigger version:
http://webf.mypicturetown.com/P2Pwe...sIYqgDd4m1Zna5t6DNyoA8d%3D1XWDs9WwC3JzdeN-uJa

As you can see they have 2 Cells.
The pack itself has 7.4V which is a bit weired and i couldt find anything with this voltage.
The measured voltage is even higher ( up to 7.9V).

Anyway my questions now are:

1. How can i find out what kind of Li-Ion they are and is this really needed to know because Lithium batterys are all charged the same way?

2. Could I just parallel this packs to my needed Ahours and leave the cells in the pack with the BMS and erverything and chage them with Docs method DC-DC?

3. If not and i have to open and rebuild the the cells, has anyone a source for BMSs?
I serached ebay and google in every corner of the world but probably with the wrong keywords.
BMS semms to be only used in the sphere and by battery pack dealers who dont sell them seperately.

4. if there are no ready to use BMSs I probably wont be able to do this because i have a lack of tech skills and sources for electrical stuff (ireland is still in the dark age
would it be worth to seperate the cells from the batt-packs and sell them in packs of 10 or so on ebay?

My general plan is to build a pack of 48V 30Ah for an e-scooter.
So btw does anyone know if i can power the brusless mars with li-ion or is the current to high for the cells?

If you need more pics or info let me know.

Thanks in advance
Sven

If you really want to make your own Li-CO battery pack, follow these principles

1. Research and get the data on your cells, their Ah capacity in particular. Calculate the number of cells needed, since you cells are free, double it.
2. Get the BMS board from batteryspace (DrBass already mentioned it). Get it so there is a reserve between the cut-off amps and your packs' capacity.
3. Overbuild, if you need to pull 30A out of them, build them in parallel, such you could pull 50A.
4. Make good oversized connections for the parallel packs themselves, them serialize them via BMS board.
5. Don't be afraid All these stories about blazing battery packs are just that stories - they served and continur to serve in the millions of laptop and camera packs around the world.
6. Don't be overconfident either. They like to be respected. For further info look for xyster's presentation of his ebike. He shows how he has build his pack there.

 

[TylerDurden]

5. Don't be afraid All these stories about blazing battery packs are just that stories - they served and continur to serve in the millions of laptop and camera packs around the world.

Well... there are those reports from the folks at RCgroups. Scores of em.

Yeah, those must just all be make-believe.

See you in the burn-ward.

 

[sharkmobil]

5. Don't be afraid All these stories about blazing battery packs are just that stories - they served and continur to serve in the millions of laptop and camera packs around the world.

Well... there are those reports from the folks at RCgroups. Scores of em.

Yeah, those must just all be make-believe.

See you in the burn-ward.

Given enough rope, there is always a chance one might hang himself.

If one got a score of free or near free Li-Co batteries in a decent shape, why would one not make an effort to produce a safe and reliable pack out of them? It is possible, maybe not as easy and forgiving as out of a123, but very doable, you know

 

[TylerDurden]

There is a reason RC flyers charge their LiCo packs outside, in contained areas (usually made of cinderblock or brick). Their packs are not nearly as large as those needed for ebikes & scoots.

Be as cavalier as you like for yourself, but advising others to take known risks is negligent.

Readers can feel free to read reports at RCgroups forums and here as well, to learn more on why very few EVers will use LiCo/LiPoly.

 

[sharkmobil]

I think the reason RC Flyers have trouble with these packs is because they are limited in size inherently. They put 2 or 4 cells in parallel and expect to pull same amperage out of them as from the LiPo pack. Putting any bigger number of cells is just not possible, as their contraptions would have trouble flying because of additional size/weight. They charge them using the LiPo chargers as well, capable of shoving up to 10A into these batteries, of course they are going to catch fire, wouldn't you?

On the eBike platform we are much less limited in size, not to the grams anyways, so we can overbuild, if need be. I'm sharing my own experiences, since when sharing personal observations is negligent?

 

[TylerDurden]

I'm sharing my own experiences, since when sharing personal observations is negligent?

If you really want to make your own Li-CO battery pack, follow these principles

1. Research and get the data on your cells, their Ah capacity in particular. Calculate the number of cells needed, since you cells are free, double it.
2. Get the BMS board from batteryspace (DrBass already mentioned it). Get it so there is a reserve between the cut-off amps and your packs' capacity.
3. Overbuild, if you need to pull 30A out of them, build them in parallel, such you could pull 50A.
4. Make good oversized connections for the parallel packs themselves, them serialize them via BMS board.
5. Don't be afraid All these stories about blazing battery packs are just that stories - they served and continur to serve in the millions of laptop and camera packs around the world.
6. Don't be overconfident either. They like to be respected. For further info look for xyster's presentation of his ebike. He shows how he has build his pack there.

I would say that is advising someone to use an unstable chemistry.

Every ES member that has used LiCo has blown-up a pack or had a near catastrophy, Xyster included. (He disappeared months ago for unknown reasons. Poof. )

 

[RealPeterPan]

I got a very convincing PM from Doc shortly after I posted my plans with the Li-Co batterys

So the batts went straight back where they came from.

Its just not worth the hassle to secure these time bombs.
With what i wanted do with them it would have cost me more in BMS, LVC, chargers, bomb shelter, etc...

Specially because the space for my packs will be between my legs i decided to go for the secure version.

Garys 16-cell BMS comes just right in time so the only thing i have to do now is to build up my own collection of DeWalt 36Vs

By the way, has anyone a good source aside from rubygirl on ebay?

Maybe one in Europe??