Try to draw the schematic....
The inductive flipping back always takes place to the opposite voltage direction vs it was forced before. If you want to feed that into cells on lower potential than the primary, then you have to isolate the secondaries.
Means you end up in a flyback.
And of course you can limit by taking energy from one cell and feeding into the series of all (or with multiple output windings into all cells except to the one which delivers ther energy).
If you implement for n cells the same amount of n flybacks then you can realize a strong and efficient limiter. But its quite a lot of efforts, just for balancing and limiting.
Trouble free Batteries:
LiFePo is not much trouble. Chargers with monitoring and/or balancers are no headache. Compared to LiPo they have no thermal runaway and are more safe.
Only if you want to use an existing lead acid charger in brute force technology, like the battery charging system in a normal car without modification for LiFePo - only then you need a balancer or limiter which can bypass quite some current.
The inductive flipping back always takes place to the opposite voltage direction vs it was forced before. If you want to feed that into cells on lower potential than the primary, then you have to isolate the secondaries.
Means you end up in a flyback.
And of course you can limit by taking energy from one cell and feeding into the series of all (or with multiple output windings into all cells except to the one which delivers ther energy).
If you implement for n cells the same amount of n flybacks then you can realize a strong and efficient limiter. But its quite a lot of efforts, just for balancing and limiting.
Trouble free Batteries:
LiFePo is not much trouble. Chargers with monitoring and/or balancers are no headache. Compared to LiPo they have no thermal runaway and are more safe.
Only if you want to use an existing lead acid charger in brute force technology, like the battery charging system in a normal car without modification for LiFePo - only then you need a balancer or limiter which can bypass quite some current.
I'm not seeing the inability of it to work on that account if the coil end connected to the MOSFET is pulled high so that its voltage can fly down. At those lower voltages, P-channel devices should work and make the circuit simpler so that gate supply bootstrapping isn't necessary. MOSFETs with low gate turn on threshold would help a lot. But this way or the flyback isolated secondary way are too much trouble for me. I like simplicity. So, I'm not building any cell balancers.
Aren't LifePO still very expensive? Is the supply of Lithium plentiful? Those are some of my concerns.
I might recheck into something like the A123 battery if I can use the present charge method that I use on the lead acid one, but with the voltage limit adjusted for the different cell potential. I don't want to deal with cell balancing, for sure. I hear that LiFePo don't flame up very easily, but I can't take even a small such risk in my house.
Aren't LifePO still very expensive? Is the supply of Lithium plentiful? Those are some of my concerns.
I might recheck into something like the A123 battery if I can use the present charge method that I use on the lead acid one, but with the voltage limit adjusted for the different cell potential. I don't want to deal with cell balancing, for sure. I hear that LiFePo don't flame up very easily, but I can't take even a small such risk in my house.
Your circuit proposal is not evident for me.
A schematic would be helpful for further discussion. Scan from a hand drawing is perfectly fine.
Pricewise they are of course more expensive than lead acid, but LiFePo is available and it is not generally unaffordable. It depends on the scale of your project if you can afford or not...
Lithium availability is currently not bad. I would say Lithium is available in large scale, but of course not unlimited.
Charging without balancer:
IMHO it should be OK for two cells in series with an voltage limit of 7.2V.
More than two cells in series? ... good question...., probably not fortunate.
For my private use, I am rating LiFePo for safe enough to be handled without much worries.
LiPo I am putting in a special safety bag during charging and storage.
BTW: Also lead acid batteries can cause fire.
If you 'cannot take even a such small risk' in your house, then almost everything may give you headache....
Wooden house, synthetic floor, dry climate?
A schematic would be helpful for further discussion. Scan from a hand drawing is perfectly fine.
Pricewise they are of course more expensive than lead acid, but LiFePo is available and it is not generally unaffordable. It depends on the scale of your project if you can afford or not...
Lithium availability is currently not bad. I would say Lithium is available in large scale, but of course not unlimited.
Charging without balancer:
IMHO it should be OK for two cells in series with an voltage limit of 7.2V.
More than two cells in series? ... good question...., probably not fortunate.
For my private use, I am rating LiFePo for safe enough to be handled without much worries.
LiPo I am putting in a special safety bag during charging and storage.
BTW: Also lead acid batteries can cause fire.
If you 'cannot take even a such small risk' in your house, then almost everything may give you headache....
Wooden house, synthetic floor, dry climate?
the reason i'm using the a123 cells is weight. i got a sportbike , lightweight is very important.
my stock battery weights 4.6kg , the a123 battery weights just 700 gram.
thats a huge saving ,and it costs less then stock battery.
see pic to see the size difference also.
this battery got less A/h capacity , but i need battery for just starting the bike , so its not a problem . it can put out 140A , my starter needs 100A .
my stock battery weights 4.6kg , the a123 battery weights just 700 gram.
thats a huge saving ,and it costs less then stock battery.
see pic to see the size difference also.
this battery got less A/h capacity , but i need battery for just starting the bike , so its not a problem . it can put out 140A , my starter needs 100A .
Attachments
as said , i will use micropower Vref , micropower op amp , and some mosfet to discharge the battery trough some resistor .
will do 4 circuits , each for 2 parallel cells.
my max charging voltageshould be 14.7V , so i will set the discharge on each cell @ 3.8V
i allready got all the parts from TI (all smt
) just need to make a circuit board and get some rwsistors .
will do 4 circuits , each for 2 parallel cells.
my max charging voltageshould be 14.7V , so i will set the discharge on each cell @ 3.8V
i allready got all the parts from TI (all smt
) just need to make a circuit board and get some rwsistors .ChocoHolic, thanks for your thoughts. I would probably do it like SSS is doing it by means of a linear circuit, anyway. That way, each cell just requires a shunt regulator which is pretty simple. But this thread reminds me that my lead acid battery probably could use some balancing after a lot of charge cycling. I think that I will put it on trickle charge for an extended period.
sss said:
Yupp, weight is also my concern.
My original battery has a weight of 3.8kg, but unfortunately my starter is drawing really 280A... and I already tried a 8 cell 2p4s pack, but as you can expect from theory: No way. Not at all.
You say 700g? Means you also have 8 cells of size 26650 in 2p4s connection?
For a starter current of approx 100A it should work fine, I fully agree.
BTW: Even if I am always pushing for propper balancing - in fact I know a guy who runs a 4s connection in his bike without balancer!! This is working now since one year without issues.
Means: Either he is a lucky hunk - or LiFePo is a quite forgiving technology.
Looking forward to hear about your experiences!
Hi SSS,
Also have a look at the LTC6802 chip. Does all you need (and a bit more)
FEATURES:
- Measures up to 12 Li-Ion Cells in Series (60V Max)
- Stackable Architecture Enables >1000V Systems
- 0.25% Maximum Total Measurement Error
- 13ms to Measure All Cells in a System
- Cell Balancing:
On-Chip Passive Cell Balancing Switches
Provision for Off-Chip Passive Balancing
- Two Thermistor Inputs Plus On-board
Temperature Sensor
- 1MHz Daisy-Chainable Serial Interface
- High EMI Immunity
- Delta Sigma Converter With Built In Noise Filter
- Open Wire Connection Fault Detection
- Low Power Modes
- 44-Lead SSOP Package
Regards,
Edmond.
Also have a look at the LTC6802 chip. Does all you need (and a bit more)
FEATURES:
- Measures up to 12 Li-Ion Cells in Series (60V Max)
- Stackable Architecture Enables >1000V Systems
- 0.25% Maximum Total Measurement Error
- 13ms to Measure All Cells in a System
- Cell Balancing:
On-Chip Passive Cell Balancing Switches
Provision for Off-Chip Passive Balancing
- Two Thermistor Inputs Plus On-board
Temperature Sensor
- 1MHz Daisy-Chainable Serial Interface
- High EMI Immunity
- Delta Sigma Converter With Built In Noise Filter
- Open Wire Connection Fault Detection
- Low Power Modes
- 44-Lead SSOP Package
Regards,
Edmond.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.