These Li-Ion batteries have been involved in massive safety recalls over the last 6 years. From major companies like Sony, HP, Dell.
The recalls started in 2006, and extended to 2011. It has involved ballpark way over 4 MILLION Li-Ion batteries.
HP
HP Expands Recall of Notebook Computer Batteries Due to Fire Hazard
Burn baby burn: HP pays out $425,000 to prevent a disco (laptop) inferno -- Engadget
Dell
Dell Recall of Notebook Computer Batteries Due To Fire Hazard
Dell To Announce Massive Laptop Battery Recall | News & Opinion | PCMag.com
etc.
The well seasoned engineers at HP, Dell, Sony, that got burned here. To the tune of over 4 million batteries. These guys are not dumb. Field use and problems were not according to their predictions. "**it happens. Perhaps they depended on app notes or articles that were based on "lab" conditions.
Whatever.
I would not want to buy your device, or anything else you may design.
The recalls started in 2006, and extended to 2011. It has involved ballpark way over 4 MILLION Li-Ion batteries.
HP
HP Expands Recall of Notebook Computer Batteries Due to Fire Hazard
Burn baby burn: HP pays out $425,000 to prevent a disco (laptop) inferno -- Engadget
Dell
Dell Recall of Notebook Computer Batteries Due To Fire Hazard
Dell To Announce Massive Laptop Battery Recall | News & Opinion | PCMag.com
etc.
The well seasoned engineers at HP, Dell, Sony, that got burned here. To the tune of over 4 million batteries. These guys are not dumb. Field use and problems were not according to their predictions. "**it happens. Perhaps they depended on app notes or articles that were based on "lab" conditions.
Whatever.
I would not want to buy your device, or anything else you may design.
Last edited:
Li-Ion is a great battery tech. Not so great for DIY's. Yes you can safely use them but if you promote their use, people less and less capable will do it too, then you wish you had discouraged it because which battery type you use isn't a life or death situation until something volatile is added to the mix.
I'm all for people making their own choices, but most DIY beginners are going to be around other people too, not just barricaded in some fireproof bunker. Respect the limits of a design effort. When dealing with explosive tech, an order of magnitude more attention is required.
Main point is, any time there is a need to discuss Li_Ion, the default response should be avoid until the asker has enough isolated experience to not ask. It is literally playing with fire otherwise.
I'm all for people making their own choices, but most DIY beginners are going to be around other people too, not just barricaded in some fireproof bunker. Respect the limits of a design effort. When dealing with explosive tech, an order of magnitude more attention is required.
Main point is, any time there is a need to discuss Li_Ion, the default response should be avoid until the asker has enough isolated experience to not ask. It is literally playing with fire otherwise.
I would consider a Vacuum tube to be much more stable device compared to a LiPo battery.
Tubes are, at times being used at voltages over their factory spec, with the result sometimes being reduced product lifetime.
Vacuum tubes can been run at lower voltages, the effect being increased distortion etc. which may not be a bad thing depending on the desired result.
A lithium battery on the other hand...
will not tolerate over voltage without catastrophic failure.
A lithium battery will be permanently damaged by allowing the voltage to go too low.
Tubes are, at times being used at voltages over their factory spec, with the result sometimes being reduced product lifetime.
Vacuum tubes can been run at lower voltages, the effect being increased distortion etc. which may not be a bad thing depending on the desired result.
A lithium battery on the other hand...
will not tolerate over voltage without catastrophic failure.
A lithium battery will be permanently damaged by allowing the voltage to go too low.
When you play with AC mains the odds are you'll only electrocute yourself. That's fair. When you have a bomb sitting around the odds are higher someone else is affected. Headamps, laptops, cell phones, etc. all *like* to travel in cars, on planes, turning a small fire bomb into a larger problem.
If you absolutely must do this...
- Add a circuit which measures the battery voltage and shuts off the load if the battery voltage gets too low.
- Set your charge voltage setpoint low (say, at 3.8 volts/cell) instead of going for 4.1 or 4.2 volts.
But honestly, change battery chemistry to SLA or NiMH. You'll pay less for batteries and it's far less likely to blow up on you.
The OP asked about "lithium" batteries. There are several types of lithium each with its own requirements.
Most laptops and cell phones use older lithium ion (LiIon) technology. It is a relatively stable chemistry that can become unstable, catch fire, or explode if mistreated. It is reccomended that the cells not be drained too far or over charged and the charge / discharge rate not be exceeded. The individual cells are usually encased in metal housings and can explode if severely mistreated. They can, and usually will catch fire if punctured.
The model plane and hellicopter industry has adopted the lithium polymer (LiPo) technology since it offers the best energy density and energy to weight ratio of the lithium family. The cells are encased in a thin plastic wrapper. LiPo cells are very dangerous, must not be discharged below a point, must not be charged continuously, or overcharged and must not be punctured. They WILL burst into flames if these conditions are violated, and the reaction leading to fire can be delayed. I don't claim to fully understand battery chemistry but it appears that violating the charge / discharge conditions of these cells can cause irreversible damage to the chemistry that makes it unstable. This can cause localized heating leading to further damage and eventual chain reaction causing cell failure. When multiple cells are installed in a pack there must be a method of equalizing the charge and monitoring the discharge to stop the discharge cycle when the weakest cell in the pack reached end of cycle. Specialized chargers are used for these packs. There is ample information on the web including several YouTube videos of what happens when you get this wrong. Model airplane enthusiasts have lost their planes, their cars, SUV's and even a few houses due to LiPo batteries. I use them in a portable tube amp, a computer, and soon a powered guitar. I use a commercially built charger and a microprocessor in the device to shut down the discharge. I charge the battery outside of the device, and use an old ammo box to hold the battery during charge. The batteries usually fail during charge, or if physically damaged, but random failures without reason have happened.
The newest lithium cells to hit the hobby market is lithium iron polymer (LiFePo) technology. These seem to have the best of both worlds. The cells are encased in metal, and are much less unstable than LiPo. They are becomming availabel to the hobbiest, but are quite expensive at this time. The charging requirements are similar to lithium ion cells.
Remember, you ARE playing with fire. Design your device accordingly. If you are using LiPo, you must play by the rules.
Most laptops and cell phones use older lithium ion (LiIon) technology. It is a relatively stable chemistry that can become unstable, catch fire, or explode if mistreated. It is reccomended that the cells not be drained too far or over charged and the charge / discharge rate not be exceeded. The individual cells are usually encased in metal housings and can explode if severely mistreated. They can, and usually will catch fire if punctured.
The model plane and hellicopter industry has adopted the lithium polymer (LiPo) technology since it offers the best energy density and energy to weight ratio of the lithium family. The cells are encased in a thin plastic wrapper. LiPo cells are very dangerous, must not be discharged below a point, must not be charged continuously, or overcharged and must not be punctured. They WILL burst into flames if these conditions are violated, and the reaction leading to fire can be delayed. I don't claim to fully understand battery chemistry but it appears that violating the charge / discharge conditions of these cells can cause irreversible damage to the chemistry that makes it unstable. This can cause localized heating leading to further damage and eventual chain reaction causing cell failure. When multiple cells are installed in a pack there must be a method of equalizing the charge and monitoring the discharge to stop the discharge cycle when the weakest cell in the pack reached end of cycle. Specialized chargers are used for these packs. There is ample information on the web including several YouTube videos of what happens when you get this wrong. Model airplane enthusiasts have lost their planes, their cars, SUV's and even a few houses due to LiPo batteries. I use them in a portable tube amp, a computer, and soon a powered guitar. I use a commercially built charger and a microprocessor in the device to shut down the discharge. I charge the battery outside of the device, and use an old ammo box to hold the battery during charge. The batteries usually fail during charge, or if physically damaged, but random failures without reason have happened.
The newest lithium cells to hit the hobby market is lithium iron polymer (LiFePo) technology. These seem to have the best of both worlds. The cells are encased in metal, and are much less unstable than LiPo. They are becomming availabel to the hobbiest, but are quite expensive at this time. The charging requirements are similar to lithium ion cells.
Remember, you ARE playing with fire. Design your device accordingly. If you are using LiPo, you must play by the rules.
Nicely said.
I will add this, everyone here has had to cut corners on occasion. For most of us we have had others around to keep us in hand and out of trouble when we make a stupid mistake.
When asked a question, we have all expressed an opinion and I believe that most have agreed that a disconnect is the way you want to go and that float charging might not be the best method.
The ultimate decision resides with the person who has to implement the design for the client. All we can do is offer an informed opinion which is the whole reason for this forum.
Tony
I spend my days as a battery engineer for a computer OEM. My focus is Li-Ion battery safety and quality so I have a little experience in this area.
I want to clear up a few misconceptions I've seen in this thread. I'll try to keep it technical without getting too detailed.
As most have said if you mistreat a Li-Ion battery, it will either degrade quicker than normal, or become unsafe. There is a big difference between the Tier 1 battery suppliers, and the cells available to most DIYers. The ones available to most DIYers overestimate capacity and specs, and have more variability cell to cell. Considering this I would treat these cells with more caution than from the major suppliers.
There has been some good advice on this thread. So I'll say some of it again with some supporting evidence.
A few things to keep in mind:
As the battery university website stated, Li-Ion is different than most battery chemistries. There aren't big changes in the anode and cathode like Lead and Nickel based chemistries. They did a bad job of explaining it, but even if it's only Lithium Ions shuttling back and forth, there are a lot of reactions that will cause a battery to degrade over time. Getting into shuttle reactions, oxidation, SEI layers, electrolyte decomposition, really just clouds the issue at hand.
Charge Rate - You need to keep the charge rate below the spec. The anode of the battery can only accept charge at a certain rate. When you go above this limit you will plate lithium onto the anode and this can make the cell much more volatile.
Float charging - Keeping a cell fully charged will cause it to degrade much faster than if you terminate charge and allow the voltage to relax. This can also also cause a condition where the cells will develop internal shorts, and cells can get quite hot during charging because the current does not taper down.
Over-discharge - When a cell is deeply discharged the copper anode current collector will actually dissolve into the electrolyte, and when recharged the copper will re-plate in other places. This can over time create internal shorts.
These are just some of the reasons you want a proper charge algorithm that holds the right charge voltage, and terminates charge when the current tapers below a threshold. Balancing between cells isn't 100% necessary, but you do need to make sure you are monitoring each cell, or somehow are making certain you keep all the cells in the proper operating range.
If I was designing a product for sale I would make sure I started with good cells. Look into battery distributors, not dealextreme or ebay. The best circuitry and protection won't stop something bad happening if the cells are terrible. Then I would make sure the battery pack had over-temperature, over-voltage, over-current, and over-discharge. I've seen some "protected" cells that were not a whole lot better than an unprotected one. There are some ICs that allow decent protection and a pretty simple circuit.
Then make sure the charge circuit has the proper voltage and current, so the protections in the pack aren't tripped unless there is a problem.
Iron phosphte cells (LifePO4) are much more tolerant to conditions, but still need to be taken good care of. The downside to them is the energy density is much lower than Li-Ion.
I'm happy to answer more detailed questions if you would like, even offer some suggestions for ICs to look at.
I want to clear up a few misconceptions I've seen in this thread. I'll try to keep it technical without getting too detailed.
As most have said if you mistreat a Li-Ion battery, it will either degrade quicker than normal, or become unsafe. There is a big difference between the Tier 1 battery suppliers, and the cells available to most DIYers. The ones available to most DIYers overestimate capacity and specs, and have more variability cell to cell. Considering this I would treat these cells with more caution than from the major suppliers.
There has been some good advice on this thread. So I'll say some of it again with some supporting evidence.
A few things to keep in mind:
As the battery university website stated, Li-Ion is different than most battery chemistries. There aren't big changes in the anode and cathode like Lead and Nickel based chemistries. They did a bad job of explaining it, but even if it's only Lithium Ions shuttling back and forth, there are a lot of reactions that will cause a battery to degrade over time. Getting into shuttle reactions, oxidation, SEI layers, electrolyte decomposition, really just clouds the issue at hand.
Charge Rate - You need to keep the charge rate below the spec. The anode of the battery can only accept charge at a certain rate. When you go above this limit you will plate lithium onto the anode and this can make the cell much more volatile.
Float charging - Keeping a cell fully charged will cause it to degrade much faster than if you terminate charge and allow the voltage to relax. This can also also cause a condition where the cells will develop internal shorts, and cells can get quite hot during charging because the current does not taper down.
Over-discharge - When a cell is deeply discharged the copper anode current collector will actually dissolve into the electrolyte, and when recharged the copper will re-plate in other places. This can over time create internal shorts.
These are just some of the reasons you want a proper charge algorithm that holds the right charge voltage, and terminates charge when the current tapers below a threshold. Balancing between cells isn't 100% necessary, but you do need to make sure you are monitoring each cell, or somehow are making certain you keep all the cells in the proper operating range.
If I was designing a product for sale I would make sure I started with good cells. Look into battery distributors, not dealextreme or ebay. The best circuitry and protection won't stop something bad happening if the cells are terrible. Then I would make sure the battery pack had over-temperature, over-voltage, over-current, and over-discharge. I've seen some "protected" cells that were not a whole lot better than an unprotected one. There are some ICs that allow decent protection and a pretty simple circuit.
Then make sure the charge circuit has the proper voltage and current, so the protections in the pack aren't tripped unless there is a problem.
Iron phosphte cells (LifePO4) are much more tolerant to conditions, but still need to be taken good care of. The downside to them is the energy density is much lower than Li-Ion.
I'm happy to answer more detailed questions if you would like, even offer some suggestions for ICs to look at.
Lithium-ion Battery Protection ICs - Seiko Instruments Inc.
Li-ion Protector IC / Electronic Devices | Ricoh Global
Those are 2 links for what I would consider to be the bare minimum for a battery protection circuit. Completely analog, a handful of comparators and some delay blocks to cover the protections.
Parts like those are used as the backup to a microcontroller in most laptop battery packs. So if the firmware has an issue you still have the analog section to protect you.
Those parts won't do charge control, so you still need a good circuit to do the charging. But again there are off the shelf parts that will do everything you need in fairly simple circuits.
Li-ion Protector IC / Electronic Devices | Ricoh Global
Those are 2 links for what I would consider to be the bare minimum for a battery protection circuit. Completely analog, a handful of comparators and some delay blocks to cover the protections.
Parts like those are used as the backup to a microcontroller in most laptop battery packs. So if the firmware has an issue you still have the analog section to protect you.
Those parts won't do charge control, so you still need a good circuit to do the charging. But again there are off the shelf parts that will do everything you need in fairly simple circuits.
JC2;
You are a man after my own heart. I like backup fault protection schemes. I like fast trips to be comparator based with a MCU backup and for slow moving events (temperature compensating the battery topoff setpoints and delay timers for shutdown were handled by the MCU. I used a thermal PTC switch under the battery to detect a serious fault. This was ran to a latching relay and was diode or'ed so that the MCU or the PTC would open the relay. Nothing I can do with the battery itself, except making sure the enclosure is designed as a fire enclosure and is self extinguishing. For SLA's I even went so far as to calculate the minimum enclosure size so that a hydrogen out-gassing would be inherantly safe (hydrogen level <10% if I remember correctly).
A calculated the total additional costs of the protection scheme was about $3USD.
Tony
You are a man after my own heart. I like backup fault protection schemes. I like fast trips to be comparator based with a MCU backup and for slow moving events (temperature compensating the battery topoff setpoints and delay timers for shutdown were handled by the MCU. I used a thermal PTC switch under the battery to detect a serious fault. This was ran to a latching relay and was diode or'ed so that the MCU or the PTC would open the relay. Nothing I can do with the battery itself, except making sure the enclosure is designed as a fire enclosure and is self extinguishing. For SLA's I even went so far as to calculate the minimum enclosure size so that a hydrogen out-gassing would be inherantly safe (hydrogen level <10% if I remember correctly).
A calculated the total additional costs of the protection scheme was about $3USD.
Tony
Good posts jc2, thank you.
Any recommendations on a smart LifePo4 charging scheme with protection and hassle-free charging?
I already have a "smart" charger with protection, balancing etc but I would very much like to be able to leave the charger always on, have over and undervoltage cut-off protection ranges and, most importantly, a scheme that would automatically disconnect the charger from batteries when the load draws current and switch back to charging when the load is switched off.
Any recommendations on a smart LifePo4 charging scheme with protection and hassle-free charging?
I already have a "smart" charger with protection, balancing etc but I would very much like to be able to leave the charger always on, have over and undervoltage cut-off protection ranges and, most importantly, a scheme that would automatically disconnect the charger from batteries when the load draws current and switch back to charging when the load is switched off.
Sorry, I don't really have any specific recommendations for that. The devices I work with are using large Power Management ASICs that handle all that. The devices I've done outside of work I used microcontrollers to control the charging. I have looked briefly at some charging circuits from TI for a project that never happened, and it looked like some were pretty well sefl contained. I'm sure they or some other vendor has parts designed for LiFePO4 batteries since they are used all over these days.
he TI part I mentioned earlier has the feature sets I would target.
http://www.ti.com/lit/ds/slus606o/slus606o.pdf
This is not an endorsement of this particular part since I haven't used it, but it at least has most of what I would be looking for in a single device.
Tony
http://www.ti.com/lit/ds/slus606o/slus606o.pdf
This is not an endorsement of this particular part since I haven't used it, but it at least has most of what I would be looking for in a single device.
Tony
TheShaman;
The MCP part doesn't appear to have a have a thermistor input for a temperature compensated CC/CV charge rate. It is also a linear part. Power dissipation in the IC could be an issue when charging a deeply discharged device.
Anyway... that's my $0.02. Take it for what it cost you.
Tony
The MCP part doesn't appear to have a have a thermistor input for a temperature compensated CC/CV charge rate. It is also a linear part. Power dissipation in the IC could be an issue when charging a deeply discharged device.
Anyway... that's my $0.02. Take it for what it cost you.
Tony
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.