Loading ...
Sorry, an error occurred while loading the content.

Re: Time for new batteries

Expand Messages
  • bkelley9401
    Lithium cost too much with no warranty on life. The only drop-in for the robomower cost $600 for 19.2ah. These would be an exact replacement no wiring change.
    Message 1 of 14 , May 1, 2012
    • 0 Attachment
      Lithium cost too much with no warranty on life. The only drop-in for the robomower cost $600 for 19.2ah.
      These would be an exact replacement no wiring change.
      http://www.all-battery.com/Tenergy12.8V19.2AhLiFePO4RechargeableBattery-31943.aspx

      If you would use these you would need to change the wiring to parallel.
      http://www.all-battery.com/Tenergy25.6V9.6AhLiFePO4RechargeableBattery-31913.aspx

      Or build your own. The parts to do so are getting to be plug and pray, but who do you blame when it fails.


      > Or... lithium....
      >
      > I'm picking up a CleanPower BMS for my Robomower; since I plan to scale
      > up this solution over the summer I wanted to try a professional-ish BMS
      > first.
      >
      > I'm also fiddling with how to mount the batteries; I think the best move
      > will be to compress copper bars with plastic insulators, held under
      > compression. Still working on the details, but a Robomower that can run
      > for 6 hours and be lighter could be nice. Right now the 8 cells weigh a
      > lot less than a single 17ah battery....
      >
      > Chris
      >
    • Chris Zach
      ... Oh for crying in the toast; I could have just BOUGHT a drop in solution? DRAT! So far I am in $240 for the cells, another $140 for the BMS, and probably
      Message 2 of 14 , May 1, 2012
      • 0 Attachment
        On 5/1/2012 10:43 AM, bkelley9401 wrote:
        >
        >
        > Lithium cost too much with no warranty on life. The only drop-in for the
        > robomower cost $600 for 19.2ah.

        Oh for crying in the toast; I could have just BOUGHT a drop in solution?
        DRAT!

        So far I am in $240 for the cells, another $140 for the BMS, and
        probably another $100 for the little parts and such. And it will look
        like dirt compared to a solution like the one above.

        > Or build your own. The parts to do so are getting to be plug and pray,
        > but who do you blame when it fails.

        Well, you deal and move on. Maybe cry a little bit or something. But
        that's what we call progress.

        I can get 3+ years on a pack with the zener regs, most people get 1 year
        on a pack. I'm currently using Lithium batteries on my Scooba and even
        though it's twice the cost of a normal pack it has already lasted 3
        times as long as NiMH. Same with the Roomba and it's lithium pack.

        For $600 including BMS and all that is a nice deal. But I need to learn
        how to build these for my elecric tractor, then my electric car. Walk
        before run and all that.

        Chris
      • Lee Hart
        ... People like Valence will sell you a drop-in replacement for about $600. It includes a built-in BMS, and has a good warranty. ... Most consumer products
        Message 3 of 14 , May 1, 2012
        • 0 Attachment
          On 5/1/2012 10:31 AM, Chris Zach wrote:
          > On 5/1/2012 10:43 AM, bkelley9401 wrote:
          > Oh for crying in the toast; I could have just BOUGHT a drop in solution?
          > DRAT!

          People like Valence will sell you a drop-in replacement for about $600.
          It includes a built-in BMS, and has a good warranty.

          > I can get 3+ years on a pack with the zener regs, most people get 1 year
          > on a pack. I'm currently using Lithium batteries on my Scooba and even
          > though it's twice the cost of a normal pack it has already lasted 3
          > times as long as NiMH. Same with the Roomba and it's lithium pack.

          Most consumer products provide the cheapest possible batteries, charger,
          and (with lithium) the minimum possible BMS that's barely good enough to
          avoid getting sued for starting fires.

          My Robomower came with el-cheapo Chinese batteries, and a crude charger
          that murdered them in one season. My replacement Hawkers and quality
          electric wheelchair charger are 3 years old and going strong.

          > For $600 including BMS and all that is a nice deal. But I need to learn
          > how to build these for my elecric tractor, then my electric car. Walk
          > before run and all that.

          Yes. We do have to pay for learning. Either in tuition and books, or
          extra time, or by wrecking things.

          --
          If we knew what we were doing, it wouldn't be called research, would it?
          -- Albert Einstein
          --
          Lee A. Hart, 814 8th Ave N, Sartell MN 56377, leeahart at earthlink.net
        • Danny Miller
          Ironically, I m pretty darn sure those batteries would NOT be a safe option. There s no proper charge controller on there that I can see, so they re
          Message 4 of 14 , May 1, 2012
          • 0 Attachment
            Ironically, I'm pretty darn sure those batteries would NOT be a "safe"
            option. There's no proper charge controller on there that I can see, so
            they're literally slapping something together that a customer might
            want, and to hell with design prudence and safety.

            There's a lot of things you can CALL "a charge controller". But if it
            can't dissipate the heat of a normal charge rate, it's doomed to
            failure. And I can tell you, bad charge controllers are a FREQUENT
            complaint, and often involve total loss of the battery, if not a fire.

            It's not just about cost. There's design constrains involved with
            installing 8x 10W resistors. Or exposing 9 low-resistance terminals for
            an external controller. It's either a funky, awkward shape with a
            heatsink hanging off it that puts the buyer off, or "complex" enough to
            put the buyer off. So the market is more than willing to sell designs
            that are basically doomed to fail. And they DO.

            Danny

            On 5/1/2012 9:43 AM, bkelley9401 wrote:
            >
            >
            >
            >
            > Lithium cost too much with no warranty on life. The only drop-in for the robomower cost $600 for 19.2ah.
            > These would be an exact replacement no wiring change.
            > http://www.all-battery.com/Tenergy12.8V19.2AhLiFePO4RechargeableBattery-31943.aspx
            >
            > If you would use these you would need to change the wiring to parallel.
            > http://www.all-battery.com/Tenergy25.6V9.6AhLiFePO4RechargeableBattery-31913.aspx
            >
            > Or build your own. The parts to do so are getting to be plug and pray, but who do you blame when it fails.
            >
            >
            >> Or... lithium....
            >>
            >> I'm picking up a CleanPower BMS for my Robomower; since I plan to scale
            >> up this solution over the summer I wanted to try a professional-ish BMS
            >> first.
            >>
            >> I'm also fiddling with how to mount the batteries; I think the best move
            >> will be to compress copper bars with plastic insulators, held under
            >> compression. Still working on the details, but a Robomower that can run
            >> for 6 hours and be lighter could be nice. Right now the 8 cells weigh a
            >> lot less than a single 17ah battery....
            >>
            >> Chris
            >>
            >
            >
            >
            > ------------------------------------
            >
            > Yahoo! Groups Links
            >
            >
            >
            >
          • Danny Miller
            Well, you re an engineer too Lee. You tell me. If you have say a 3A charge current, an internal BMS , how s that gonna work? You shunt 3A at 3.65v, that s
            Message 5 of 14 , May 1, 2012
            • 0 Attachment
              Well, you're an engineer too Lee. You tell me. If you have say a 3A
              charge current, an internal "BMS", how's that gonna work?
              You shunt 3A at 3.65v, that's 11W, potentially on multiple cells.
              Where's that gonna go inside a plastic case?

              In theory you could have a series pass transistor so if any ONE cell
              started bypassing, it drops the charge current down to like 5% while
              shunting & balancing. That will work, but I'd be surprised if they did
              that. And such a battery might not work well in series with another
              battery, I'm not sure.

              There's also a thing where you need the shutoff transistor. But that's
              a mess because the transistor needs to shut off in the opposite
              direction, and a MOSFET with its intrinsic body diode won't do that.
              You'd need a series pair of either an NMOS-PMOS combination or an
              NMOS-reversed NMOS pair to be able to control current in both
              directions. Again, it's possible, sure. But I'd be surprised if they
              did that. I wouldn't believe it without seeing a teardown of what they
              did, I have very low faith in LiFePO4 vendors doing this.

              Danny

              On 5/1/2012 1:48 PM, Lee Hart wrote:
              > On 5/1/2012 10:31 AM, Chris Zach wrote:
              >> On 5/1/2012 10:43 AM, bkelley9401 wrote:
              >> Oh for crying in the toast; I could have just BOUGHT a drop in solution?
              >> DRAT!
              > People like Valence will sell you a drop-in replacement for about $600.
              > It includes a built-in BMS, and has a good warranty.
              >
              >> I can get 3+ years on a pack with the zener regs, most people get 1 year
              >> on a pack. I'm currently using Lithium batteries on my Scooba and even
              >> though it's twice the cost of a normal pack it has already lasted 3
              >> times as long as NiMH. Same with the Roomba and it's lithium pack.
              > Most consumer products provide the cheapest possible batteries, charger,
              > and (with lithium) the minimum possible BMS that's barely good enough to
              > avoid getting sued for starting fires.
              >
              > My Robomower came with el-cheapo Chinese batteries, and a crude charger
              > that murdered them in one season. My replacement Hawkers and quality
              > electric wheelchair charger are 3 years old and going strong.
              >
              >> For $600 including BMS and all that is a nice deal. But I need to learn
              >> how to build these for my elecric tractor, then my electric car. Walk
              >> before run and all that.
              > Yes. We do have to pay for learning. Either in tuition and books, or
              > extra time, or by wrecking things.
              >
            • Lee Hart
              ... I m an EE, and design battery chargers and management systems professionally. One of my clients was building a replacement lithium battery, intended for
              Message 6 of 14 , May 1, 2012
              • 0 Attachment
                On 5/1/2012 1:52 PM, Danny Miller wrote:
                > Ironically, I'm pretty darn sure those batteries would NOT be a "safe"
                > option. There's no proper charge controller on there that I can see, so
                > they're literally slapping something together that a customer might
                > want, and to hell with design prudence and safety.
                >
                > There's a lot of things you can CALL "a charge controller". But if it
                > can't dissipate the heat of a normal charge rate, it's doomed to
                > failure. And I can tell you, bad charge controllers are a FREQUENT
                > complaint, and often involve total loss of the battery, if not a fire.
                >
                > It's not just about cost. There's design constrains involved with
                > installing 8x 10W resistors. Or exposing 9 low-resistance terminals for
                > an external controller. It's either a funky, awkward shape with a
                > heatsink hanging off it that puts the buyer off, or "complex" enough to
                > put the buyer off. So the market is more than willing to sell designs
                > that are basically doomed to fail. And they DO.

                I'm an EE, and design battery chargers and management systems
                professionally. One of my clients was building a replacement lithium
                battery, intended for racers that wanted a super-light engine starting
                battery. The racers will pay "big bucks" for such batteries as they can
                easily take 10-50 pounds off the weight of their car or motorcycle.

                They first tried selling them with no BMS (just lots of BS). They soon
                had angry customers calling. Batteries would catch fire, sometimes
                destroying entire vehicles!

                I designed a BMS for them that made their lithium battery behave exactly
                the same as a lead-acid battery, and that would fail safe. It would quit
                working if you tried to run it dead, draw too much current, or
                overcharge it; but it wouldn't start a fire.

                Unfortunately, the BMS added cost, which reduced their sales.
                Competitors (mainly fly-by-night outfits selling junk) undercut them.
                They outsourced the design to India to cut costs. But then they had
                quality problems.

                What was their solution? They removed most of the safety features of the
                BMS to minimize cost, and then sold their batteries without labels and
                no guarantee to the fly-by-night resellers to market under their own
                names. :-(
                --
                If you would not be forgotten
                When your body's dead and rotten
                Then write of great deeds worth the reading
                Or do these great deeds, worth repeating.
                -- Ben Franklin, from Poor Richard's Almanac
                --
                Lee A. Hart, 814 8th Ave N, Sartell MN 56377, leeahart at earthlink.net
              • Dan Barclay
                Warnings also say: Do not let battery come in contact with water. Might be a problem in the rain? It s hard to know what they define as contact . Most of us
                Message 7 of 14 , May 1, 2012
                • 0 Attachment
                  Warnings also say:



                  Do not let battery come in contact with water.



                  Might be a problem in the rain? It's hard to know what they define as
                  "contact". Most of us have RL1000's that almost never come indoors. It
                  needs to withstand hard rain and wind.



                  In any case, $600 is a bit much. I assume that is for the pair!



                  Dan



                  From: RoboMower@yahoogroups.com [mailto:RoboMower@yahoogroups.com] On Behalf
                  Of Danny Miller
                  Sent: Tuesday, May 01, 2012 1:53 PM
                  To: RoboMower@yahoogroups.com
                  Subject: Re: [RoboMower] Re: Time for new batteries





                  Ironically, I'm pretty darn sure those batteries would NOT be a "safe"
                  option. There's no proper charge controller on there that I can see, so
                  they're literally slapping something together that a customer might
                  want, and to hell with design prudence and safety.

                  There's a lot of things you can CALL "a charge controller". But if it
                  can't dissipate the heat of a normal charge rate, it's doomed to
                  failure. And I can tell you, bad charge controllers are a FREQUENT
                  complaint, and often involve total loss of the battery, if not a fire.

                  It's not just about cost. There's design constrains involved with
                  installing 8x 10W resistors. Or exposing 9 low-resistance terminals for
                  an external controller. It's either a funky, awkward shape with a
                  heatsink hanging off it that puts the buyer off, or "complex" enough to
                  put the buyer off. So the market is more than willing to sell designs
                  that are basically doomed to fail. And they DO.

                  Danny

                  On 5/1/2012 9:43 AM, bkelley9401 wrote:
                  >
                  >
                  >
                  >
                  > Lithium cost too much with no warranty on life. The only drop-in for the
                  robomower cost $600 for 19.2ah.
                  > These would be an exact replacement no wiring change.
                  >
                  http://www.all-battery.com/Tenergy12.8V19.2AhLiFePO4RechargeableBattery-3194
                  3.aspx
                  >
                  > If you would use these you would need to change the wiring to parallel.
                  >
                  http://www.all-battery.com/Tenergy25.6V9.6AhLiFePO4RechargeableBattery-31913
                  .aspx
                  >
                  > Or build your own. The parts to do so are getting to be plug and pray, but
                  who do you blame when it fails.
                  >
                  >
                  >> Or... lithium....
                  >>
                  >> I'm picking up a CleanPower BMS for my Robomower; since I plan to scale
                  >> up this solution over the summer I wanted to try a professional-ish BMS
                  >> first.
                  >>
                  >> I'm also fiddling with how to mount the batteries; I think the best move
                  >> will be to compress copper bars with plastic insulators, held under
                  >> compression. Still working on the details, but a Robomower that can run
                  >> for 6 hours and be lighter could be nice. Right now the 8 cells weigh a
                  >> lot less than a single 17ah battery....
                  >>
                  >> Chris
                  >>
                  >
                  >
                  >
                  > ------------------------------------
                  >
                  > Yahoo! Groups Links
                  >
                  >
                  >
                  >





                  [Non-text portions of this message have been removed]
                • Danny Miller
                  The A123 20AH pouch cells are a far better deal. Then you CAN put a proper BMS on it. But some hacker fu knowledge is required. Danny
                  Message 8 of 14 , May 1, 2012
                  • 0 Attachment
                    The A123 20AH pouch cells are a far better deal. Then you CAN put a
                    proper BMS on it. But some hacker fu knowledge is required.

                    Danny

                    On 5/1/2012 3:26 PM, Dan Barclay wrote:
                    > Warnings also say:
                    >
                    >
                    >
                    > Do not let battery come in contact with water.
                    >
                    >
                    >
                    > Might be a problem in the rain? It's hard to know what they define as
                    > "contact". Most of us have RL1000's that almost never come indoors. It
                    > needs to withstand hard rain and wind.
                    >
                    >
                    >
                    > In any case, $600 is a bit much. I assume that is for the pair!
                    >
                    >
                    >
                    > Dan
                    >
                    >
                    >
                    > From: RoboMower@yahoogroups.com [mailto:RoboMower@yahoogroups.com] On Behalf
                    > Of Danny Miller
                    > Sent: Tuesday, May 01, 2012 1:53 PM
                    > To: RoboMower@yahoogroups.com
                    > Subject: Re: [RoboMower] Re: Time for new batteries
                    >
                    >
                    >
                    >
                    >
                    > Ironically, I'm pretty darn sure those batteries would NOT be a "safe"
                    > option. There's no proper charge controller on there that I can see, so
                    > they're literally slapping something together that a customer might
                    > want, and to hell with design prudence and safety.
                    >
                    > There's a lot of things you can CALL "a charge controller". But if it
                    > can't dissipate the heat of a normal charge rate, it's doomed to
                    > failure. And I can tell you, bad charge controllers are a FREQUENT
                    > complaint, and often involve total loss of the battery, if not a fire.
                    >
                    > It's not just about cost. There's design constrains involved with
                    > installing 8x 10W resistors. Or exposing 9 low-resistance terminals for
                    > an external controller. It's either a funky, awkward shape with a
                    > heatsink hanging off it that puts the buyer off, or "complex" enough to
                    > put the buyer off. So the market is more than willing to sell designs
                    > that are basically doomed to fail. And they DO.
                    >
                    > Danny
                    >
                    > On 5/1/2012 9:43 AM, bkelley9401 wrote:
                    >>
                    >>
                    >>
                    >> Lithium cost too much with no warranty on life. The only drop-in for the
                    > robomower cost $600 for 19.2ah.
                    >> These would be an exact replacement no wiring change.
                    >>
                    > http://www.all-battery.com/Tenergy12.8V19.2AhLiFePO4RechargeableBattery-3194
                    > 3.aspx
                    >> If you would use these you would need to change the wiring to parallel.
                    >>
                    > http://www.all-battery.com/Tenergy25.6V9.6AhLiFePO4RechargeableBattery-31913
                    > .aspx
                    >> Or build your own. The parts to do so are getting to be plug and pray, but
                    > who do you blame when it fails.
                    >>
                    >>> Or... lithium....
                    >>>
                    >>> I'm picking up a CleanPower BMS for my Robomower; since I plan to scale
                    >>> up this solution over the summer I wanted to try a professional-ish BMS
                    >>> first.
                    >>>
                    >>> I'm also fiddling with how to mount the batteries; I think the best move
                    >>> will be to compress copper bars with plastic insulators, held under
                    >>> compression. Still working on the details, but a Robomower that can run
                    >>> for 6 hours and be lighter could be nice. Right now the 8 cells weigh a
                    >>> lot less than a single 17ah battery....
                    >>>
                    >>> Chris
                    >>>
                    >>
                    >>
                    >> ------------------------------------
                    >>
                    >> Yahoo! Groups Links
                    >>
                    >>
                    >>
                    >>
                    >
                    >
                    >
                    >
                    > [Non-text portions of this message have been removed]
                    >
                    >
                    >
                    > ------------------------------------
                    >
                    > Yahoo! Groups Links
                    >
                    >
                    >
                    >
                  • Lee Hart
                    ... You re right. The cheapest (and thus most popular) BMS is a simple shunt regulator that bypasses excess charging current, and burns it up as heat. You
                    Message 9 of 14 , May 1, 2012
                    • 0 Attachment
                      On 5/1/2012 2:05 PM, Danny Miller wrote:
                      > Well, you're an engineer too Lee. You tell me. If you have say a 3A
                      > charge current, an internal "BMS", how's that gonna work?
                      > You shunt 3A at 3.65v, that's 11W, potentially on multiple cells.
                      > Where's that gonna go inside a plastic case?

                      You're right. The cheapest (and thus most popular) BMS is a simple shunt
                      regulator that bypasses excess charging current, and burns it up as
                      heat. You can't have very many cells doing this in a pack, or for very
                      long before something gets very hot. And we know that heat significantly
                      shortens the life of lithium cells.

                      > In theory you could have a series pass transistor so if any ONE cell
                      > started bypassing, it drops the charge current down to like 5% while
                      > shunting & balancing. That will work, but I'd be surprised if they did
                      > that.

                      That is the next step up in BMS systems. The individual shunt regulators
                      have some kind of buss. When a cell begins shunting current, the bus
                      signals the charger to cut back to a low current. It roughly doubles the
                      cost and number of wires to be installed on the BMS. You see them;
                      they're just less common.

                      It's a challenge to make it fail-safe and reliable. Many of these bus
                      systems send a signal on the buss when there's a problem. That means if
                      a wire breaks or a regulator fails, there is *no* signal to tell the
                      charger to cut back or stop. But if they make it send signals for "OK",
                      then the regulators are drawing current all the time from the cells to
                      run the buss. The cells can be run dead and destroyed just from the "OK"
                      power requirements of the BMS.

                      Reliability starts to become a problem because you have hundreds of
                      parts and connections in these BMS systems. When they get built with the
                      cheapest possible parts and wiring, you get a high failure rate.

                      > And such a battery might not work well in series with another
                      > battery, I'm not sure.

                      Most BMS ICs are designed for laptops and other applications where the
                      total pack voltage is no more than 24v. Over this, you need multiple
                      ICs, and some way for them to cooperate.

                      There are problems dealing with failure modes. For instance, a cell can
                      fail open. That puts the full pack voltage across its terminals. The
                      parts between that cell and its BMS gets "hit* with full pack voltage.
                      The 300v pack in an EV can put 300v across any input to the BMS!

                      > There's also a thing where you need the shutoff transistor. But that's
                      > a mess because the transistor needs to shut off in the opposite
                      > direction, and a MOSFET with its intrinsic body diode won't do that.
                      > You'd need a series pair of either an NMOS-PMOS combination or an
                      > NMOS-reversed NMOS pair to be able to control current in both
                      > directions.

                      Right again. The BMS I designed for the starting battery (see earlier
                      post) had two anti-series NMOS MOSFETs to act as a switch. It could
                      handle the high starting current (like 500a for 10 sec), and the smaller
                      steady state currents (like 50a for an hour, or 10a indefinitely).

                      But that was just a single 12v battery, so I could use 30v MOSFETs. If
                      you put more than two batteries in series, these would have to be be
                      changed to higher-voltage MOSFETs, which would significantly increase
                      the cost. For higher voltage setups, you need to use a contactor or
                      relay to disconnect the battery in case of trouble.

                      > I have very low faith in LiFePO4 vendors doing this.

                      Same here. I read all the engineering journals and data sheets and
                      patents on BMS systems. I also reverse-engineer a lot of stuff, just to
                      see what's being sold and how it works.

                      The overwhelming majority of stuff is cheap junk that is *barely* safe
                      enough for the seller to dodge lawsuits if something goes wrong. And,
                      there are no concessions for longer battery life. This is why it is rare
                      to find anything with lithium cells more than a few years old where the
                      cells aren't shot.

                      This attitude on the part of venders worries me about lithiums in
                      electric cars. Will the automakers do it "right"? Or, will they use the
                      cheapest possible system that will barely get past their lawyers? From
                      the automaker's point of view, a $10,000 battery pack that must be
                      replaced every few years is a *big* profit opportunity for them!

                      The scheme I like best is the one I use in my own Battery Balancers (
                      See http://www3.telus.net/nook/balancerland). It uses relays so a single
                      central controller can select any cell, measure its voltage, and charge
                      or discharge it as needed, using power from the pack as a whole. This
                      way, no power is lost in wasted heat, and you don't have dangerous
                      failure modes.
                      --
                      First they ignore you; then they mock you; then they fight you; then you
                      win.
                      -- Mahatma Gandhi
                      --
                      Lee A. Hart, 814 8th Ave N, Sartell MN 56377, leeahart at earthlink.net
                    • Danny Miller
                      Well I think the strategy they re using is assuming that there will only be a trivial imbalance in the cells, and the charger will be tapering off because
                      Message 10 of 14 , May 1, 2012
                      • 0 Attachment
                        Well I think the "strategy" they're using is assuming that there will
                        only be a trivial imbalance in the cells, and the charger will be
                        tapering off because the charge is nearly full.

                        So they're probably betting the shunt will take 10% of the full charge
                        current and/or only for a minute or two as it tops off. But:
                        a) there's no guarantee the balance will be "close" like that. It
                        depends on differences in cell leakage and time since last balancing.
                        It can be significant.
                        b) it's assuming the charger will taper off. Not all charging schemes
                        do, in fact "constant current" which doesn't taper is a common design.
                        An algorithm of "3A until voltage= 3.65v * 4 = 14.6v", then "shut off
                        entirely".
                        c) the mfg generally doesn't control the charger that's put on it.
                        Tenergy for example is making a rather bold claim that it works with any
                        12v lead-acid charger. Well WTF does that limit the possible situations
                        to? There's many different charger types.

                        With a series transistor putting it in trickle-balance mode, that
                        transistor could itself be burning a LOT of heat, assuming it doesn't
                        have an inductor and use switching mode. The situation gets more
                        serious when you try to put multiple batts in series, because the
                        voltage margin needed to limit the current gets higher.

                        The series transistor could conflict with a SLA charging algorithm.
                        3.45v * 3 cells + 3.65v * 1 cell= 14v, but the transistor tries to limit
                        current to 150mA to avoid overheating the shunts. In doing so it allows
                        the batt voltage to rise above 14.8v and the charger drops out of Bulk
                        charge and into Float, which could be anything below 100mA, it could be
                        10mA, which will take days and days to complete a balancing.

                        Danny

                        On 5/1/2012 4:05 PM, Lee Hart wrote:
                        > The overwhelming majority of stuff is cheap junk that is *barely* safe
                        > enough for the seller to dodge lawsuits if something goes wrong. And,
                        > there are no concessions for longer battery life. This is why it is
                        > rare to find anything with lithium cells more than a few years old
                        > where the cells aren't shot. This attitude on the part of venders
                        > worries me about lithiums in electric cars. Will the automakers do it
                        > "right"? Or, will they use the cheapest possible system that will
                        > barely get past their lawyers? From the automaker's point of view, a
                        > $10,000 battery pack that must be replaced every few years is a *big*
                        > profit opportunity for them! The scheme I like best is the one I use
                        > in my own Battery Balancers ( See
                        > http://www3.telus.net/nook/balancerland). It uses relays so a single
                        > central controller can select any cell, measure its voltage, and
                        > charge or discharge it as needed, using power from the pack as a
                        > whole. This way, no power is lost in wasted heat, and you don't have
                        > dangerous failure modes.
                      • Lee Hart
                        ... Exactly right. They assume all cells are good, new, and well matched. But what you ll find in an old laptop pack is that as the pack aged, one or more
                        Message 11 of 14 , May 1, 2012
                        • 0 Attachment
                          On 5/1/2012 4:54 PM, Danny Miller wrote:
                          > Well I think the "strategy" they're using is assuming that there will
                          > only be a trivial imbalance in the cells, and the charger will be
                          > tapering off because the charge is nearly full.
                          >
                          > So they're probably betting the shunt will take 10% of the full charge
                          > current and/or only for a minute or two as it tops off.

                          Exactly right. They assume all cells are good, new, and well matched.

                          But what you'll find in an old laptop pack is that as the pack aged, one
                          or more cells has drifted away from the others. The amount of current or
                          time needed to try to balance it exceeded some limit, and the BMS shut
                          it down. The back is "bad" and must be replaced, even if the only
                          problem is that one cell has 80% of the amphour capacity of the others.

                          > a) there's no guarantee the balance will be "close" like that. It
                          > depends on differences in cell leakage and time since last balancing.
                          > It can be significant.
                          > b) it's assuming the charger will taper off. Not all charging schemes
                          > do, in fact "constant current" which doesn't taper is a common design.
                          > An algorithm of "3A until voltage= 3.65v * 4 = 14.6v", then "shut off
                          > entirely".

                          All correct.

                          > c) the mfg generally doesn't control the charger that's put on it.

                          Most products with lithiums *do* provide the charger, for exactly that
                          reason.

                          > Tenergy for example is making a rather bold claim that it works with any
                          > 12v lead-acid charger. Well WTF does that limit the possible situations
                          > to? There's many different charger types.

                          Exactly. Many 12v lead-acid chargers will happily output 17v forever!
                          They are nothing but a transformer and rectifier, and the light-load
                          voltage is very high.

                          > With a series transistor putting it in trickle-balance mode, that
                          > transistor could itself be burning a LOT of heat, assuming it doesn't
                          > have an inductor and use switching mode. The situation gets more
                          > serious when you try to put multiple batts in series, because the
                          > voltage margin needed to limit the current gets higher.
                          >
                          > The series transistor could conflict with a SLA charging algorithm.
                          > 3.45v * 3 cells + 3.65v * 1 cell= 14v, but the transistor tries to limit
                          > current to 150mA to avoid overheating the shunts. In doing so it allows
                          > the batt voltage to rise above 14.8v and the charger drops out of Bulk
                          > charge and into Float, which could be anything below 100mA, it could be
                          > 10mA, which will take days and days to complete a balancing.

                          Like I said, this is why people sell these batteries on eBay through
                          anonymous "fronts" so they can disappear when things go wrong.
                          --
                          Ring the bells that still can ring
                          Forget your perfect offering
                          There is a crack in everything
                          That's how the light gets in.
                          -- Leonard Cohen, from "Anthem"
                          --
                          Lee A. Hart, 814 8th Ave N, Sartell MN 56377, leeahart at earthlink.net
                        Your message has been successfully submitted and would be delivered to recipients shortly.