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

Re: [RoboMower] Switching to lithiums for RL500

Expand Messages
  • Danny Miller
    I would strongly caution you against building a large pack like that out of cobalt-cathode 18650s. That s just dangerous. Among 90 cells, the failure of a
    Message 1 of 8 , Jan 4, 2013
    • 0 Attachment
      I would strongly caution you against building a large pack like that out
      of cobalt-cathode 18650s. That's just dangerous. Among 90 cells, the
      failure of a single cell is high, and a single-cell failure can
      definitely ignite the entire pack. Lithium packs burn hot and regular
      fire extinguishers may not be appropriate for burning lithium.

      Yes you can certainly get superior capacity over lead-acid, esp if you
      DON'T weight it back down. The lighter mower requires substantially
      less power to the drive motors. But you do need to consider the overall
      layout and space-efficiency because the volumetric density is only
      moderately better but the ability to utilize the volume is always much less.

      Volumetrically, when you're trying to parallel small cells into
      macrocells, there's a problem where you STILL want a macrocell to be
      rectangular for practical reasons. There's no law against making
      nonrectangular macrocells but it's just awkward to assemble Tetris-piece
      shapes into a pack.

      Danny

      On 1/4/2013 2:41 PM, wolfbob wrote:
      > I don't understand the problem, Danny I
      >
      > took a picture showing all the 18650s I have (90) only filling about 60% of the RL500 battery compartment. Even with the charger, BMS and some lead weights in there it should be no problem to get 24 volts 20Ah. The SLAs as you point out only furnish around 10Ah out of the 17 Ah rating and have any life.
      >
      > Gotta run now, but I can post the picture when I figure out how..
      >
      > WBob
      >
      > ----- Original Message -----
      > From: Danny Miller
      > To: RoboMower@yahoogroups.com
      > Sent: Friday, January 04, 2013 11:36 AM
      > Subject: Re: [RoboMower] Switching to lithiums for RL500
      >
      >
      >
      > Unfortunately the question is now of VOLUMETRIC energy density. And you
      > have a problem.
      >
      > The 12v 20ah half-U1 lead acid batts you're replacing are 1.72166
      > watt-hr/cu in.
      >
      > Now a large-format rectangular lifepo4 cell can be 2.1 watt-hr/cu in.
      > The 26650 2.3AH A123's come to 1.44 wh/cu in.
      >
      > The problem is the old case has almost 100% utilization of the volume
      > inside the pack. Cylindrical cells have considerable airspace AND will
      > have a remainder of space on the sides because the sizes won't line up
      > just right with the volume.
      >
      > Rectangular cells will likely not be able to use all the space either.
      > And you have to have exactly 8 cells, not 7, not 9. You can't just add
      > as many cells as the pack can hold, if you can only fit 7 with the "big"
      > size a mfg has, so you select a much smaller one and you see you can
      > stack 8 but there's still room for 2 on the side and one flat across the
      > top, you will still have to just use 8 and ignore the extra volume.
      >
      > It's just very likely you're going to have a significant volume problem
      > with any arbitrary LiFePO4 shape (but rectangulars will usually make
      > much better use of the space).
      > So rectangulars are 22% denser by volume, but having a case that's only
      > 2/3rd battery and 1/3rd air would be a pretty lucky fit the way I see
      > it. It will be less dense. Of course lead-acid always had trouble
      > actually making the rated capacity repeatedly so the energy density for
      > lead-acid was a bit optimistic.
      >
      > A123 LiFePO4 is 3.3v per cell not 3.6v, you need 8 for 24v. That's
      > 26.4v, or 13.2v per "batt" on the old lead-acid pair, a LITTLE high but
      > that's the voltage you see when it's on the charger. Most other mfgs
      > rate their cells at 3.2v though.
      >
      > For rectangulars go to Thundersky/Winston batt, or better yet CALB batt.
      >
      > I don't know of an A123 "3AH" cell. The common 26650 A123 is 2.3AH.
      > They do make a AHR32113 cell which is 4.4AH in their gen-2 version of
      > that cell, but those would be hard to obtain anyways, they're not sold
      > on ebay or online or anything.
      >
      > Danny
      >
      > On 1/4/2013 9:30 AM, Christopher Zach wrote:
      > > Last year I tried running a set of 20ah A123 pouch cells in my RL500
      > > with a BMS on them. Not only did they work, the biggest problem was that
      > > I couldn't run them down. Mower just kept going and going....
      > >
      > > It convinced me that Lithium is a way to go, keeping one old dead lead
      > > battery in there as ballast. However the A123 cells are too tall and
      > > have *way* too much power, so I'm going to try something else.
      > >
      > > Anyone thought about building a pack of smaller 3ah A123 cylinder cells
      > > in parallel? A 17ah lead pack will put out only around 12ah under load
      > > tops, so 4 strings of 3ah cells would do it nicely. 3.6 volt cells would
      > > allow a 7 cell string to provide 24+ volts, 8 cells would make the
      > > motors *sing* at 28 volts but might be too big to charge with the
      > > on-board charger.
      > >
      > > I'm going to start work this winter, to have something for March/April.
      > >
      > > C
      > >
      > >
      > > ------------------------------------
      > >
      > > Yahoo! Groups Links
      > >
      > >
      > >
      > >
      >
      >
      >
      >
      >
      > [Non-text portions of this message have been removed]
      >
      >
      >
      > ------------------------------------
      >
      > Yahoo! Groups Links
      >
      >
      >
      >
    • wolfbob
      Back for a moment... The LiIon have metal cases instead of the easily punctured and easily shorted LiFe etc. cells and are the one that fills all of those 50
      Message 2 of 8 , Jan 4, 2013
      • 0 Attachment
        Back for a moment...

        The LiIon have metal cases instead of the easily punctured and easily shorted LiFe etc. cells and are the one that fills all of those 50 million laptops. There is no Lithium metal in LiIon cells just some compound of Lithium that takes a lot to ignite. Read "Battery University"...It is by far the safest of the Lithium cells. Unlike the LiFe cells that the manufacturer warns you to never charge them without physical protection and never leave them unattended. My helicopter manufacturer (China) specifically warns about charging the LiFe batteries near anything valuable like people.

        By all means rectangular are more space efficient but not very good for cooling as the middle cells have no exposure. The 18650s get all of 10 deg F temp rise at 1.5A, hardly warm.

        Australia has banned individuals from shipping any Lithium batteries in their post. You can still ship them but you must use commercial carriers. I have several customers working on trying to educate the POLs who made such an arbitrary ruling.

        WBob


        ----- Original Message -----
        From: Danny Miller
        To: RoboMower@yahoogroups.com
        Sent: Friday, January 04, 2013 1:20 PM
        Subject: Re: [RoboMower] Switching to lithiums for RL500



        I would strongly caution you against building a large pack like that out
        of cobalt-cathode 18650s. That's just dangerous. Among 90 cells, the
        failure of a single cell is high, and a single-cell failure can
        definitely ignite the entire pack. Lithium packs burn hot and regular
        fire extinguishers may not be appropriate for burning lithium.

        Yes you can certainly get superior capacity over lead-acid, esp if you
        DON'T weight it back down. The lighter mower requires substantially
        less power to the drive motors. But you do need to consider the overall
        layout and space-efficiency because the volumetric density is only
        moderately better but the ability to utilize the volume is always much less.

        Volumetrically, when you're trying to parallel small cells into
        macrocells, there's a problem where you STILL want a macrocell to be
        rectangular for practical reasons. There's no law against making
        nonrectangular macrocells but it's just awkward to assemble Tetris-piece
        shapes into a pack.

        Danny

        On 1/4/2013 2:41 PM, wolfbob wrote:
        > I don't understand the problem, Danny I
        >
        > took a picture showing all the 18650s I have (90) only filling about 60% of the RL500 battery compartment. Even with the charger, BMS and some lead weights in there it should be no problem to get 24 volts 20Ah. The SLAs as you point out only furnish around 10Ah out of the 17 Ah rating and have any life.
        >
        > Gotta run now, but I can post the picture when I figure out how..
        >
        > WBob
        >
        > ----- Original Message -----
        > From: Danny Miller
        > To: RoboMower@yahoogroups.com
        > Sent: Friday, January 04, 2013 11:36 AM
        > Subject: Re: [RoboMower] Switching to lithiums for RL500
        >
        >
        >
        > Unfortunately the question is now of VOLUMETRIC energy density. And you
        > have a problem.
        >
        > The 12v 20ah half-U1 lead acid batts you're replacing are 1.72166
        > watt-hr/cu in.
        >
        > Now a large-format rectangular lifepo4 cell can be 2.1 watt-hr/cu in.
        > The 26650 2.3AH A123's come to 1.44 wh/cu in.
        >
        > The problem is the old case has almost 100% utilization of the volume
        > inside the pack. Cylindrical cells have considerable airspace AND will
        > have a remainder of space on the sides because the sizes won't line up
        > just right with the volume.
        >
        > Rectangular cells will likely not be able to use all the space either.
        > And you have to have exactly 8 cells, not 7, not 9. You can't just add
        > as many cells as the pack can hold, if you can only fit 7 with the "big"
        > size a mfg has, so you select a much smaller one and you see you can
        > stack 8 but there's still room for 2 on the side and one flat across the
        > top, you will still have to just use 8 and ignore the extra volume.
        >
        > It's just very likely you're going to have a significant volume problem
        > with any arbitrary LiFePO4 shape (but rectangulars will usually make
        > much better use of the space).
        > So rectangulars are 22% denser by volume, but having a case that's only
        > 2/3rd battery and 1/3rd air would be a pretty lucky fit the way I see
        > it. It will be less dense. Of course lead-acid always had trouble
        > actually making the rated capacity repeatedly so the energy density for
        > lead-acid was a bit optimistic.
        >
        > A123 LiFePO4 is 3.3v per cell not 3.6v, you need 8 for 24v. That's
        > 26.4v, or 13.2v per "batt" on the old lead-acid pair, a LITTLE high but
        > that's the voltage you see when it's on the charger. Most other mfgs
        > rate their cells at 3.2v though.
        >
        > For rectangulars go to Thundersky/Winston batt, or better yet CALB batt.
        >
        > I don't know of an A123 "3AH" cell. The common 26650 A123 is 2.3AH.
        > They do make a AHR32113 cell which is 4.4AH in their gen-2 version of
        > that cell, but those would be hard to obtain anyways, they're not sold
        > on ebay or online or anything.
        >
        > Danny
        >
        > On 1/4/2013 9:30 AM, Christopher Zach wrote:
        > > Last year I tried running a set of 20ah A123 pouch cells in my RL500
        > > with a BMS on them. Not only did they work, the biggest problem was that
        > > I couldn't run them down. Mower just kept going and going....
        > >
        > > It convinced me that Lithium is a way to go, keeping one old dead lead
        > > battery in there as ballast. However the A123 cells are too tall and
        > > have *way* too much power, so I'm going to try something else.
        > >
        > > Anyone thought about building a pack of smaller 3ah A123 cylinder cells
        > > in parallel? A 17ah lead pack will put out only around 12ah under load
        > > tops, so 4 strings of 3ah cells would do it nicely. 3.6 volt cells would
        > > allow a 7 cell string to provide 24+ volts, 8 cells would make the
        > > motors *sing* at 28 volts but might be too big to charge with the
        > > on-board charger.
        > >
        > > I'm going to start work this winter, to have something for March/April.
        > >
        > > C
        > >
        > >
        > > ------------------------------------
        > >
        > > Yahoo! Groups Links
        > >
        > >
        > >
        > >
        >
        >
        >
        >
        >
        > [Non-text portions of this message have been removed]
        >
        >
        >
        > ------------------------------------
        >
        > Yahoo! Groups Links
        >
        >
        >
        >





        [Non-text portions of this message have been removed]
      • Danny Miller
        You ve got many misconceptions, and you misread something somewhere. You ve stated most things backwards from fact. Li-ion, the 18650 type, most of those are
        Message 3 of 8 , Jan 4, 2013
        • 0 Attachment
          You've got many misconceptions, and you misread something somewhere.
          You've stated most things backwards from fact.

          Li-ion, the 18650 type, most of those are li-cobalt (they make a
          lithium-manganese-spinel type which is safer but 1/2 the capacity in the
          18650 case), are THE most dangerous cell. Hands down. They've got the
          highest initial storage density and the best cost-to-capacity ratio but
          don't easily safely scale up to large batts.

          LiFePO4 is not commonly used in laptops. Lithium cobalt or
          manganese-spinel are.

          Helicopter cells are li-polymer which is physically very similar to
          li-ion, and yes they're risky, but much of that comes from the very
          aggressive charging cycles in RC. They are flammable and do explode
          (pop) if overcharged.

          Li-cobalt is very very flammable, the metal case doesn't protect much
          against that. If there is a problem in the charging system where they
          get overcharged, they plate out metallic lithium in the anode and
          basically turn it into a little firebomb. And they're very vulnerable
          to catching fire when heated by another burning cell (thermal runaway).

          LiFePO4 doesn't "burn", but its energy will turn into heat if the
          separator fails. The don't react nearly as badly when overcharged, and
          they're much more resistant to thermal runaway (melting down because
          they were exposed to the heat of an adjacent batt melting).

          LiFePO4 is widely recognized as the safer chemistry by far. But that's
          not to say they're totally safe and could never catch fire. I don't
          know how you got the idea that the li-ion cobalt type was safer, but
          you're very much mistaken. No battery expert is going to back you on that.

          Danny


          On 1/4/2013 6:00 PM, wolfbob wrote:
          > Back for a moment...
          >
          > The LiIon have metal cases instead of the easily punctured and easily shorted LiFe etc. cells and are the one that fills all of those 50 million laptops. There is no Lithium metal in LiIon cells just some compound of Lithium that takes a lot to ignite. Read "Battery University"...It is by far the safest of the Lithium cells. Unlike the LiFe cells that the manufacturer warns you to never charge them without physical protection and never leave them unattended. My helicopter manufacturer (China) specifically warns about charging the LiFe batteries near anything valuable like people.
          >
          > By all means rectangular are more space efficient but not very good for cooling as the middle cells have no exposure. The 18650s get all of 10 deg F temp rise at 1.5A, hardly warm.
          >
          > Australia has banned individuals from shipping any Lithium batteries in their post. You can still ship them but you must use commercial carriers. I have several customers working on trying to educate the POLs who made such an arbitrary ruling.
          >
          > WBob
          >
          >
          > ----- Original Message -----
          > From: Danny Miller
          > To: RoboMower@yahoogroups.com
          > Sent: Friday, January 04, 2013 1:20 PM
          > Subject: Re: [RoboMower] Switching to lithiums for RL500
          >
          >
          >
          > I would strongly caution you against building a large pack like that out
          > of cobalt-cathode 18650s. That's just dangerous. Among 90 cells, the
          > failure of a single cell is high, and a single-cell failure can
          > definitely ignite the entire pack. Lithium packs burn hot and regular
          > fire extinguishers may not be appropriate for burning lithium.
          >
          > Yes you can certainly get superior capacity over lead-acid, esp if you
          > DON'T weight it back down. The lighter mower requires substantially
          > less power to the drive motors. But you do need to consider the overall
          > layout and space-efficiency because the volumetric density is only
          > moderately better but the ability to utilize the volume is always much less.
          >
          > Volumetrically, when you're trying to parallel small cells into
          > macrocells, there's a problem where you STILL want a macrocell to be
          > rectangular for practical reasons. There's no law against making
          > nonrectangular macrocells but it's just awkward to assemble Tetris-piece
          > shapes into a pack.
          >
          > Danny
          >
          > On 1/4/2013 2:41 PM, wolfbob wrote:
          > > I don't understand the problem, Danny I
          > >
          > > took a picture showing all the 18650s I have (90) only filling about 60% of the RL500 battery compartment. Even with the charger, BMS and some lead weights in there it should be no problem to get 24 volts 20Ah. The SLAs as you point out only furnish around 10Ah out of the 17 Ah rating and have any life.
          > >
          > > Gotta run now, but I can post the picture when I figure out how..
          > >
          > > WBob
          > >
          > > ----- Original Message -----
          > > From: Danny Miller
          > > To: RoboMower@yahoogroups.com
          > > Sent: Friday, January 04, 2013 11:36 AM
          > > Subject: Re: [RoboMower] Switching to lithiums for RL500
          > >
          > >
          > >
          > > Unfortunately the question is now of VOLUMETRIC energy density. And you
          > > have a problem.
          > >
          > > The 12v 20ah half-U1 lead acid batts you're replacing are 1.72166
          > > watt-hr/cu in.
          > >
          > > Now a large-format rectangular lifepo4 cell can be 2.1 watt-hr/cu in.
          > > The 26650 2.3AH A123's come to 1.44 wh/cu in.
          > >
          > > The problem is the old case has almost 100% utilization of the volume
          > > inside the pack. Cylindrical cells have considerable airspace AND will
          > > have a remainder of space on the sides because the sizes won't line up
          > > just right with the volume.
          > >
          > > Rectangular cells will likely not be able to use all the space either.
          > > And you have to have exactly 8 cells, not 7, not 9. You can't just add
          > > as many cells as the pack can hold, if you can only fit 7 with the "big"
          > > size a mfg has, so you select a much smaller one and you see you can
          > > stack 8 but there's still room for 2 on the side and one flat across the
          > > top, you will still have to just use 8 and ignore the extra volume.
          > >
          > > It's just very likely you're going to have a significant volume problem
          > > with any arbitrary LiFePO4 shape (but rectangulars will usually make
          > > much better use of the space).
          > > So rectangulars are 22% denser by volume, but having a case that's only
          > > 2/3rd battery and 1/3rd air would be a pretty lucky fit the way I see
          > > it. It will be less dense. Of course lead-acid always had trouble
          > > actually making the rated capacity repeatedly so the energy density for
          > > lead-acid was a bit optimistic.
          > >
          > > A123 LiFePO4 is 3.3v per cell not 3.6v, you need 8 for 24v. That's
          > > 26.4v, or 13.2v per "batt" on the old lead-acid pair, a LITTLE high but
          > > that's the voltage you see when it's on the charger. Most other mfgs
          > > rate their cells at 3.2v though.
          > >
          > > For rectangulars go to Thundersky/Winston batt, or better yet CALB batt.
          > >
          > > I don't know of an A123 "3AH" cell. The common 26650 A123 is 2.3AH.
          > > They do make a AHR32113 cell which is 4.4AH in their gen-2 version of
          > > that cell, but those would be hard to obtain anyways, they're not sold
          > > on ebay or online or anything.
          > >
          > > Danny
          > >
          > > On 1/4/2013 9:30 AM, Christopher Zach wrote:
          > > > Last year I tried running a set of 20ah A123 pouch cells in my RL500
          > > > with a BMS on them. Not only did they work, the biggest problem was that
          > > > I couldn't run them down. Mower just kept going and going....
          > > >
          > > > It convinced me that Lithium is a way to go, keeping one old dead lead
          > > > battery in there as ballast. However the A123 cells are too tall and
          > > > have *way* too much power, so I'm going to try something else.
          > > >
          > > > Anyone thought about building a pack of smaller 3ah A123 cylinder cells
          > > > in parallel? A 17ah lead pack will put out only around 12ah under load
          > > > tops, so 4 strings of 3ah cells would do it nicely. 3.6 volt cells would
          > > > allow a 7 cell string to provide 24+ volts, 8 cells would make the
          > > > motors *sing* at 28 volts but might be too big to charge with the
          > > > on-board charger.
          > > >
          > > > I'm going to start work this winter, to have something for March/April.
          > > >
          > > > C
          > > >
          > > >
          > > > ------------------------------------
          > > >
          > > > Yahoo! Groups Links
          > > >
          > > >
          > > >
          > > >
          > >
          > >
          > >
          > >
          > >
          > > [Non-text portions of this message have been removed]
          > >
          > >
          > >
          > > ------------------------------------
          > >
          > > Yahoo! Groups Links
          > >
          > >
          > >
          > >
          >
          >
          >
          >
          >
          > [Non-text portions of this message have been removed]
          >
          >
          >
          > ------------------------------------
          >
          > Yahoo! Groups Links
          >
          >
          >
          >
        • Lee Hart
          ... I agree with Danny. Your typical 18650 laptop cell is *highly* combustible. Extensive safety measures need to be taken to prevent them from ever catching
          Message 4 of 8 , Jan 5, 2013
          • 0 Attachment
            On 1/4/2013 3:20 PM, Danny Miller wrote:
            > I would strongly caution you against building a large pack like that out
            > of cobalt-cathode 18650s. That's just dangerous.

            I agree with Danny. Your typical 18650 laptop cell is *highly*
            combustible. Extensive safety measures need to be taken to prevent them
            from ever catching fire, because if they do, it will be spectacular.
            Think of them like a box of matches.

            The LiFe cells still burn, but it's harder to get them started, and the
            resulting fire spreads much slower. Think of them more like a box of
            kindling.

            > you need to consider the overall layout and space-efficiency because
            > the volumetric density is only moderately better, but the ability to
            > utilize the volume is always much less.

            Again, I agree. The watthours per pound of lithium is about 3 times
            better than lead acid, but the watthours per cubic foot is only slightly
            more. When I do talks, I have three batteries that graphically
            illustrate this: One Hawker lead-acid 12v 60ah, one GM/Ovonic nimh 12v
            90ah, and one Thundersky 12v 90ah. All three are the same size. But the
            lead-acid weighs 60 lbs, the nimh 40 lbs, and the lithium 22.5 lbs.

            When you have a fixed size battery box, you'll wind up with about the
            same watthour capacity no matter which battery chemistry you pick.

            --
            Results! Why man, I have gotten a lot of results. I know several
            thousand things that won't work. -- Thomas A. Edison
            --
            Lee A. Hart, http://www.sunrise-ev.com/LeesEVs.htm
          Your message has been successfully submitted and would be delivered to recipients shortly.