What came of the Milbay battery pitchman?

softdown
softdown Solar Expert Posts: 3,812 ✭✭✭✭
At first I was transfixed by information about a newish lithium based battery technology. Then I was "impressed" by the constant pitching with a lack of concrete facts. Then I became concerned about all of the electronics and possible issues with the technology. Then...poof...no more Milbay sales pitching.

If I am stepping on toes....feel free to delete this post. Just curious....
First Bank:16 180 watt Grape Solar with  FM80 controller and 3648 Inverter....Fullriver 8D AGM solar batteries. Second Bank/MacGyver Special: 10 165(?) watt BP Solar with Renogy MPPT 40A controller/ and Xantrex C-35 PWM controller/ and Morningstar PWM controller...Cotek 24V PSW inverter....forklift and diesel locomotive batteries

Comments

  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    All I did was close the one thread... I have done nothing else and not heard from him since.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • Mountain Don
    Mountain Don Solar Expert Posts: 494 ✭✭✭
    Goes to show his heart was set on self promotion and not the actual exchange of useful information in general. 
    Northern NM, 624 watts PV, The Kid CC, GC-2 batteries @ 24 VDC, Outback VFX3524M
  • zoneblue
    zoneblue Solar Expert Posts: 1,220 ✭✭✭✭
    sshh, lets not encourage the guy.
    1.8kWp CSUN, 10kWh AGM, Midnite Classic 150, Outback VFX3024E,
    http://zoneblue.org/cms/page.php?view=off-grid-solar


  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    Since I cannot read people's minds or hearts through the laptop--I just try to give everyone a fair chance/hearing.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • westbranch
    westbranch Solar Expert Posts: 5,183 ✭✭✭✭
    All I saw was one individual that got stuck in the (uninformative) groove and wouldn't step out :'(
     
    KID #51B  4s 140W to 24V 900Ah C&D AGM
    CL#29032 FW 2126/ 2073/ 2133 175A E-Panel WBjr, 3 x 4s 140W to 24V 900Ah C&D AGM 
    Cotek ST1500W 24V Inverter,OmniCharge 3024,
    2 x Cisco WRT54GL i/c DD-WRT Rtr & Bridge,
    Eu3/2/1000i Gens, 1680W & E-Panel/WBjr to come, CL #647 asleep
    West Chilcotin, BC, Canada
  • softdown
    softdown Solar Expert Posts: 3,812 ✭✭✭✭
    Did anyone else find the sheer amount of electronics to be worrisome? I did not go off the grid to surround myself with such technology. Sometimes I consider going Amish.
    First Bank:16 180 watt Grape Solar with  FM80 controller and 3648 Inverter....Fullriver 8D AGM solar batteries. Second Bank/MacGyver Special: 10 165(?) watt BP Solar with Renogy MPPT 40A controller/ and Xantrex C-35 PWM controller/ and Morningstar PWM controller...Cotek 24V PSW inverter....forklift and diesel locomotive batteries
  • BB.
    BB. Super Moderators, Administrators Posts: 33,431 admin
    It was always a problem for me... When designing equipment with hot standby, redundant power supplies, etc... I always worried that while throwing more electronics at stuff, I was doubling (or more) the amount of hardware and creating more new and wonderful ways for things to fail. And making things more difficult to debug.

    Even went so far that one co-engineer of ours put in a defeat jumper (the system had a hard shutdown if any basic thing failed) so the system could be kept running until the problem was found (for example, one million dollar system had a badly assembled cable). But the folks who wrote the service manual refused to put anything about the override in the the service manual. Had at least one system mechanically replaced because of a connector pin not fully pushed home (and could not keep the system running until the service people could try replacing the harness).

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • westbranch
    westbranch Solar Expert Posts: 5,183 ✭✭✭✭
    softdown said:
    Did anyone else find the sheer amount of electronics to be worrisome?
    That is a main reason that I have been following the users here that are advocating '' Bottom Balancing" rather than the industry standard '' Top Balancing'', complex and IMO more prone to failure.  Bottom balancing however does not give you the max output but rather safety...!!..  ie unlikely to have an overcharge shutdown, that the Top balancing can do...
     
    KID #51B  4s 140W to 24V 900Ah C&D AGM
    CL#29032 FW 2126/ 2073/ 2133 175A E-Panel WBjr, 3 x 4s 140W to 24V 900Ah C&D AGM 
    Cotek ST1500W 24V Inverter,OmniCharge 3024,
    2 x Cisco WRT54GL i/c DD-WRT Rtr & Bridge,
    Eu3/2/1000i Gens, 1680W & E-Panel/WBjr to come, CL #647 asleep
    West Chilcotin, BC, Canada
  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    Top vs bottom balancing quickly turns into a religious debate - which is of most concern to EV/RC or other high-amperage motive power situations.

    In our application, top-balancing can be easily achieved, but it too is often misunderstood and puny current-shunt bleeder boards are another source of failure.  And all too often, the TIME factor necessary to perform this type of balancing isn't taken into account - holding the cells at full charge longer than necessary over the lifetime of the pack.

    We have the luxury of conservatism and KISS in our low-amperage application.

    Consider a simple 12 / 24v system where one does an initial charge on each cell to full.  (typically 3.6v / 0.05C end current).  Bench supply makes this easy, but there are "single cell" chargers out there too.  (lurkeres, these are 3.2v LFP cells/chargers, NOT the higher 3.7v!!!)

    Now, never run it into the upper knee.  Set your charger for 3.4v per cell (13.6v for a nominal 12v battery of 4 cells).  It will never fully charge.  And no need to!  If you leave it on for long periods, since 13.6v (or 3.4v under charge) is never fully charged, you don't have to freak out if you forget, or if your solar array goes weak - you'll never be holding these cells at full charge by accident.  While there is no need to "float" LFP at 3.4/cell (13.6v for 12v nominal), many have reported no damage long-term.  13.8v is a different story, as that WILL actually fully charge given enough time!

    Utilize your own LVD, set for say 10.7v max under load.  Normally, stop at about say 12.5v under load for near 80% DOD.  Run even narrower window of psoc if you like.

    At any rate, if anyone is serious, just grab four, 40ah GBS, Winston, Calb's.   If that is too much, then my recent educational build made from Kingsolar 3300mah 26650 cells (again, energy cells and not power cells) I found on Amazon can cut down on the cost for an educational pack.  Keystone makes some nice holders for these.

  • Blackcherry04
    Blackcherry04 Solar Expert Posts: 2,490 ✭✭✭
    edited January 2016 #11


    Now, never run it into the upper knee.  Set your charger for 3.4v per cell (13.6v for a nominal 12v battery of 4 cells).  It will never fully charge.  And no need to!  If you leave it on for long periods, since 13.6v (or 3.4v under charge) is never fully charged, you don't have to freak out if you forget, or if your solar array goes weak - you'll never be holding these cells at full charge by accident.  While there is no need to "float" LFP at 3.4/cell (13.6v for 12v nominal), many have reported no damage long-term.  13.8v is a different story, as that WILL actually fully charge given enough time!

    Utilize your own LVD, set for say 10.7v max under load.  Normally, stop at about say 12.5v under load for near 80% DOD.  Run even narrower window of psoc if you like.

    I pretty much follow PNJ's charging / discharging thoughts, they give you the best chance for long term life from your cells. Do take note that when you look at the dod that 100% to 20% is 80% of capacity, but that not what we're using. Being conservative in charging / discharging gives you 60-70% of capacity in reality, so size your bank accordingly.

    Absorb Phase : Once you reach the 13.6v @ 12v when charging, you then have to absorb to allow the ion's time to reshuffle and repopulate. I see recommendations from 3% of Capacity to 50% for EA. I have been on the high end ( 50% ) because I am not using the capacity I have, so no reason to tempt mother nature, If you don't absorb you'll notice a loss in capacity, as the pack voltage will drop faster. Absorbs do take time and you have to figure that into your charging scheme with sun hours and PV.

    Usable Capacity is real $$$ when your dealing with LiFePO4.
  • zoneblue
    zoneblue Solar Expert Posts: 1,220 ✭✭✭✭
    Well Bc4, if youve embraced LFP, then we know its come of age, and its safe for the rest of us ;)
    1.8kWp CSUN, 10kWh AGM, Midnite Classic 150, Outback VFX3024E,
    http://zoneblue.org/cms/page.php?view=off-grid-solar


  • Blackcherry04
    Blackcherry04 Solar Expert Posts: 2,490 ✭✭✭
    edited January 2016 #13
    zoneblue said:
    Well Bc4, if youve embraced LFP, then we know its come of age, and its safe for the rest of us ;)
    lol, I am still in the test phase, I will reserve that decision till later. It's cost and fiddle factor for me. I am at 16 months and no issues yet and I am not looking for any. Right now I'd say if they drop to 2X cost, you could justify them , at 3X cost, they are expensive. I tried to point out above that the 80% dod is deceiving and the charge time is almost the same as FLA.

    PSOC charging could be beneficial to some and if your recouping fuel costs that could make them more attractive.  You can do that with FLA, but it gives you less capacity to play with. If you use a Inverter / Generator and charge at a lower rate you can probably beat that cost difference. You can now charge and power loads over a 10 hr period with a Honda for less than you use to be able to charge for a hour .

  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    Don't forget that the cost/comparisons for things like cycle life against Pb are always under lab-perfect conditions.  Most of us murder our Pb batteries and will never reach those lab / marketing specs.  So the cost differential could actually be even cheaper in the long run under real world conditions as long as you have your LFP under control.  Conservatism helps us achieve this simply.

    Re absorb:  under charge at 3.45 volts per cell the cells will eventually fully charge as you watch the absorb taper down to near nothing.  Still, this is too high if left unattended for days, weeks, months.

    By dropping the voltage just a tad to 3.4v *under charge*, even though you will still witness absorption (not recombination of course like with Pb), when the current tapers off to nothing, you still won't be fully charged.  Seems ideal to me for a varying source like the sun.

    So with a little conservatism, and giving up some minor capacity at the top end, we can really keep it simple.  Here charging each cell individually to full first, and then running conservative at 3.4v under charge overall means we are relying on the QC of the cell manufacturer to be reasonably close in both capacity and internal resistance.  If you are waay out, then we have a QC issue, much like we would if we dropped a cell in a Pb bank.

  • karrak
    karrak Solar Expert Posts: 326 ✭✭✭✭
    I pretty much follow PNJ's charging / discharging thoughts, they give you the best chance for long term life from your cells. Do take note that when you look at the dod that 100% to 20% is 80% of capacity, but that not what we're using. Being conservative in charging / discharging gives you 60-70% of capacity in reality, so size your bank accordingly.


    I think you can get much better utilisation that 60-70%. In my case I have battery capacity around three times my winter daily requirements which allows me to run my system without a generator. If there is enough sun I charge my battery to 3.4 volts/cell allowing the current to drop to C/50. I used to charge to an end current of C/20 but found that the variation in SOC could be greater that 10% depending on the variability of the power from the solar panels. I have found that the difference between charging to 3.4 volts/cell and 3.5 volts/cell at this end current is only a few %. This implies to me that charging to 3.4 volts/cell at an end current of C/50 will give you an SOC in excess of 95%. Most of the time my battery is more than 50% full, the average SOC being around 75%. Fairly regularly the battery will dip into the high 20s%, and occasionally the battery will get down to ~15%. My total utilisation is around 80%.

    If I were running a generator I would go for one day of battery storage and cycle the battery 50-60% on a daily basis and keep the battery between 95% and 20% SOC.
    Simon
    Off-Grid with LFP (LiFePO4) battery, battery Installed April 2013
    32x90Ah Winston cells 2p16s (48V), MPP Solar PIP5048MS 5kW Inverter/80A MPPT controller/60A charger, 1900W of Solar Panels
    modified BMS based on TI bq769x0 cell monitors.
    Homemade overall system monitoring and power management  https://github.com/simat/BatteryMonitor
     

  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    I have actually changed my conservative voltage just a *wee* bit.

    I now ride the edge at 3.425v per cell (13.7v for you 12v nominal guys).  Just under 3.45.  Using a TRUSTED voltmeter to make my settings measured at the terminals, not the CC display!

    The reason for this is that at 3.4v under charge, the absorption comes on a bit too early for my tastes.  I'm not striving to get to 95% SOC quickly, but the absorption curve slows things down approaching that point - and with solar, I often don't have the luxury of unlimited time like an ac charger.  It's just too slow.  Very safe, but I wanted a bit more efficiency even though it isn't strictly necessary.

    Raising the voltage from 3.4 to 3.425 per cell delays the absorption a bit, and makes bulk a bit more efficient when clouds, haze, or other shadows come into play.

    So here it is a matter of fine-tuning for my setup.  That's the thing about LFP - plenty of room to tweak if one wants according to their needs.

    Still, charging underway at 3.4v per cell would be probably the "baseline" for the most conservative setup.  For me, 3.425v made the porridge "just right". :)


  • westbranch
    westbranch Solar Expert Posts: 5,183 ✭✭✭✭
    PNj, what do you feel the SoC you can achieve (will achieve? )  after this TWEAK is?
     
    KID #51B  4s 140W to 24V 900Ah C&D AGM
    CL#29032 FW 2126/ 2073/ 2133 175A E-Panel WBjr, 3 x 4s 140W to 24V 900Ah C&D AGM 
    Cotek ST1500W 24V Inverter,OmniCharge 3024,
    2 x Cisco WRT54GL i/c DD-WRT Rtr & Bridge,
    Eu3/2/1000i Gens, 1680W & E-Panel/WBjr to come, CL #647 asleep
    West Chilcotin, BC, Canada
  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    Glad you asked!  The actual SOC when all is done is about the same, so I'm not striving for an increase there.

    What I'm doing is shortening the time taken to get to 95%, by raising the voltage just slightly.  Ie, when in absorb at 3.4v steady under charge, I might be pushing 2A initially.  By changing to 3.425v, that increases to 4a or a little more initially.  Essentially I'm reducing the time factor by about half in absorb.  I duplicated this test with a bench supply and that kind of opened my eyes.

    What that really meant was that I wanted to keep the speed efficient when going from 80%-95% especially when I have intermittent solar.

    True, it is a hard thing to stop thinking in Pb terms of 100 - xx% SOC, and for many, the slower climb from 80-95% might be negligable, especially if their bank is large enough and properly designed for some autonomy in the first place.

    For me, I just wanted to make sure I could get to 95% if I truly felt I needed to on a daily basis based on my battery capacity, load, insolation and so forth.  Kind of a worst-case thing to make sure I don't exceed 80% DOD.  Which of course is possible but a poor design parameter with no autonomy.

    The message is clear - I need more LFP! :)


  • westbranch
    westbranch Solar Expert Posts: 5,183 ✭✭✭✭
    Like Karrak said, I have battery capacity around three times my winter daily requirements.    
    So the 3 day autonomy rule of thumb, still applies like I tried to get across to the Milbay guy
     
    KID #51B  4s 140W to 24V 900Ah C&D AGM
    CL#29032 FW 2126/ 2073/ 2133 175A E-Panel WBjr, 3 x 4s 140W to 24V 900Ah C&D AGM 
    Cotek ST1500W 24V Inverter,OmniCharge 3024,
    2 x Cisco WRT54GL i/c DD-WRT Rtr & Bridge,
    Eu3/2/1000i Gens, 1680W & E-Panel/WBjr to come, CL #647 asleep
    West Chilcotin, BC, Canada
  • PNjunction
    PNjunction Solar Expert Posts: 762 ✭✭✭
    EXACTLY!  With my setup with nearly no autonomy, that forces me to ride the knife edge.

    A lot of guys get suckered into designing an LFP system for 80% DOD, and even though you CAN with good cycle life,  you now have no autonomy, and it can take away the luxury of conservatism and simplicity.

  • Blackcherry04
    Blackcherry04 Solar Expert Posts: 2,490 ✭✭✭
    edited January 2016 #21
    Simon , in your case if your charging to 95% and getting 75-80% in return and are happy , more power ( no pun ) to you. In my case I figure I am cutting off about 10% on both ends so I end up at 60%. I am willing to be a little more conservative. Everyone will have to judge their level of comfort and pocketbook. I really don't think my A/C or pump could kick on at 20% without dragging the voltage below the LVD, but this is a learning system and if I eventually get to a larger system, 20% soc should be doable.

    Ugg, I just looked and my battery shed temperature has dropped to 34° F. I am going to have to start thinking about adding some heat or moving them today. Just another adjustment for a 30 year FLA user.

    One thing you can see, is that PNJ is talking about 100's of a volt adjustments. It's really hard to even calibrate all your readings in your mind for all the equiment. Being off by .020 is a like success or failure. Can you tell the difference in 3.4v and 3.425v per cell on all your equipment across the board ?? It's really easy to off by 10% in what you think you have and what you really have.
  • karrak
    karrak Solar Expert Posts: 326 ✭✭✭✭
    Simon , in your case if your charging to 95% and getting 75-80% in return and are happy , more power ( no pun ) to you. In my case I figure I am cutting off about 10% on both ends so I end up at 60%. I am willing to be a little more conservative. Everyone will have to judge their level of comfort and pocketbook.
    As far as I can see the major issue with LFP batteries is cost so getting the best possible utilisation from them is very important. From the literature I have read and my and other peoples experience I can't see that charging an LFP battery up to 95%+ versus say 80%  at a conservative charging voltage of 3.4-3.45 volts will make much difference to their lifespan. Now charging them at 3.6 volts at C/2 to 95% is another matter. I don't think discharging them occasionally to 3.0 volts will cause any problems either. Time will tell.
    I really don't think my A/C or pump could kick on at 20% without dragging the voltage below the LVD,
    My experience is that I can draw around C/3 out of the battery that is at around 20% SOC and only drop the voltage to just above 3 volts/cell. I have my LVD set at 2.8 volts/cell
    One thing you can see, is that PNJ is talking about 100's of a volt adjustments. It's really hard to even calibrate all your readings in your mind for all the equiment. Being off by .020 is a like success or failure. Can you tell the difference in 3.4v and 3.425v per cell on all your equipment across the board ?? It's really easy to off by 10% in what you think you have and what you really have.
    I only start getting concerned when the  voltage difference between the cells gets to more that 100mV (0.1 volts). If you have a multimeter with better than 1% accuracy on the voltage range that should be sufficient. My multimeter is accurate to around +/-0.1% and my BMS around +/-0.2%. I have found that measuring the SOC of the battery by measuring the current going into and out of the battery to be accurate to within a few percent
    Simon
    Off-Grid with LFP (LiFePO4) battery, battery Installed April 2013
    32x90Ah Winston cells 2p16s (48V), MPP Solar PIP5048MS 5kW Inverter/80A MPPT controller/60A charger, 1900W of Solar Panels
    modified BMS based on TI bq769x0 cell monitors.
    Homemade overall system monitoring and power management  https://github.com/simat/BatteryMonitor
     

  • Blackcherry04
    Blackcherry04 Solar Expert Posts: 2,490 ✭✭✭
    edited January 2016 #23
    Being off by .020 is a like success or failure. Can you tell the difference in 3.4v and 3.425v per cell on all your equipment across the board ?? It's really easy to off by 10% in what you think you have and what you really have.
    Let me clarify, I wasn't talking about cell balance and Multimeters, I was talking about Inverter, SCP, Mate, Charge Controllers, Stand Alone chargers, Generator AGS systems, Battery Monitors. To ever get them all calibrated to the same exact voltage at .020 - .20 is near impossible and it does effect cell voltage. It's real easy to end up way off from where you want to be using the equipment we currently have if your not sitting there baby sitting them. I have noticed changes as the ambient temperature has changed. While the cell charging voltage may not need compensation the charging is a little different as it's got colder. Example : I was at 13.24v and now it is resting at 13.17v @ 40° F. as it warms up, it will go back up. Of course the more cells you have and the higher the voltage the more it's spread out across the board.