Edison NiFe batteries

littleharbor2littleharbor2 Solar Expert Posts: 704 ✭✭✭✭
This question is for anybody here who recognizes and knows these. Are these typical Nickel-Iron batteries? The cases and vents? look interesting.

2.1 Kw Suntech 175 mono, Classic 200, Trace SW 4024 ( 15 years old  but brand new out of sealed factory box Jan. 2015), Bogart Tri-metric, 700 ah @24 volt AGM battery bank. Plenty of Baja Sea of Cortez sunshine.

Comments

  • PhotowhitPhotowhit Solar Expert Posts: 4,281 ✭✭✭✭
    Hard to tell just by looking at the outside, but if there are just 10 cells used together, I'd bet on NiCD that have a 1.2 nominal voltage.
    Home system 4000 watt (Evergreen) array standing, with 2 Midnite Classic Lites,  Midnite E-panel, Prosine 1800 and Exeltech 1100, ForkLift battery. Off grid for @13 of last 14 years. 1000 watts being added to current CC, @2700 watts to be added with an additional CC.
  • littleharbor2littleharbor2 Solar Expert Posts: 704 ✭✭✭✭
    They're listed on eBay as "Edison - Exide NiFe Battery 337.5 Amp Hour " batteries

    2.1 Kw Suntech 175 mono, Classic 200, Trace SW 4024 ( 15 years old  but brand new out of sealed factory box Jan. 2015), Bogart Tri-metric, 700 ah @24 volt AGM battery bank. Plenty of Baja Sea of Cortez sunshine.

  • softdownsoftdown Solar Expert Posts: 1,560 ✭✭✭✭
    I think NiFe battery cells are rated at 1.2 volts instead of the 1.5 we all know and love. Other than that, their appearance seems normal enough. Too bad the cables were cut....they were pretty burly.
    First Bank:16 180 watt Grape Solar with  FM80 controller and 3648 Inverter....Fullriver 8D AGM solar batteries. Second Bank/MacGyver Special: 10 150 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
  • mike95490mike95490 Solar Expert Posts: 7,308 ✭✭✭✭
    NiCad or NiFe, they both are nearly indestructible, are 1.22V  and take the same electrolyte - KoH potassium hydroxide.  NiCad has the added cachet of being Toxic.  I'd ask for a photo of the nameplate.
     If anyone ever added acid, or if the cells were drained and sat empty, the iron turns to iron oxide and then they are useless.  But the outer shell looks intact, so I doubt that's the case.
    Powerfab top of pole PV mount | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

    solar: http://tinyurl.com/LMR-Solar
    gen: http://tinyurl.com/LMR-Lister ,

  • SylphhawkinsSylphhawkins Registered Users Posts: 8 ✭✭
    Those are Nickel Iron cells from the Edison Battery Company in Orange, NJ. You should be able to see the metal engraving on the lid caps. He used metal (steel) cases rather than plastic (right up to 1972 when Exide corp bought out the company and ceased NiFe battery production). The metal cases were isolated from each other when the cells were grouped in series by using an ebonite-tar coating (the black outside coating)
    They charge at 1.7v per cell, float at around 1.4v p/c and have a NOMINAL cell voltage of 1.2v
    They should ideally be charged at the 7-Hour rate, Edison makes a big note of this is some of his patents on the cells.
    Nickel-Iron was originally invented by a brilliant electro-chemist called Ernst Waldemar Jungner, from Sweden, Edison did what he did best and "built on" the invention and patented it in the USA in late 1890's to early 1900's.

    He did make some notable improvements on the original invention though, and the use of Nickel Oxyhydroxide at the key active compound on the Positive electrode increased cell capacity and cell voltage over the use of straight Nickel Hydroxide.
    In the batteries you've pictured above, the + electrodes are made of nickel with alternating "flakes" of nickel and nickel oxyhydroxide pounded into round cylindrical long tubes and then pressed onto flat plates. 
    The - electrodes are pure Iron (Fe) flakes also pounded into cylindrical tubes and pressed onto flat plates.

    The electrolyte can be any of the alkali hydroxides, though you will achieve the highest cell voltages using Potassium Hydroxide (KOH), but NaOH can also be used if needed. Modern versions add 10% quantities of LiOH (Lithium Hydroxide) for added smoother voltage plateau's under heavy load.

    The batteries would last forever chemically (not counting physical abuse) if not for the fact that the highly concentrated KOH absorbs CO2 from the air and forms Potassium Carbonate (KCO3) in the electrolyte. This interacts with the Iron (the - electrode) under normal charge conditions and forms Iron Carbonates (FeCO3) which impede the capacity of the cell and eventually lead to a slow but consistent decrease in amp-hour capacity. One of Edison's later revised patents on the NiFe cell specifically mentions using the steel metal case to act as an electrode when performing a "Refreshing charge cycle", where the nickel and iron electrodes are shorted/joined together and are charged as a single + electrode, with the steel case becoming the new - negative electrode. This process helps to push out the Carbonates from the iron due to the positive charge effect on the plate, and transfers the carbonates to the outside case. You then give them quite a sloshing and pour out the electrolyte and replace with fresh KOH. 

    His notes comment on this process being able to effectively restore almost the entire amp-hour capacity (a feature that is missing in the new versions of the NiFe batteries that are coming out of China in recent years in plastic cases).

    The KOH electrolyte doesn't take part as a reactant during normal operation.

    Upon normal discharging (from a fully charged state) the process that follows is:
    Fe(s) + NiOOH(s) + 2H2O <=> Fe(OH)2 + Ni(OH)2 + H+
    Then a further decomposition reaction of the Iron electrode (at a half-equation voltage of approximately 0.9v) of;
    3Fe(OH)2 + 2NiOOH <=> 2Ni(OH)2 + Fe3O4 + 2H2O 

    The - iron electrode usually ends up becoming a form of Iron oxide, and this is fully reversible during the next charge as the iron undergoes reduction back to straight metallic Fe(s).

    The cells can sit empty and dry for years and be revitalised back to near full capacity.

    The comment in the reply above about the problems that could occur if the cells have been filled with an acid electrolyte instead of the alkaline electrolytes is due to the solubility of many of Nickel's acid-based compounds such as Nickel Sulphate (highly soluble), which forms if the cell has ever had sulphuric acid used in it.
    The chances of this are very slim, but you could tell immediately if this has happened by giving a water rinse of the cell and seeing if any blue-green colour liquid comes out (Nickel Sulphate dissolved in water).

    I have a few of these cells myself, dated from 1916 from the Edison Battery Storage Co.
    They are fantastic cells, and really put our overall battery inventions since, to shame on many levels.
    They have a low energy density (approx. 30-50Wh per Kg) which is close to Flooded Lead Acid (~40Wh per Kg), but their extremely long life span (25 years + before needing the "refresh charge" to be performed due to the iron carbonate build up) and their potential of being restored again many times over makes them extremely vital.

    If you have the opportunity to play with these old cells, I highly recommend it.
    You can learn everything you need to learn about managing them and their chemistry from reading the patents of both Jungner and Edison, simply do google patent searching for both names and look for "Alkaline rechargeable / secondary battery patents".

    - Sylph Hawkins, Australia


  • BB.BB. Super Moderators Posts: 26,872 admin
    Thank you for the information Sylph Hawkins.

    That is very interesting bit of history and chemistry.

    -Bill
    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • Dave AngeliniDave Angelini Solar Expert Posts: 3,333 ✭✭✭✭
    I agree it was a nice post! Thanks!
    That is why I have always thought (and was told) that these were doomsday batteries!

    I am not sure Nikola Tesla would go down this path as he was interested in the efficiency end of the Engineering. He might look at the graph below and shake his head.



    "we go where power lines don't" Sierra Mountains near Mariposa/Yosemite CA
     http://members.sti.net/offgridsolar/
    E-mail offgridsolar@sti.net

  • SylphhawkinsSylphhawkins Registered Users Posts: 8 ✭✭
    Re: Nickel Iron battery 65% coulomb charge efficiency. Thinking about the chemistry, this is going to be mostly a function of the large Hydrogen gassing I reckon due to water electrolysis, which is, incidentally, a function of the low H+ evolution voltage from the iron electrode (I have read that this sits at around 1.4v the pure Ferrous electrode). Textbook average water electrolysis voltages that I have found are given between 1.7v and 2.5v depending on the metals used for the electrodes. 

    Lead acid also has this problem, which is why Calcium/Silver began being used in the alloy for the positive electrode (it raises the Oxygen evolution voltage, which means that at standard charge voltages, less off-gassing should occur, and more of the charge will be specifically oxidising the lead positive plate rather than to water electrolysis). Using simple plate lead/lead in sulphuric acid has a lower charge efficiency than a calcium alloy positive lead/calcium cell of the same amp-hour rating.

    Perhaps someone will discover an alloy that can be made with the iron which will raise the Hydrogen evolution voltage above 1.7v, and could probably then present a NiFe battery with an overall coulombic charge efficiency of >80%.
    Graphite comes to mind, and any other metals that form insoluble hydroxides in alkaline environments under -1.0v charges.

    Still, having said that, when taking into consideration of coulomb charge efficiencies, we must also consider the long term life span and cycle-ability of the cell, a charge efficiency of 100% is completely pointless if the cell lasts for 100 charge cycles, is made of rare earth metals and cost 4x times the price of an equivalent capacity battery bank, and conversely, a cell that could last for 30 years + (or over a century if you continue too force out the iron carbonates), is made from 2 of the most abundant metals we have access to, is non toxic, easily recycled and can withstand almost constant complete abuse should be factored into the story.

    We need a new unit of measurement that considers overall charge efficiency with expect life span, cost of input materials, environmental concerns of recyclability and safety handling and embodied energy in cell production, then the figures would be more useful from an investment concern.

    Given the ease of rejuvenation of alkaline cells such as these and the seemingly near-indestructibility of the cells, I would simply invest the extra money that I save (by not purchasing battery banks every 10 years) into a 35% increase in my solar/wind/hydro/generator charging system.

    :wink:

    I have a bank of 1960's NiCd wet fill cells that I experiment with as well as a small bank of 1916 Edison Battery Company cells (identical to the cells above in the first post) which I play with and explore. I have not personally experimented with the brand new NiFe cells coming out of China with the clear plastic cases, but It sure would be interesting to put them to a side-by-side test with the 1916 Edison cells.

    - Sylph H
  • Dave AngeliniDave Angelini Solar Expert Posts: 3,333 ✭✭✭✭
    Another great post Sylph!

    You left out the maintenance part of NiFe and for Offgrid the part about winter when electricity is at premium and charging is a chore and the word "play".is not what one would use to describe the task.

    Simply over designing the charge source is not always possible although this is my primary design tool :)

    The battery below is being tested here (a few others also) and I have cycled it about 4 times in the year to about 30% Soc. I expect it to last twice the 10 year warranty at this rate cycling most days down to 70% Soc.

    I go back to my doomsday description of NiFe and do appreciate the part about it being able to last the years in a storage mode and of course the lack of  BMS electronics and their failure modes.

    Where in Australia? I have a few clients down under.





    "we go where power lines don't" Sierra Mountains near Mariposa/Yosemite CA
     http://members.sti.net/offgridsolar/
    E-mail offgridsolar@sti.net

  • SylphhawkinsSylphhawkins Registered Users Posts: 8 ✭✭
    Hi Dave, 
    Thanks for your reply. I'll write my reply over 2 posts, :)

    Yes, totally agree that the build does not always allow for oversizing the charging source (when retrofitting), although at this stage we are talking about battery banks that cost in the $10,000-$30,000 range for, say, a FLA (Flooded Lead Acid) 30kWh bank (depending on tubular or plate cells) and I expect that for people ready to invest this amount of money into their system, there is (I have noticed) budget to either size up the system, or provide alternative energy source inputs.

    Re: the maintenance of NiFe, this is similar to the ongoing maintenance of FLA or wet-filled NiCd, that being a usual water top up every 6 to 12 months depending on DOD cycling and equalisation charging, however, unlike FLA, the KOH/LiOH electrolyte of the NiFe does require dumping out and changing once the cell has absorbed a substantial amount of CO2 from the atmosphere. 

    I read about (not seen in person) NiFe setup where the owners use a thin film of mineral oil in the cells (this is usually recommended by the cell manufacturers I have noticed) to reduce the CO2 absorption by the electrolyte, and that this alone has sometimes pushed out the "electrolyte change over period" to be around the 10 year mark, rather than after 3 to 4 years. I'm sure that must be more that could be done around this from some inventive laboratory.

    If done correctly, and especially when using "tall" cells that are designed to house double the amount of electrolyte so that the watering only has to take place every 1 to 2 years instead, plus the use of mineral oil as a barrier/air sealant, it seems that a NiFe battery bank might be able to get away with:


    * Water top-ups once every 1 to 2 years
    * Full electrolyte change every 10 years
    * A "de-carbonating charge" (by shorting out the + and - electrodes and using that metal case as the new negative electrode for 1x full charge cycle) every 20-25 years

    It's starting to sound more and more doable, just like FLA always has been for off-grid systems.

    I know that the hype of maintenance free batteries have taken off in the last 10 years, especially for the semi-rural / off-grid lifestyle choice demographic, and more and more off-grid lead-acid battery banks are using AGM and GEL cells (using GEL for a stationary setup is a bad idea if you ask me, given that the slightest over-charging can cause the gel to be pushed away from the plates rather permanently and that there is no need to have the batteries installed up-side down or sideways. Given the cost at approx. 2x-3x the price of an equivalent capacity FLA setup, I think that it is more sales marketing that drives GEL for off-grid systems that actual sense. Even though the usual argument is the benefit of a better life-cycle at greater DOD (50-70% giving still over 1000 cycles), I have not yet come across a GEL Lead-Acid system for off-grid where the owners are cycling them at more than 50% anyway, and could have done just as well with a thick-plate FLA or a Tubular FLA, but I digress. :p), but it still stands that an equivalent FLA bank, when maintained correctly and setup and sized correctly, will outlive its GEL or AGM counterparts by many years, is more readily re-juvenatable, and can tolerate moderate DOD and overcharging

    I recently read a fascinating paper that was looking at the effect on overall cell life and lead-plate durability in Valve Regulated Sealed Lead-acid (VRLA) and so-called Maintenance free lead-acid, and it showed that the internal cell heating (including the trapped heat of the Lead plates themselves) was shown to be nearly double that of the FLA equivalent, mainly due to the heat energy that would have escaped the cell during gassing (where the H2 and O2 carry heat energy with them out of the cell) but was trapped inside the cell instead when the cell pressure and stopper-caps forced the H2 and O2 to recombine into water under pressure. Some of the effects that were noted were a large increase in cell deterioration, mainly due to the low melting point of Lead and the higher cell temperatures under normal use.


    Re: the Li-ion battery you are currently testing,
    I am the co-owner of an Australian company called the "Melbourne Custom Battery Co.", where I present talks on electro-chemistry around Australia and Melbourne, and my business partner and I consult on / design and assembly battery packs for a range of activities, from off-grid home setups to electric bikes / electric cars / electric motorbikes, photographer day-hiking packs, etc..

    When I saw the >6000 cycles to 90% DOD note in your graphic of the Li-ion batteries you posted, I immediately assumed that it must either be a LiFePO4 (Lithium Ferro Phosphate) cell or one of the new and hard to get Li4Ti5O12 (Lithium Titanate Dioxide) cells, which often have life cycles in the 4000-6000 cycles (LiFePO4) and >10,000 cycles (Li4Ti5O12).

    Once I saw the next few lines where it notes "NMC" chemistry I took a double take.

    Yes, granted that Nickel-Manganese-Cobalt li-ion chemistry has the highest specific energy and energy density of the current Li-ion range, but it does so due to the incredible reactivity of the cobalt (There are currently patented versions of NiFe cells in small scale production which use a cobalt/Nickel alloy for the positive electrode which also can provide specific energy values in the range of over 240wH/Kg).

    The cells use carbon/aluminium as the negative electrodes, and the electrolyte tends to be a Lithium Fluoride paste with organic solvents which form the SEI layer upon the initial formatting charges. This SEI layer is what makes the cell work as an ion-transfer battery (instead of a traditional chemical non-displacement reaction or oxygen-lift cell).
    This SEI layer that coats the negative electrodes is rather fragile, beginning to decompose at 100*c (internal cell temperatures) and causing thermal-chemical runaway problems when the lithium begins to plate onto the negative electrode when the SEI layer fails, and when the cobalt then begins to interact with the Fluoride paste itself.

    Having used these cells for many years and tested them in hundreds of e-bike builds and bench top testing I now advocate for a lower per-cell charge rate of 4.10-4.15v with a BMS activation voltage of 4.08v, to reduce the flammability risk by over 60% (in my in-lab tests) and push out the cycle life by 2x. We have BMS and Chargers for this change produced for us in China as well which we sell to our customers and our assembled li-ion battery packs.

    One of the key things that I discovered with NMC cells (and the NCA cells that Tesla corp. use) is that we have been able to pull 400-700 cycles at most before the cell capacity has dropped to below 80%. The usual data sheet for the cells (18650 cells from Samsung/LG/Panasonic) make specific notes about the cycle life being up to 1000cycles but it is given at a 0.02C rating (50 hour charge/discharge rating) (this is ridiculous for most current applications of Li-ion cells, such as power tools an e-bikes/e-cars as the discharge rates are closer to 1C or 2C and the charge rates are usually at 0.5C).

    My testing has showed that when run at the common 1C or 2C discharge rates that the cycle life of the cells drops back to between 300-400 cycles, not any better than thin-plate VRLA (Lead acid). This has equated to a general "life span" of 3 to 4 years in our testing, including battery packs that we assembled and sent out years ago that are still coming back now with huge capacity drops, and are showing us that when the data-sheets say "10 year life-span" that we are getting 3 to 4 year spans at real-world usages. (Cont. in PART 2)
  • SylphhawkinsSylphhawkins Registered Users Posts: 8 ✭✭
    (Cont. from PART 1)
    I have also noted in our research that the Lithium Fluoride electrolyte paste has its' own given shelf-life, which we are finding might be around 7 years. It seems that the paste (potentially) dries out/reacts unfavourably with the organic solvents which form the SEI layer/slowly reacts with the positive electrode metals to form un-helpful metal-fluoride compounds.

    Way more research is needed in this particular area, but I highly doubt the 10 year warranties that current Australian Li-ion distributors are providing, and I hope that the distributors have insurance against the manufacturers claims if they turn out not to be correct.

    I assume that you are acting as the distributor of these cells for your solar-installation business, do you have the spec-sheets / data sheets for the exact type of cell that is inside the casing from LG?, I would love to read it to see their claim on >6000 cycles.

    Re: the DOD / SOC performance, the fantastic point about all of the current li-ion chemistries is the ~90% DOD rates and to have them still fit into their 80% capacity warranties after x-cycles, however, given the incredibly low life cycles compared to the advertised data (of all of the cells that we test and use in-house at MCB co.) when the cells are used in day-to-day activities, I advocate people away from Li-ion cells for off-grid setups, given that there is usually no problem with weight of the battery bank, and that for 1/2 the price customers can have systems that will far outstrip Li-ion's overall lifespan.
    I don't know about you guys, but here in Australia we still have no national or state recycling programs for Li-ion cells. We have to pay to have our used/dead/burnt/broken li-ion cells removed by a private company who ship them back to China for "recycling", and I have seen many references to the open-cut coal mines that litter Australia's western state being converted into open-cut lithium mines for the new battery boom that is occurring.

    Personally, I am interested in battery systems that are made of abundant cheap minerals, not rare earth metals, and especially made of materials that are easy to recycle and non-toxic to living systems and water ways.

    I definitely see the place for them in the current tech landscape of being perfectly suited to electric-vehicle transport requiring high specific energy / energy densities to reduce payload and extend drive range, but I'm not a fan of them for stationary off-grid systems.

    Going back to NiFe cell chemistries, these too have the advantage of being able to be drawn down to 90% DOD (10% SOC) repeatedly without marked degradation to the electrodes, and can withstand day-to-day overcharging.
    They do have a limited temperature operating range which is due to the wet electrolyte's properties, and they do off-gas the largest amount of Hydrogen of any of the current wet-filled battery chemistries.

    I saw one patent pending that was for a system of capturing the hydrogen produced to be used as a cooking gas to supplement the LPG for an off-grid home, very cool idea!.

    NiFe's days of operating electric vehicles are long gone, given their usual specific energy of ~30wH/Kg, they sit below Lead Acid's 40wH/Kg, and despite their great properties in situations of great DOD and high cycling rates such as in an electric vehicle, the payload weight simply makes them totally impractical. On top of this, they have rather low crank-rates of short-ciruit connections or heavy motor loads (this is due to the low solubility of the Iron-Hydroxide / Iron-Oxides which cause a chemical buffer to the cycle of charge), and so they lack the ability to drive a high torque electric motor in harsh conditions uphill at any great speed.

    I see them as perfectly situated in the market as a solid off-grid stationary battery, the kind that you pass onto your children after you die,
    and they could very well run the battery backend of the new wave of microgrid / hybrid battery system houses that are becoming popular, while Li-ion should continue to take their place as the battery of choice for low weight / high energy / high discharge-rates for electric-vehicles and will no doubt help us to transition from petroleum based vehicles to electric until such time as a high specific energy, high discharge rate battery is produced that uses abundant minerals, such as the work that Dr Goodenough is doing with Sodium-ion batteries after moving on from his patent on Li-ion and Sony Corp in the 1990's.

    I had better wrap up this post before it becomes a book, :p
     
    - Sylph H
    P.S. I am based in Melbourne, Australia (the southern-most city on the mainland)
  • mcgivormcgivor Registered Users Posts: 1,305 ✭✭✭✭
    Very interesting post, thanks for sharing your information.
      1500W, 6× Schutten 250W Poly panels , Schneider 150 60 CC, Schneider SW 2524 inverter, 8×T105 GC 24V nominal 

  • Dave AngeliniDave Angelini Solar Expert Posts: 3,333 ✭✭✭✭
    Yes Thanks Sylph!  I will try to hook you up with my Australian client Warwick if you would like. He has been offgrid longer than my 25 years and is into the chemical end professionally. My Dad was in Melbourne during ww2. It was is last land base  working in a hospital before the Navy shipped him to the Philippines. His ship was sunk inbound by a Kamikaze.

    The LG info you asked about is in this thread I started here. 
    http://forum.solar-electric.com/discussion/351297/conext-bridge-for-xw-li-ions-from-lg-and-hoppecke-announced-for-2nd-half-2017/p1


    "we go where power lines don't" Sierra Mountains near Mariposa/Yosemite CA
     http://members.sti.net/offgridsolar/
    E-mail offgridsolar@sti.net

  • SylphhawkinsSylphhawkins Registered Users Posts: 8 ✭✭
    Hi Dave, Thanks for the link to your other thread on the LG batteries, i'll jump on and take a look.
    Interesting about your father being stationed here in Melbourne during the war, what was his profession with the hospital?
    - S
  • westbranchwestbranch Solar Expert Posts: 4,877 ✭✭✭✭
    Sylph, what is your opinion on NiCd   batteries. I have a functioning 100A 12 v system in the guest cabin 20 cells and a whack more (~64 SAFT 100A cells) of them in storage that I tested a few years back and found only ~ 5 that needed to have the KOH replaced.  Also recall that they also recommended using a Mineral Oil  layer on top of the cell electrolyte for CO2 protection  - minimizing the degradation of KOH.
    Comments welcome, these are '' tough as nails'' as the expression goes, almost indestructible under normal use IMHO.
     
    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, Hughes1100 Sat Modem
    Eu3/2/1000i Gens, 1680W & E-Panel/WBjr to come, CL #647 asleep
    West Chilcotin, BC, Canada
  • clockmanfranclockmanfran Registered Users Posts: 8 ✭✭

    Sylphhawkins,

    I do not often comment on this particular forum, but ! .......................

    Thanks very much for your excellence posts, you certainly explain and describe in a very realistic practical manner. 

    And Yes,  even a dim whitted fool like me can understand.

    Thanks.

     

    Everything is possible, just give me Time.

    The OzInverter man. Normandy France.

    3off Hugh P's 3.7m dia wind turbines, (9 years running).  ... 5kW PV on 3 Trackers, (5 years) .... 9kW PV AC coupled using Used/second hand GTI's, on my OzInverter created Grid, and back charging with the AC Coupling and OzInverter to my 48v 1300ah batteries. 

  • SylphhawkinsSylphhawkins Registered Users Posts: 8 ✭✭
    Sylph, what is your opinion on NiCd   batteries. I have a functioning 100A 12 v system in the guest cabin 20 cells and a whack more (~64 SAFT 100A cells) of them in storage that I tested a few years back and found only ~ 5 that needed to have the KOH replaced.  Also recall that they also recommended using a Mineral Oil  layer on top of the cell electrolyte for CO2 protection  - minimizing the degradation of KOH.
    Comments welcome, these are '' tough as nails'' as the expression goes, almost indestructible under normal use IMHO.
    Hi Westbranch, 
    I also have a set of NiCd wet-fill cells, ~100Ah, mine are the brand "ALCAD" G8 model, do you know the manufacturer and model # of the cells you have?, are they also SAFT cells?

    I obtained these cells before my first set of old NiFe, and I found these NiCd cells on a friends farm property here in Australia, just lying around in some long grass in a pile, I have about 18 cells.
    When I found them, they were completely empty and dry, the insides of the cases had big black crusty deposits (I later found this to be non-stoichiometric black coloured Nickel Oxide). My friend told me that they had been sitting in the paddock for over 20 years untouched and that he bought them back in the 1980's from a man who selling the banks of decommissioned NiCd cells from a local medical hospital (apparently, it was part of their backup power system in the 1960's?).

    I did nothing more than simply pull out the top caps and plates so that I could remove the hard crusty deposits that were clinging to the inside of the plastic case, then prepare a NaOH solution of ~30% NaOH/H2O, fill them, string 11 of them in series for a 13.2v NOM bank voltage (1.2v NOM x 11 = 13.2v), then charge them up from the solar without a regulator, at the approximate 7-hour rate (for the 100Ah cells, I roughed this out to be a 15amp charge). I decided to use NaOH instead of KOH simply because I had 30L of NaOH sitting around in my shed, :p, I discovered that when using NaOH the cells would work as designed, but seemingly with a lower NOMINAL voltage of around 1.1v when full, instead of the classic 1.21v noted when using KOH. I don't know if the overall Ah capacity would also be reduced by using NaOH.

    9 of the 11 cells came up well over the 7 hours to around 0.7 - 1.0v, so figuring that perhaps they needed extra charge time after sitting dry for so long, (I couldn't find any great info about whether or not the Cd(OH)2 would become Cadmium Oxide like how the Iron does in NiFe cells, and would require some extra charge to push it back towards plate metal Cd(s)).

    I decided to push it for another 10 hours at 15amp (solar, done over 2 sunny days), and all of the cells came up to between 1.0-1.1v
    I was easily able to pull ~50Ah from the entire string before half of the cells had dropped voltages down to 0.7v and below.

    Having read so much about NiFe's tolerance to overcharge and discharge, I decided to just throw hours of charge at the NiCd to see if they would come up. So, a week later of full sun and easily over 1kW into the cells (adding water every day, :p) and I ended up with a bank of cells that were all between 1.10v to 1.15v
    They would quickly rise to 1.60v when under charge from them on, and I noticed that I could push them up to 1.7v using the solar direct with no regulator once they were at this stage.

    I was so impressed, I immediately put them into service running a 12v run of house lights and a DC-AC inverter in my testing room.
    -----------------
    Electrically, the cells seemed to be un-killable, however, I returned home one day after work to find that there was a surprisingly large amount of NaOH all over the floor (the floor in this testing room had yellow-lounge chip board laid on timber beams) and I discovered that 2/3 of the cell cases now had huge cracks running down the sides and bottoms where the electrolyte had leaked out from. 
    I could only imagine that this was because they sat for 2 decades exposed in the Australian sun and that the plastic had weakened. Perhaps the hardcore charging and gassing had finally pushed the old cases to their end?

    I started to find other people on internet forums saying the same things about the old cases cracking. Some of those users suggested using 2-part epoxy to coat the cases in a few layers and returning them to service. I tried this on one case and it seemed to hold up, but I ended up purchasing a range of PP plastic tupperware containers that were larger than the original cells, and melting electrode holes in the top lids and translating the plate assembly into these, since the PP is resistant to NaOH/KOH.
    Unfortunately I've only done a few of these, and so most of the bank is sitting on a shelf empty and dry again waiting for me to fix them up.

    I have tried a few times to contact ALCAD to enquire about purchasing brand new clear cases (with no plates) in them for these to be transplanted into, but I get no real response from them.

    -------------------
    The last comment I'll make is that there are specific notes from Edison, and he has a patent where we recommends this practice, that for his Edison NiFe cells, he would use the iron metal case to act as the negative electrode, while shorting out and combining the Fe (-) and NiOOH (+) electrodes to become one single positive electrode and he would charge the cells up like this to "push out" the Iron Carbonate that forms on the negative electrode over time (as the KOH electrolyte absorbs CO2 from the atmosphere and then this interacts with the negative electrode under normal charge conditions to form FeCO3). The idea being that the carbonates would leave the original iron electrode (which is now being positively charge up to Iron Oxide (II,III), and instead, migrate their way over to "the case", after which the electrolyte is flushed, the cells are rinsed and then the electrolyte replaced for normal capacity again.

    I wonder if a similar thing occurs with the NiCd wet fill batteries, since we know that the KOH is also absorbing CO2 from the atmosphere and forming K2CO3  (2KOH + CO2 → K2CO3 + H2O).
    Perhaps this means that there is also Cadmium Carbonate forming over years?, and perhaps this same idea could be used to restore amp-hour capacity to the old NiCd wet filled cells?
    Since these NiCd's have plastic cases, it would require the addition of an iron sheet into the cell, but isolated from the other plates, to be used as the new negative electrode. The Cd shouldn't migrate over to plate the Fe electrode sheet since the solubility of Cd is basically insoluble in both the Cd(OH)2 and CdO compounds. (The two different states of Cd which would occur when it is charged to either a - or + charge voltage)

    - Sylph H



  • SylphhawkinsSylphhawkins Registered Users Posts: 8 ✭✭

    Sylphhawkins,

    I do not often comment on this particular forum, but ! .......................

    Thanks very much for your excellence posts, you certainly explain and describe in a very realistic practical manner. 

    And Yes,  even a dim whitted fool like me can understand.

    Thanks.

     

    Thanks clockmanfran, I appreciate that. :)

    I certainly love and enjoy electro-chemistry and electro-physics, to the detriment of my wife and kids who are constantly surrounded by battery experiments and piles of research papers, and I give talks around Melbourne to various groups on these topics. I am generally pretty astounded at how many people in the off-grid industry in Oz have very little, if any, chemical knowledge of how the battery systems operate, but I suppose that it is simply horses for courses and there is a place for everyone. :)

    - S
  • Dave AngeliniDave Angelini Solar Expert Posts: 3,333 ✭✭✭✭
    Horses for courses...Roger Waters?
    Most of the 100's I know off the grid just want consistent reliability and to be outside away from the grid :)
    "we go where power lines don't" Sierra Mountains near Mariposa/Yosemite CA
     http://members.sti.net/offgridsolar/
    E-mail offgridsolar@sti.net

  • westbranchwestbranch Solar Expert Posts: 4,877 ✭✭✭✭
    Vintage is from the 1980's, installed in FM 2 way radio repeater sites with primitive solar panels and controllers.  Yes SAFT brand, No model #, I contacted a SAFT dealer here in Canada and he remembered that there were cells made in Canada at that time, but as usual all documentation had disappeared, other than a small 6 or so page manual that obviously assumed you would have to recharge the fluid  yourself , vs the new idea of 'turf and buy NEW ones'...
    !! and a warning about exposure to sunlight !! I'll dredge my HD to see what I still have on 'electro-paper'
     
    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, Hughes1100 Sat Modem
    Eu3/2/1000i Gens, 1680W & E-Panel/WBjr to come, CL #647 asleep
    West Chilcotin, BC, Canada
  • mike95490mike95490 Solar Expert Posts: 7,308 ✭✭✭✭
    i would like to add to the NiFe data:
    high internal resistance
    high water usage
    low efficiency, 60%
    no partial charge deterioration
    wide voltage swing from low in use, to topping off charge.  (46v-67v my 42 cell bank)

    Powerfab top of pole PV mount | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

    solar: http://tinyurl.com/LMR-Solar
    gen: http://tinyurl.com/LMR-Lister ,

  • SylphhawkinsSylphhawkins Registered Users Posts: 8 ✭✭
    mike95490, it is awesome that you have a 42 cell, 800Ah NiFe battery bank. Is that one of the new style / plastic case types from Iron Edison or one of the similar new distributors?

    Re: your comments
    -NiFe Internal cell resistance must be similar to Flooded Lead Acid, since FLA specific energy is between 40-47Wh/Kg and NiFe is usually listed at and average of 20-30Wh/Kg. Obviously considering that internal cell resistance changes between charged and discharged state.
    Lead Acid is generally noted as having only 5-15% of the total metal weight of Lead in use electro-chemically. Increasing this percentage (by making the lead sheets thin and perforated, decreasing electrode spacing with thin separators, etc..) takes a massive toll on cycle life. Internal resistance (and therefore corresponding specific energy Wh/Kg) must be generalised, 

    -I have found that the water usage varies massively depending on maximum charge voltage. From your voltage range of 46-67v it seems that you have a per/cell charge voltage of 1.595v (1.60v). Have you tried slightly lowering the max bank charge voltage to reduce the gas evolution and water loss?

    I have found:

    1.2-1.45v develops the lowest gas evolution, but the bank will slowly loose capacity over time.

    1.45-1.55v develops the average amount of gas evolution, and therefore water loss, which seems to amount to requiring water addition every 6 months (for tall cells), but seems to maintain cell capacity over the long term.

    1.55v-1.75v develops the greatest amount of gas evolution and water loss is massive, requiring water top ups every 3 to 4 weeks. Cell capacity is probably kept at 100% for the longest span, but difficult to maintain with the water loss.

    -The round-trip coloumbic efficiency is usually given at around 60%, with the Discharge efficiency given at 85% and the charge efficiency given at 60%. I think that the NiFe/Co cell chemistry could add an interesting twist to this, as it pushes the specific energy up to around 200Wh/Kg, and the coloumbic efficiency up to 95%, since I assume that the lower coloumbic efficiency of NiFe is entirely due to the water electrolysis at, ironically, the perfect charge voltage.

    In theory, Pure Iron would have a higher Hydrogen evolution voltage (and therefore less water loss at the same charge voltage, and a higher coloumbic efficiency) than iron with a carbon content. I wonder what quality / source of Iron is used in the production of modern NiFe cells?

    -Granted that NiFe is noted to work best with lower gas-evolution at a 7-hour charge rate, and that this is unlikely to be consistently deliverable from Solar PV other than in summer. The winter 3-hour peak-sun average (that we have here in Melbourne, Australia), is obviously out-of-sync with the 7-hour charge rate that NiFe work best with. This obviously either requires the "plan" of full battery charging over a 2-day period by Solar PV (in the case of dealing with days of autonomy and cloud cover) or oversizing the battery system to such a degree that the total discharge kWh per cycle, only average to around the 7-hour rate over 2 days, and would be pretty cost prohibitive.

    Obviously this wouldn't be a problem for a hydro set up, or even a fairly consistent Wind generator.

    I still advocate that Coloumbic efficiency %'s need to be coupled along with the overall battery expected life-span, ability to tolerate accidents (humans are terrible at managing systems long term in perfect order without possible fault) and cost of production/materials.

    -The no-partial charge deterioration is a great advantage, and adds it's piece to the long life span of the chemistry, like LiFePO4 and Lithium-ion NMC/NCA (between 3.2v and 4.1v).

    I have read about, but not yet tested myself, that setting up a super capacitor bank in parallel to a Pb/Pb set can help to increase the charge efficiency and reduce the high voltage peaks during charging to give a more stable charge at slightly lower voltages. I assume that this would also work for NiFe.

    The wide voltage swing is a problem (46v-67v range being a 21v difference), with the lights getting super bright in the day time, :p, and the current electronics not managing this well, but comparing with some of the other current available chemistries we can see that it is only between 6.6v to 9v greater than these in use anyway)

    A 24 cell Lead Acid 48v set, with an electrolyte specific gravity of 1.275 has a voltage range of: 45.6v (LOW at 1.9v p/cell), 50.64v (NOMINAL at 2.11v p/cell) and 57.6v (MAX CHARGE at 2.4v p/cell).This shows a voltage range between LOW and MAX CHARGE as being 57.6v - 45.6v (a 12v difference), still a big difference for standard electronics and light brightness.

    A 24 cell 48v Lead acid set using traction batteries (like forklifts, or proper deep-cycle storage cells for stationary use) using an electrolyte specific gravity of 1.200 - 1.170 would have a voltage range of 43.2v (LOW at 1.8v p/cell), 48v (NOMINAL at 2.0v p/cell) and 57.6v (MAX CHARGE at 2.4v p/cell). This shows a voltage range between LOW and MAX CHARGE as being 57.6v - 43.2v (a 14.4v difference).

    A 13 cell, 48v Lithium-ion NMC bank could be given at 41.6v (LOW at 3.2v p/cell), 47.45v (NOMINAL at 3.65v p/cell) and 54.6v (MAX CHARGE at 4.2v p/cell). So this still shows a voltage range between LOW and MAX CHARGE as being 54.6v - 41.6v (a 13v difference).

    - Sylph
Sign In or Register to comment.