Batteries and New Battery Technologies

Discussion in 'Boat Design' started by brian eiland, Mar 28, 2008.

  1. BertKu
    Joined: May 2009
    Posts: 2,521
    Likes: 47, Points: 58, Legacy Rep: 223
    Location: South Africa Little Brak River

    BertKu Senior Member

    I personnaly would put the 8 x 6 Volt cells first in series and then in those 4 banks in parallel. The same way, a manufacturer makes a battery, however the system proposed by Jeremy is new to me and may very well be a better methode on long term. If I think logical, it could indeed be a better methode.

    I will only try it with Lithium, because I have a problem with up to 1000 batteries to monitor them, and Jeremy solution is as superglue from heaven. Bert
     
  2. BertKu
    Joined: May 2009
    Posts: 2,521
    Likes: 47, Points: 58, Legacy Rep: 223
    Location: South Africa Little Brak River

    BertKu Senior Member

    Ray, let me think logical. If I put 2 x 6 volt batteries in serial and I accidental drop a spanner on the one battery's terminal and I have shortened the battery for a few seconds. My experience is that the two batteries will never be charged properly to the same level, as they are being charged in series. Thus the conclusion is, Jeremy is probably right and just that nobody ever thought about it, we did it not do it that way. Maybe if your batteries are new, same kind, to put them indeed all 4 x 6 Volt parallel and then those two banks in series. Bert
     
  3. CDK
    Joined: Aug 2007
    Posts: 3,324
    Likes: 148, Points: 63, Legacy Rep: 1819
    Location: Adriatic sea

    CDK retired engineer

    A battery maker connects several plates in parallel to obtain the desired capacity and several cells in series to get the voltage. What you want to do with 6 volt batteries is very similar and doesn't present any issues as long as all batteries are identical in condition and technology.

    The problem starts when after a number of years you discover a drop in capacity. You disconnect the batteries and will find a few of them no longer contribute but only take space. That's when you contemplate exchanging the bad ones, but should rigorously throw them all out.
    With diodes between all members of the bank you loose some power and limit the maximum current allowed, but it allows you to use unequal batteries. And you can check their performance without unscrewing the terminals.
     
    1 person likes this.
  4. BertKu
    Joined: May 2009
    Posts: 2,521
    Likes: 47, Points: 58, Legacy Rep: 223
    Location: South Africa Little Brak River

    BertKu Senior Member

    Ray, CDK has as usual it said very well. In my case I will connect up to 40 Lithium 3.3 Volt batteries parallel. However my batteries are magnetically electrical connected. Thus at any time, without much sweat, I can lift a battery out of the bank of 480 batteries and either replace or re-charge it in a separate charger. This all without I have to stop the motors or whatever.

    I agree with CDK, the single package 2 x 40 Ampere Schottky cost me 1.5 Dollar each. To make a decent 400 Ampere diode, used at 100 Ampere are only 5 Schottky diodes on a piece of 70 Micron pcb. This for you x 4 batteries banks is only, what 10 dollar ?Loss? 100 Ampere x 0.4V loss = 40 watt on 1.2 Kw usage = 3.3% loss. For me not an issue. You also have an advantage that you have basically a protection fuse. If the diode is badly overloaded by a short circuit or whatever, it wil pop.

    But for my Lithium batteries, I am definitely going to follow Jeremy's solution.
     
  5. pool
    Joined: Sep 2010
    Posts: 59
    Likes: 3, Points: 8, Legacy Rep: 38
    Location: outbound

    pool Junior Member

    The good suppliers offer to assemble a bank of batteries from the same production batch, which helps a bit with equalizing useful service life for the 2V units.
    When ordering our last 1,200Ah@24V bank, we got a nicely engineered package for the available space, complete with cabeling, ready for the yard to drop in.
     
  6. Landlubber
    Joined: Jun 2007
    Posts: 2,640
    Likes: 125, Points: 0, Legacy Rep: 1802
    Location: Brisbane

    Landlubber Senior Member

    ...thanks fellas, yes they are all brand new batteries, so we leave it as it is. Ta, JOhn
     
  7. Jeremy Harris
    Joined: Jun 2009
    Posts: 978
    Likes: 60, Points: 28, Legacy Rep: 711
    Location: Salisbury, UK

    Jeremy Harris Senior Member

    One thing I should have mentioned, Bert, was the capacity of the charge control circuit I gave earlier.

    If mounted on a heatsink, the TIP105 shunt transistors are OK for a few watts, maybe 10 to 12 W each if the heatsink is reasonably good. This means that the maximum shunt current when they are conducting (cell fully charged) needs to be below about 3 amps (3.65V x 3A = ~11W). This is fine for the final stage of the charge on a big pack, as all that will be going on at this stage is cell balancing - the pack will be almost fully charged at this point.

    I limited the current during the final balancing phase by switching the 5 ohm resistor in series with the charger. During the "fast" charge phase, before any cell group has reached 3.65V, the charge current flows through the poer FET, which is OK for up to around 30A if mounted on a small heatsink (at 30A charge current when turned on it will dissipate about 4W). The resistor that keeps this FET turned on during fast charge, the 7k5 one, and the one that holds the gate off, the 100k one, need to have their value adjusted if you are using a higher voltage. The maximum gate voltage on the IRFB4110 FET is 20V and my charger can't ever exceed about 18V. The easy way to maintain the gate voltage to a safe level (ideally between 10V and 12V) is just to adjust the values of the 100K and 7k5 resistors so that the gate can't ever get above 20V. For example, if using a 50V power supply as a charger, then resistor values of 82K in place of the 7K5 one and 22K for the 100K one. This would limit the maximum gate voltage to just over 10V with a 50V supply which would be fine.

    Similarly, the value of the 5 ohm resistor and its power rating may need to be changed if you are using a different charge voltage and current. The maximum current that flows through the 5 ohm resistor on my charger is determined by the difference in voltage between the charger maximum voltage (18V) and the battery voltage with three of the shunts turned off (and their cells at 3.3V) and one of the shunts turned on (at 3.65V). In this state my battery pack could possible be at around (3 x 3.3) + (3.65V) = 13.55V. The current through the 5 ohm resistor is then (18V - 13.55V) / 5 = 0.89A. This current drops as the other cell groups come up to balance, but is generally fine for balancing my 80 Ah pack. For the worst case of 0.89A through the 5W resistor and 4.45V across it, the power it is dissipating is 3.96W, again this reduces as the other cell groups increase in voltage with charge.

    The key thing here is the worst case difference between the pack voltage (when it has one shunt active and all the rest off) and the charger voltage. If you keep this difference to around 5V then the circuit should be OK as it is, just remembering to change the values of the two FET gate resistors.
     
    1 person likes this.
  8. BertKu
    Joined: May 2009
    Posts: 2,521
    Likes: 47, Points: 58, Legacy Rep: 223
    Location: South Africa Little Brak River

    BertKu Senior Member

    Hi Jeremy,
    Yes , I figured that al out already. having 4 parallel or 40 batteries parallel is a different charging current. In my first experiment I used a few BD676 parallel and IRFP064N. My problem is, I know very little about the LiFePo4 batteries. The Chinese people are very helpful, like I demonstrated in the e-mail conversation (Supplier kept confidential ) , but I don't get direct answers on my questions. I still don't know what the lifetime is of a LiFePo4 battery when charged with low currents i.e. 500 mA and only twelve times per year. Will a LiFePo4 charge itself when charged with low currents? Will the batteries survive, like my NiCad batteries from Varta for 40 years and are still working, or will the battery dry up after 5 -7 years, like Porta mentioned that they will dry up. In the latter, I am not willing to spent 10.000 dollar on a large battery pack. I may as well use cheaper SLABS, they also last only for 5 - 7 years. The only advantage I have with this tubing method, should I be so lucky, that new advancements will give me 30 watt or more per cell instead of 16, I have a winner with my system. So far those ones I bought (40) 2,5 years ago, they behaving themselves. In the meantime I am preparing the square tubing battery bank with an extra tube for the electronics and heat sinks . I will charge at about 20 Ampere, i.e. 500 mA each battery in that case. Will it work??? time will learn. Thanks so much for all your information Jeremy. Very much appreciated.
    Bert
     
  9. Jeremy Harris
    Joined: Jun 2009
    Posts: 978
    Likes: 60, Points: 28, Legacy Rep: 711
    Location: Salisbury, UK

    Jeremy Harris Senior Member

    I know what you mean about getting information from Chinese suppliers. They are helpful and polite, but often fail to give you the information you really need.

    The lifetime for LiFePO4 cells isn't really known, but I do know for sure that they lose capacity as they age, whether cycled or not. The general consensus seems to be that they lose capacity at around 5% per year or so if looked after. I have some that have lost around 10% per year, but they were early cells from Headway that were not of very good quality, the later ones are a bit better. I believe that a 5 to 7 year life is probably realistic, less for other types of lithium cell I suspect, but perhaps more for the newer lithium cells that are now coming out. Nissan claim that the LiMn2O4 battery in the LEAF will still give 80% of new capacity after 8 years, for example, which is significantly better than LiFePO4. I believe this figure is for a battery used in cool conditions, though, and the capacity drop with age is greater if the batteries are used in a hot climate.

    Battery life for most of the newer chemistries like lithium seems to be dominated by calendar life rather than discharge cycles. The longest life cells I know of are NiFe cells, that will usually last a few decades with little loss in capacity. They are big and heavy though, plus they use potassium hydroxide solution as the electrolyte. OK for things like off-grid domestic supplies by perhaps no ideal for a boat.
     
  10. BertKu
    Joined: May 2009
    Posts: 2,521
    Likes: 47, Points: 58, Legacy Rep: 223
    Location: South Africa Little Brak River

    BertKu Senior Member

    Thanks heaven I only bought 40 of them and not a few hundred. Is there a minimum charge current needed, to charge a near empty LiFePo4 battery?
    i.e. 1 Cell = 3.3 Volt, 16 watt, C = 5 Ampere max discharge 5C till 2 Volt. As you know, I like to charge them with about 0.5 Amp i.e. 0.1 C The manufacturer has confirmed < 1C is good, but now I need to know is this from 0.001C - 1 C or does it has a minimum. Do you have an opinion on this question?
    Bert
     
  11. Jeremy Harris
    Joined: Jun 2009
    Posts: 978
    Likes: 60, Points: 28, Legacy Rep: 711
    Location: Salisbury, UK

    Jeremy Harris Senior Member

    From the testing I've done I say there is no minimum charge current at all. I used a single lithium cell with a small solar charger (it runs a sealed home made wireless weather station transmitter) and that seemed to accept charge quite happily at a few mA. Similarly, my 80Ah LiFePO4 pack frequently drops down to charge rates of just a couple of hundred mA at times, yet the cell voltages carry on increasing (very slowly), which shows that charge is being accepted. 200mA would be a charge at 0.0025 C, so I can say with certainty that you're OK down at that level, maybe lower.
     
  12. BertKu
    Joined: May 2009
    Posts: 2,521
    Likes: 47, Points: 58, Legacy Rep: 223
    Location: South Africa Little Brak River

    BertKu Senior Member

    Thank you so much Jeremy, now I can make a decision.
    Bert
     
  13. portacruise
    Joined: Jun 2009
    Posts: 1,475
    Likes: 178, Points: 63, Legacy Rep: 218
    Location: USA

    portacruise Senior Member

    Bert:

    This reference is pretty close to my experience on various battery types over 45 years, including the latest lithiums. Browse the website and you can find answers to most questions on batteries types but not much on connecting them. http://batteryuniversity.com/learn/article/how_to_prolong_lithium_based_batteries

    Temperature and depth of discharge are two of the most important. They are even more prominent with delicate chemistries, like lithium. Even with premium MATCHED cells available for RC model racing, there will be temperature issues because it is impossible to have the same temperature throughout the battery pack. I don't know how you could correct for this regardless of electronics, so that all individual cells are charged to their proper proportion. I don't see how mechanical connections (series-parallel) can change this, much especially at higher voltages. All battery types operate with caustic chemicals, either acids or bases (alkalines), so the hazards are about equal if they happen to leak (phosphoric acid in lithiums is nasty stuff). One thing that is mentioned in various references is that it is better to undercharge constantly used lithiums and not keep them connected to their charger, unless it is set for low cutoff.

    You mention 40 year nicads, and I am a fan of nicads as well, because of their extreme ruggedness and longevity. An equalizing over charge can be applied periodically to bring all cells to their maximum level, without damage, unlike other chemistries. They tolerate and even thrive on full charges and deep discharges, so FULL rated capacity is always available without decreasing longevity of more delicate types. You have to discount the rated capacity of other types, meaning you have to buy up to 50% more capacity just to get a reasonable life. No complex electronic monitoring of individual cells, which is subject to corrosive and heat effects with time. They can be built to give at least twice the capacity of over the counter cells nicads and some high quality ones are probably still running satelites today, after several decades. My latest B&D nicads have almost twice the capacity and much lower self discharge rates. They are still preferred by some for RC racing. I think they are particularly well suited to heavy, slow vehicles like many boats that are not for racing.

    The Prius uses specially tweaked nimh that have higher capacity and ruggedness. Cadmium in nicads is extremely bad for the environment, so maybe that's why they went with nimh. Seems as the Leaf is doing something similar with lithiums, but I doubt these cells are available over the counter. China is well known for low prices on most things, but sometimes they cut corners and use false advertising, while being very personable. Caveats.

    Porta



     
  14. BertKu
    Joined: May 2009
    Posts: 2,521
    Likes: 47, Points: 58, Legacy Rep: 223
    Location: South Africa Little Brak River

    BertKu Senior Member


    Thanks Porta, Interesting site. The answers I have been looking for has been answered by you and Jeremy. Thanks very much.
    Bert
     

  15. philSweet
    Joined: May 2008
    Posts: 2,682
    Likes: 451, Points: 83, Legacy Rep: 1082
    Location: Beaufort, SC and H'ville, NC

    philSweet Senior Member

Loading...
Forum posts represent the experience, opinion, and view of individual users. Boat Design Net does not necessarily endorse nor share the view of each individual post.
When making potentially dangerous or financial decisions, always employ and consult appropriate professionals. Your circumstances or experience may be different.