Motorcycle Batteries .. AGM, GEL, Wet, Lithium Iron Phosphate (LiFePO4)

Discussion in 'The Garage' started by _cy_, Jan 20, 2012.

  1. _cy_

    _cy_ Long timer

    Joined:
    Sep 25, 2011
    Oddometer:
    6,468
    Location:
    Tulsa, Oklahoma
    learning how to properly charge and learning cold start procedure for LiFePO4 greatly increases odds of success, using li-ion batteries in our motorcycles.

    let's visit a commonly used term in batteries... "C" ... example, A123 has a recommended max charge rate of 4C. a discharge rate of 30C, etc.

    so what exactly does "C" mean. actual amp hour rating = 1C ... one A123 26650 cell = 2.3 AH @ 3.6v fully charged. 4x A123 wired in series = 2.3 AH @ 14.4v fully charged. (AH=amp hour) (26650 = 26mm x 650mm)

    4x A123 2.3 AH LiFePO4 battery pack is spec'd at 4C .. so max charge rate is 10amps.... but not so fast... folks in the RC world has been charging A123 packs at 10C 1,000+ cycles with no ill effects.

    Shorai LFX21 AH (pb eq) or 7 AH actual is rated at 3C conservatively. Shorai's prismatic cells can easily take 5C with no sweat.

    not telling you to purposely overcharge LiFePO4 cells... but these cells are very rugged and are able to withstand a huge amount of abuse. including operating in unbalanced condition.

    where all this matters is due to relatively low AH ratings of LiFePO4 vs lead acid batteries. Peukerts law says as internal resistance goes down, losses from high amp discharges goes down accordingly. this heavily favors LiFePO4 vs lead acid batteries during starting discharge when experiencing 70-185amps.

    at lower amps Peukerts losses are very close. in other words current draw by accessories like lights, heated gear, GPS, etc. discharge efficiency are close for both types of batteries.

    alternator will be constantly recharging for both types of batteries. but LiFePO4 will be drawing larger amps loads for shorter durations. using test mule R80G/S .. current delivered to battery is 22.5amp with higher output regulator @ 3,000 rpm. at that rate battery will be charged back to 85% level within minutes. then battery will take on less amps to fully charged condition.

    whew... sorry about drilling in ... but really helps to understand basic terms

    SYCL 2.3AH on top of Shorai LFX21 AH(pb eq) for size comparison
    [​IMG]
  2. Anorak

    Anorak Woolf Barnato

    Joined:
    Jul 23, 2004
    Oddometer:
    35,155
    Location:
    OAK
    Your testing is admirable but the batteries you are testing that use the pb eq ratings, can they replace a 94R battery rated at 725 cranking amps and 85 amp hours. I've been driving around with it in my 1991 Mercedes 420SEL. It will start it at least ten times in a row with the headlights on. The Lightning is rated at 4.6 amp hours. I think that's a good battery stress test.
    [​IMG]

    [​IMG]
  3. Rajin Cajun

    Rajin Cajun Been here awhile

    Joined:
    Jan 30, 2007
    Oddometer:
    661
    Location:
    The DEEP South
    My GF said my new battery came in today :clap Now I have to wait until I get home next week to try it out. :cry

    Thanks for all the help Scott!

    [​IMG][​IMG]
    [​IMG]
  4. _cy_

    _cy_ Long timer

    Joined:
    Sep 25, 2011
    Oddometer:
    6,468
    Location:
    Tulsa, Oklahoma
    doing regenerative discharge testing with Cellpro Powerlab 8 V2
    this means target battery that is normally charged becomes power source.
    roles are reversed with total discharged amps measured.

    sequence of detailed photos below, documents a discharge/charge cycle
    20 AH rated LiFePO4 prismatic battery is charged to 99% before discharge starts.

    8.5amp discharge rate measured ... about 1.5 AH so far
    too bad having trouble with PC able to communicate with charger.
    hopefully will have that issue resolved.

    note volts measured by meter is under 8.5amp load

    [​IMG]

    37 minutes, 4.367 amp hour discharged in regenerative discharge mode.

    [​IMG]

    1hr 14min elapsed... 9.4 amp hour capacity discharged so far

    [​IMG]

    1hr 35min .. 11.8 amp hour capacity measured

    [​IMG]

    2h 9 minutes .. 16.22 amp hour capacity measured

    [​IMG]

    2hr 24minutes .. discharged to 0% .... 64f degree... 18.29 amp hour total capacity discharged from 99%. rated at 20 AH actual.

    [​IMG]

    [​IMG]

    battery recovering quickly

    [​IMG]

    Charging back up at 1C = 20amps ... 13.18v after 2-3 minutes of charging

    [​IMG]

    charged received 7.18 amp at 1C = 20amp charge rate ... 24minutes

    [​IMG]

    49 minutes @ 20amp charge rate ... 16.26 amp hour charged into battery .. 80% full, charger goes into top off mode

    [​IMG]

    1hr 15minutes ... 100% charged .. 19.57 amp hour delivered to battery ... Fin

    [​IMG]

    [​IMG]
  5. _cy_

    _cy_ Long timer

    Joined:
    Sep 25, 2011
    Oddometer:
    6,468
    Location:
    Tulsa, Oklahoma
    Tulsa 6:47am, R80G/S sitting outside all night .. 27f degree .. resting voltage 13.79 (clip slipped off in pic) 10 seconds warmup with lights and heated grips on. then turned off for cranking tests.

    choke off, first crank draws 168 amps for 4 seconds. cranks over medium speed (would have started if choke was on), . next 3 sets draws 145amp for 5 seconds. wait one minute between cranks set. engine cranking over fast. next 3 sets of cranks for 5 seconds draws 138 amps, engine cranks over fast.

    watching meter for recovery over 13v before cranking. seems to be more efficient. recovery gets faster as battery warms up. after 7 sets of cranking which all would have started test mule G/S. battery is allowed to rest to measure for recovery resting voltage. This is the first chance at cold weather for weeks in Tulsa.

    Shorai LFX21 AH (pb eq) was fully charged last night with Cellpro Powerlab 8. Installed before putting bike outside for the night. Very impressive performance by Shorai LFX21 AH (pb eq)!

    7:25am .. battery has recovered to 13.19v after resting for 30 minutes. temp has risen outside to 32f degree. 10 second warmup with lights and heated grips... first crank draws 145amp for 5 seconds. turns over strong. wait about a minute for battery to recover over 13v before next crank .... another 3 more sets ...current draw goes down to 135amp, engine is cranking over fast.

    what an impressive performance by Shorai LFX21 AH (pb eq)! note each crank cycle including first one at medium speed. All crank cycles would have started test mule R80G/S if choke was set.


    27f degree 13.79v resting outside all night.. first crank 168amp for 4 seconds.
    [​IMG]

    battery warms up to 30f degree, watching voltage recover over 13v in-between 5 second crank cycles

    [​IMG]

    7:25am temp is now 32f degree for second set of crank tests

    [​IMG]

    waiting about a minute for battery to recover to over 13v between crank cycles

    [​IMG]

    Tulsa, 11:30am .. resting voltage 13.17v after recovering for about 3 hours. temp 57f degree
    this means above brutal crank cycles used about 30% of Shorai LFX21 AH (pb eq) capacity.

    [​IMG]

    Shorai remaining capacity chart
    [​IMG]
  6. 2old2Bbold

    2old2Bbold was 2bold2getold

    Joined:
    Dec 10, 2011
    Oddometer:
    1,969
    Location:
    DFW TX
    It looks like Lithium batteries need a 14.3 volts charging rate. This is a higher level than some bikes produce. My old Transalp puts out about 13.8 at it's highest point, observed. I noticed that a common problem, from reading the Transalp sites, is over charging and boiling the battery dry. The general concensus is the problem is resistance in the voltage sensing circuit to the regulator/rectifyer, and the cure is to clean all of the connectors and/ or run a new wire from the RR to the battery so there's no resistance build up. So if too much resistance to the RR causes the regulator to charge higher, then why not install a quality variable resistor in that circuit, and an accurate volt meter on the insturment pannel, so you can set the charge level at an optimium level. I suspect newer bikes have different configurations, and wouldn't allow this.
  7. _cy_

    _cy_ Long timer

    Joined:
    Sep 25, 2011
    Oddometer:
    6,468
    Location:
    Tulsa, Oklahoma
    it's becoming apparent how an existing motorcycle's 12volt charging system interacts with LiFePO4's higher operating volt range needs to explored. this will be old hat for some, but how a charging system works will be covered shortly.

    above is but one of the difference between real world testing vs load testing.
    ____________________________

    followup next day without touching battery resting voltage 13.17v. went for a short night ride ... about 15 minutes to a riding buddy's house.

    then another 20 minutes to a gas station, then home. no heated gear, only lights. took 1.3gal to go 71 miles or 54 mpg. been really trying to go easy on the throttle. normal riding nets 33 mpg. In other words, short ride combined with low RPM is not a problem charging LiFePO4 battery back to full charge with higher 14.5v voltage regulator. (no heated gear on)

    next morning resting voltage 13.7v .. fully charged
    a LiFePO4 battery fully charged to 14.4v will settle to about 13.7v resting voltage overnight.

    according to Shorai's chart above. voltage drop from fully charged 14.34v to 13.3v represents 10% of total capacity. then discharge voltage is extremely flat for next 80% of remaining capacity.

    [​IMG]
  8. Antigravity

    Antigravity Been here awhile

    Joined:
    Sep 6, 2010
    Oddometer:
    161
    Location:
    Los Angeles
    Hey Bold, Actually if it was indeed 13.8v max output and that was in fact consistent, that would be fine for an A123 based lithium battery... it does not need to be topped off at 14.4 or 14.6... the nominal voltage is 13.2v for the A123 Based packs and that is fine for the performance required of the battery, and that performance would be great at 13.2v. After a full charge up to 14.7 it will drop down into the 13.3-5 range by the next day or so, and that will still flip the bike extremely well...no one should be looking at the 14v range as the realistic operating range of the battery though it is a nice charge to point.. You want the bike to start effectively at is nominal voltage and a little below that because a majority of riders actually let the bike sit a couple days/weeks in between rides...so they really need it to work even after a little parasitic drain.

    This goes back to my Power is King rule....For example...if you have an 240 CCA battery at 13.2v it will have trouble starting a bike above 1300cc.... but if you have a 360 CCA battery at 12.8v it will still flip that 1300cc bike even though its only at roughly 70% capacity... this is because it more powerful it can deliver more power with a lower state of charge.... So in fact voltage is only relevant to a certain degree itself...which I will discuss later....
  9. JoelWisman

    JoelWisman Long timer

    Joined:
    Apr 22, 2009
    Oddometer:
    1,392
    Location:
    STL, MO, USA
  10. _cy_

    _cy_ Long timer

    Joined:
    Sep 25, 2011
    Oddometer:
    6,468
    Location:
    Tulsa, Oklahoma
    @Joel thanks for dropping in ... please consider merging your information here. have always considered you one of the best technical resources on ADV. my prediction is your experience with computer controlled electronics will be invaluable when drilling in interaction between LiFePO4's higher voltage range and a modern motorcycle.

    this is brand new ground we are covering right here on ADV!

    if you go back in this thread... most of your info posted cannot help but overlap with what's already posted. for instance due to LiFePO4's flat discharge curve falling within a very narrow range. accuracy of your meter becomes paramount.

    the unbalanced cell issue has been briefly covered in differences between prismatic cells and cylindrical cells. with a promise to drill in at another point. to me it's very important to convey how things work in the least technical manner possible.

    Glazed eyeballs ... means info gets gleamed over instead getting read. Education on how LiFePO4 chemistry works goes a long ways towards a successful implementation of LiFePO4 in our motorcycles.

    LiFePO4 in a race bike is a no brain fit. But successful use of LiFePO4 batteries in an adventure bike requires knowing your charging system is mated to LiFePO4's higher voltage. just as important ... learning how to start your motorcycle in cold conditions with LiFePO4 is a requirement.

    question that comes to mind ... is it worth going to all this trouble to save 10lbs off my R80G/S?

    getting ready to go for a ride from Tulsa, OK to Elgin Kansas with Shorai LFX21 AH (pb eq) doing the duties.

    --------------

    Martel MC-1000 Multifunction Calibrator is a handy tool to have around when accuracy is a concern.
    Cellpro Powerlab 8 is the most powerful battery workstation available. surprisingly affordable compared to Fluke prices :eek1

    [​IMG]
  11. 2old2Bbold

    2old2Bbold was 2bold2getold

    Joined:
    Dec 10, 2011
    Oddometer:
    1,969
    Location:
    DFW TX
    Thanks Antigravity. That's kinda what I suspected. Like most of us ordinary ADV guys, I was probably going to just buy one (or not) and see if it worked---but (there's always a "but") a lot of my rideing is in CO above 8000 ft with morning temps near 30 degrees, so $200 bucks is a questionably amount to experiment with.
  12. JoelWisman

    JoelWisman Long timer

    Joined:
    Apr 22, 2009
    Oddometer:
    1,392
    Location:
    STL, MO, USA
    _CY_ I am following your thread and will pipe up any time I have something to add or a question, such as, WOW, this battery is only 6 months old, how did it get in such an imbalanced condition?

    Also I love that your riding an R80GS! Between you, me, and 10,000 ADV members, I think the R80GS was the best air cooled GS ever. The 100 was a step back, the 1150 a claw foot bathtub on wheels, and the 1200 a fat pig, elegant on super slab, surprisingly good on the twisties, and like any other pig, likes to wallow on it's side in the mud.

    Nice :) I couldn't say things like that while I worked for BMW :D

    It is also pretty easy on the amps draw being low displacement, low compression and very smooth for a 2 banger compared to the downright buzzy F800.

    The thread I started is to answer specific charging system and battery questions relevant to the F8 for an adventurer that is a personal hero and sent me his Shorai, but as I said. I'll post here when the question is generic.
  13. Anorak

    Anorak Woolf Barnato

    Joined:
    Jul 23, 2004
    Oddometer:
    35,155
    Location:
    OAK
    Eight A123 cells start the 4.2 liter V-8 in my Mercedes at 46 degrees ambient with no lag and will do it 10 times in a row with the headlights on without slowing. Your 12 cell battery is almost absurd.
  14. _cy_

    _cy_ Long timer

    Joined:
    Sep 25, 2011
    Oddometer:
    6,468
    Location:
    Tulsa, Oklahoma
    just got back from a 236 mile bike ride ... super windy conditions, not bad as yesterday driving up to Kansas. Gusts 30-50 mph... starting out... 40-60 mph gusts towards Kansas.

    needless to say crosswind sections were ridden slower down towards 30 mph, leaning sideways. winds were from south, so due north sections were not as bad... still the most windy conditions I've ridden in so far. got to be good training for Patagonia.

    took a small Fluke VOM, measurement on arrival was 13.40v temp low 70's only lights on, engine rev's low .. big time fun riding gravel with 50 mph gusts coming at you sideways. winds were so bad, detoured several times on side roads that led to gravel.

    winds blew hard all night only slowing down by 7am next morning. then winds started back up as the day warmed up. not bad as yesterday, but strong headwinds all the way back. rode back roads much as possible back. did about 35 miles of gravel exploring back Kansas farm roads.

    13.57v resting after getting home 2-3 hours ago. higher (14.5v) voltage regulator seems to be doing a good job of charging without overcharging. Shorai LFX21 AH (pb eq) didn't miss a beat the entire trip.

    [​IMG]

    13.57v resting after getting home 2-3 hours ago. Shorai LFX21 AH (pb eq) didn't miss a beat the entire trip.

    [​IMG]
  15. _cy_

    _cy_ Long timer

    Joined:
    Sep 25, 2011
    Oddometer:
    6,468
    Location:
    Tulsa, Oklahoma
    Let's explore the wealth of available data for AGM battery's cold weather performance. which would be battery of choice over Gel and wet PB before LIFePO4 came along. Peukert's law is also explained in detail link below.

    note at 15f degree temp losses are about 7%. fortunately Motorcycles don't have to worry about starting at much below 15f. couldn't say the same for Snowmobiles.

    in terms of cold weather performance for Motorcycles. AGM sets the bar to beat. it's also 10+ lb heavier than LiFePO4 for R80G/S.

    Temperature Losses

    [​IMG] Generally speaking, as a molecule is reduced in temperature, the less active atomically it becomes. This affects many exothermic (heat inducing) reactions including the power generated by lead acid batteries. In cold Colorado winters, be they AGM, GEL or flooded in composition, all batteries efficiency are negatively affected. The graph shows the negative correlation between temperature and battery capacity in an AGM battery. Notice how a 100 amp/hour battery at 90° F has only 55% of that capacity at 4° F and 43% capacity at -22.

    http://www.bdbatteries.com/peukert.php
  16. Antigravity

    Antigravity Been here awhile

    Joined:
    Sep 6, 2010
    Oddometer:
    161
    Location:
    Los Angeles
    I have the same results Anorak, and have a 12-cell under the truck seat just for emergencies... But there is factor here that allows that great performance in cars and it is usually the great starter reduction gearing in cars with that super large ring gear on the flywheel/torque converter area, and also the efficiency of the system in general with the nice grounding and fat cable usually a miniumum of 4 awg and the starters also seem very efficient. Some bikes also have great starter reduction gear though it varies alot and the efficiency of the starter plays a part too... for example our 4-cell starts the GSXR 750 fine... yet it is slower to turn over the CBR 600.... it turns out the CBR has not as good starter reduction gearing and also a starter that isn't as efficient.

    Interesting fact----
    I heard on the new super killer Ducati Superbike they did some engine efficiency work to allow for easier turn over so they could use a much smaller battery than normal... so the bike is able to use a smaller battery to achieve the good starting results by making the engine easier to turn over on the start attemp,t where-as normally twins are a harder turn-over than inline 4 cylinders....

    I guess the point is just that it not just the battery but the vehicle also that can make for a harder start...
  17. Anorak

    Anorak Woolf Barnato

    Joined:
    Jul 23, 2004
    Oddometer:
    35,155
    Location:
    OAK

    I've got the battery wired to the bolts on the regular lead cable ends that would normally connect to the posts so the connection is compromised. Also, would a YTX-14BS start my car repeatedly?

    Interesting rumor on the Ducati. The early 1098's had starting issues. Part of it was due to too much ignition advance during cranking caused by a software glitch but Ducati made a running change to the starter drive ratio also.
  18. Antigravity

    Antigravity Been here awhile

    Joined:
    Sep 6, 2010
    Oddometer:
    161
    Location:
    Los Angeles
    The Ducati fact I brought up i is from the new Cycle News write up, and the Ducati engineers explain what they did, pretty cool stuff and I guess the bike is the best thing since sliced bread according to all the raves... cool stuff.

    As far at the YTX14 doing the repeated starts..... Maybe but probably not... You may know this stuff but for those who don't.....One of the issues with lead acid batteries is that they do not handle high amp discharges well, nor do they handle higher constant current draws very well... The amp hour rating for a lead acid battery is given on a very low amp draw (cant remember the actual number, maybe its one amp)... but the fact is leads capacity/amp hours will be diminished much more by high amp discharges than a lithium battery would be (using roughly equivalent batteries).
  19. Antigravity

    Antigravity Been here awhile

    Joined:
    Sep 6, 2010
    Oddometer:
    161
    Location:
    Los Angeles
    Hey Cy and the hard core adventure riders...

    I'm not understanding something, most likely because I'm not an Adventure Rider...but it seems sort of obvoius...

    I keep seeing testing of voltage, an comments about starting in cold weather and questions, if Lifepo4 can handle a trip around the world. But what I'm not getting is that you have a technology that due to its extreme small size and high power would allow for a redundant battery system, yet I don't see anyone speaking on that...

    If I was going into the deep places of Africa or where ever I would take two 12-Cells or a 12 and an 8 and laugh away that day knowing I have a system an a back-up. I would want security...Though I preach about getting the most power you can having a redundant system would trump that and provide for the massive power and amp hours if I needed itplus give security . Keep in mind the battery that sits doing nothing is not going to lose voltage/energy during the whole adventure... but if some trouble should start just a have a quick parallel connection for starting in the most freezing of weather at 720 Cranking amps and laugh at Jack Frost or just a back up source of power for a GPS...
  20. _cy_

    _cy_ Long timer

    Joined:
    Sep 25, 2011
    Oddometer:
    6,468
    Location:
    Tulsa, Oklahoma
    totally agree... light weight of LiFePO4 batteries makes carrying a backup feasible. in post #117 covered that when weighing SYCL battery. fortunately R80G/S has a kick starter so already got a manual backup. which almost no modern Adventure bike has any more.

    testing data has shown Shorai LFX21 AH (pd eq) is up to the task reliably starting a motorcycle down 30f degree. which still doesn't leave enough margin for error for an adventure bike in case one should be using heated gear combined with a short ride. By doubling AH rating with Shorai 36 AH (pb eq) 540 cranking amps. it barely breaks a sweat starting R80G/S.

    for the short trip to Elgin Kansas. predicted low temp was 40f + so starting R80G/S was not even a concern with Shorai LFX21 AH. Data on how charging system fared with 14.5v voltage regulator was what I was after. Just so happens due to high wind conditions, average speed for 100+ miles was much lower than normal highway speeds. ended up with ending resting voltage of 13.4v after 100+ miles. fully charged resting is 13.7v after charge dissipates from 14.5v with higher regulator.

    got a trip planned soon ... R80G/S down to Florida from Tulsa, OK with Shorai 36 AH (pb eq) and 14.5v voltage regulator for a proper shake out ride. while SYCL 2.3 AH is the ideal battery for race bikes. it'd make an excellent backup battery for Adventure bikes too.