AA size 3.2V LiFePO4 rechargeable - solar charging circuitry?

[manuka]

I'm considering a portable wireless 08M2 PICAXE project to be housed in a 2 x AA switched battery box, powered by a single LiFePO4 rechargeable 14505 cell. These lightweight cells are rated at 3.2V 600 mAh, 0.6A rate, 2.2Wh, ~2000 cycles & are standard AA size. Although the capacity is low (but adequate as current drain will max at ~50mA & normally be "SLEEPing") they retain a pretty flat 3.0-3.2V discharge curve. The idea is to use the other battery box cell space for the PICAXE circuitry.

I'm well aware of the increasing number of AA sized LiFePO4 now available (often for just a few $ & even turning up in better solar garden lamps),but my pondering relates to their charging needs. Compared with other Lithium's they seem pretty charge tolerant, but as I may use solar "top ups" (with naturally a varying supply), I'll hence not have the luxury of a specific 3.6V mains smart charger.

Although Google shows diverse LiFePO4 smart ICs & circuitry, many are overkills for my modest needs, while others tend almost laughably simple. Hence any LiFePO4 personal experiences & cost effective charging insights would be welcomed ! Stan.

 

[premelec]

Hi Stan... my recent considerations about using these cells came upon 2 things - a large part of their charge is in the last tenths of volts AND don't overdischarge them or they are screwed... There are a lot of step up converters on ebay showing low voltage turn off to keep the latter from happening - and you can use any good stable regulator to keep overcharge OK [assuming some reasonable current limitation]. One thing I haven't found data on is temperature change in top charge voltage - perhaps someone here can offer advice on this [they definitely aren't like lead acid's high temperature variation...]... I did find a step up unit on ebay with <.2ma idle current but no low voltage cutout and I haven't fully tested it yet...

 

[manuka]

Premelec: Many thanks - I've considered both "last tenths of volts" capacity & (LM4041 based) over discharge OK, but (like you) haven't yet looked at temperature.

Compared with other Li cell types (Li-ion & Li-Po), anecdotal user feedback says LiFePO4 are very tolerant. I've shamefully had little to do with them however, which is largely the reason for my posting. I'm increasingly taken with their relative safety, low cost, high terminal voltage & (compared with NiMH/NiCd) nominally very long cycle life. Stan.

 

[premelec]

That circuit looks like something I've tried but not fully tested using a P MOSFET and TL431 [no bipolar like yours] where you have to either have capacitor or push button start as the P hold up is derived from it's Drain so it cuts out on low voltage till reset. The batteries I have were from various old computer packs etc so I don't know their chemistry. There are lots of charge your phone up converter devices with USB output connector, for single Li input, but one 'feature' is a minimum current to keep them on... There have been some discussions here about the LiFePO4s I think - finding hard data sheets on these ubiquitous cells has so far eluded me.

 

[manuka]

I've now pinned down 4 different brands of 14500 AA sized LiFePO4 ("LFP") rechargeable cells - a nameless locally sourced "blue", Coolook, SoShine & Powerlion. Amazingly the latter name brands are freight free from Hong Kong - see attached. LiFePO4 cell prices have been MOST gratifying (averaging ~US$2 each), & even their dedicated chargers are <US$10. Tests should reveal if they cut the mustard however.

It's rather to early to firmly say, but indications are that 3.2V LFP may be the future cell of choice for many multiple AA users -a dummy AA used in series with a single LFP should give real benefits in traditional 2 x C-Zn applications (flashlights, AA powered digital cameras etc). 4 x LFP will give 12.8 V as well. Stay tuned!

Extra: I can just see those dummy placeholder cells suiting April Fool variations on "long weights" & "tins of elbow grease". I wonder how many end up in the rubbish bin before folks realise their function!? Stan.

 

[srnet]

What sort of capacity are the cells you have found ?

The AA sized LiFePO4 I have seen are in the 800maH region, which is not a lot compared to a decent NiMh.

 

[manuka]

LFP cell capacity indeed looks only about 800mAh for AA types, but of course they've almost 3 times the voltage (3.2V versus 1.2V). LFP power availability hence looks comparable to decent NiMH/NiCd, but in a single cell rather than three.

 

[SAborn]

Stan, im not sure if im on the same page for the cells your are looking at using, but a while back i came across some lithum cell recharge PCB's on ebay and saved the location, you might like a look here as it could suit your purpose ...

http://www.ebay.com.au/itm/1-pcs-PCB-3-6V-16340-18650-Li-ion-Lipo-Battery-Pack-/180675632196?pt=AU_B_I_Electrical_Test_Equipment&hash=item2a111b5c44

and a far bigger range here...........

http://stores.ebay.com.au/GreenForceStoreAU/_i.html?_nkw=Li-ion Lipo Battery&rt=nc&_sid=215589483&_trksid=p4634.c0.m14.l1513&_pgn=2

 

[manuka]

SAborn: Thanks - but those are all for classic 3.6V Li-Po and Li-Ion types It's the lesser known 3.2V LiFePO4 ("LFP") that I've been focusing on, since that type is inherently safer that Li-Ion/Li-Po and is stated as having a much larger cycle life (>1000 times).

 

[Paix]

Of course, LiFePo4 were the battery technology that Boeing didn't select for their Dreamliner and consequently ran into a few problems, unless I got that wrong?

 

[manuka]

Yes -bad PR & VERY costly for Boeing - the mere thought of such batteries going ballistic when half way across the Atlantic is sobering ...

 

[srnet]

Sold to Boeing on the basis of lower operating costs apparently .........

And of course it saves the weight of approx two small bits of luggage, 50lbs, per plane.

So given the cost and weight savings, you can understand why they chose Lithium Ion.

 

[Paix]

I had also read a short time after the incidents, that the same technology had been rejected by automotive manufacturers for use in vehicles up to two years previously for because of the volatility problems, in favour of the LiFePO variant. Quite why the aircraft manufacturer decided that this information wasn't important given it's publicity is not quite understood.

In one of the more safety conscious industries on the planet, it beggars belief that they seemed to live in ignorance of the facts. I guess that someone will be quietly picking up their P45 on the way home . . .

 

[manuka]

Fortunately no 787 "Dreamliner" crashes or fatalties occured,but the ongoing costly (rumoured US$50 million a week) 787 grounding has at least served as a great wake up on Lithium cell types & associated issues.

FWIW the design of Boeing's Yuasa made Lithium Cobalt Oxide batteries dates from ~2005, which was BEFORE safer LiFePO4 types became available.

 

[Paix]

That of course is very significant as superficially the decision to use Lithium Cobalt instead of LiFe battery technology in the light of the automotive rejection of the same two years ago seems absolutely mad, but your point that the Dreamliner batteries were essentially being specified around eight years ago, when most of us were totally unfamiliar with even the basics of the technology never mind the flavours that were yet to emerge.

Eight years ago how many of us understood the ins and outs of the Euro, so very clear to us today in retrospect.

As always Stan, you have put the perspective on the discussion. Thank you for that. The decision has a significant cost in terms of operating loss and damage to reputation. Hopefully looking back in a few years it will be but a blip on the Boeing balance sheet as they jockey with Airbus for slices of the aviation market. It just goes to show that the decisions made today are apt to rattle around a lot in the future if there is more than a margin in it. Particularly where safety is concerned.

That which doesn't kill us makes us stronger, or at least a lot wiser it seems, but then 20:20 hindsight is pretty much a universal capability. How many little guys as a result of the Boeing experience will tread the safer path with nary a thought of the cost that kept them on the straight and narrow? Out of adversity my friends . . .

 

[manuka]

I've now had my arm twisted for an up to date LiFePO4 ("LFP") article,and (according to my rapidly amassing draft material) LFP attributes were at least known about even before the y2k era -

1979 American researcher John B. Goodenough working at Oxford university perfects Lithium-ion rechargeable battery technology using Lithium Cobalt Oxide (LiCoO2) and Lithium Manganese Dioxide (LiMn2O4) based cathodes but the patents for the invention were awarded to the UK Atomic Energy Commission (now AEA Technology) who funded the research. Sony of Japan however were the first to commercialise the technology. Goodenough did not benefit financially from the patents but in December 2000 he was awarded the Japan Prize (and $450,000) by The Science and Technology Foundation of Japan for his invention.
1996 Subsequently working at the University of Texas,Goodenough patented the more stable Lithium Iron Phosphate (LiFePO4) cathode chemistry.

My first involvement with LFP was in 2008 with the now infamous OLPC (One Laptop Per Child), when I contributed significant power drain & solar charging experiences for it's LiFePO4 battery pack. However the OLPC featured a BMS (Battery Management System) & I was not concerned at that stage with standalone LFP cells or their treatment. Stan.

 

[manuka]

Update: LiFEPO4 "LFP" cells & charger have arrived hassle free & VERY well packaged (each AA was staunchly bubble wrapped) from Hong Kong. The US$22 total is a real deal - it covered the dedicated AAA/AA LFE charger, 6 LiFePO4 cells, 2 unlabelled dummies AND free p&p -itself HKD40 (~US$6) according to the postage stamps.

Alert for other importers. It's a tad concerning however that FASTTECH's Customs Declaration (of "Adapter, Electronic Parts, Case") did NOT indicate Lithium cells were within, & that the "Gift" value was only HKD22 (~US$3). I've been importing assorted hardware for over 50 years & am well versed in diverse labelling skullduggery,but the potential concerns arising from airfreighting lithium cells (even these "safe" LFP ones) justify lucid labelling IMHO. NZ & China now have extensive mutual tariff free trade, but such a liberal approach may backfire & NOT be tolerated in even nearby "nanny state" Australia.

Anyway -all cells were well charged & delivered ~3.2 V under typical loads OK. Revitalisation of 2 & 3 x AA NiMH powered devices at hand (cameras,flashlights & battery shavers especially) has been dramatic. 2 x AA powered Canon digital cameras have notably sat up & resumed polonged snapping duties using just a single LFP & dummy. Naturally a clear head is needed to avoid over supply nightmares-it'd be easy to stuff 2 x 3.2V (= 6.4V) total into a digital camera that expected under half this supply!

The 2 x AA charger performs admirably, with it's individual red LEDs changing to green when the cells reach 3.6V. Diverse experts say 2.5V is the LFP low voltage cutoff, so I'm still considering cost effective low voltage sensing circuitry. Although dedicated ICs exist for this,by chance white LEDs usually stop emitting light at 2.5V, so some simple LED-LDR circuitry may have mileage? But rather than such a crude "OFF when too low" approach,"ON when too low" TL431 programmable shunt regulator (~25 cents) setups naturally have greater appeal. (Check circuit below - yet to be modified for LFP cells)

My initial quest was to use a gutted 2 x AA battery box, but may change to a 3 x AA type instead, as this more traditional size offers greater room for single LFP powered PICAXE circuits. Naturally I'm pondering some (solar powered?) circuitry show off, perhaps using a PICAXE-08M2 controlled Dorji ASK transmitter (which is good for supplies 1.8-5.5V). Any other proof of concept 1 or 2 x AA LiFePO4 circuitry suggestions welcomed. Stan.

 

[premelec]

As I mentioned [post #4] I've used a TL-431 + PMOS - you could use a BJT or PMOS with gate/base through white LED to V- ; not very accurate but could be ok [need gate to source R if using PMOS]

 

[JimPerry]

Update: LiFEPO4 "LFP" cells & charger have arrived hassle free & VERY well packaged .

Where did you get them

 

[eclectic]

Where did you get them

Possible = post#5 Basket summary pic?

e

 

[manuka]

Yes - sourced from FastTech in Hong Kong- refer their LiFePO4 page.

Note: Electrochemical cells,quaintly still known as A,B,C,D variants from the ancient valve electronics era,are increasingly being titled by their metric dimensions. The "AA" 14500 size is nominally 14mm wide x 50mm long. You'll occasionally see 14505 which are 14mm x 50.5mm long - that extra ½mm can be crucial for making reliable battery contact of course. AAA sized cells, which are also available in LiFePO4 chemistry, are known as 10440 (10mm x 44mm)

More powerful 18650 offerings (thus 18mm x 65mm & commonly used to stuff laptop battery packs),are of course much larger & hence will not fit common AA sized devices,battery holders or chargers.

 

[manuka]

Herewith a proof of concept 3 x AA battery box housed single AA LiFePO4 powered PICAXE/Dorji circuit. I've pondered assorted low voltage cutoff sensors & PICAXE driven software (especially the 08M2 "CALIBADC" command), but initially have just used a dumb test LED for simplicity. It reminded me of the external "see the power" powercheck strips featured on some Duracell alkaline AAs in fact.

I'd only 5mm white LEDs at hand (although naturally a smaller white could be used), but all showed significant dimming below 3V & (MOST usefully for LiFEPO4 !) a total light cutoff by 2.5V -a near perfect matching! Such a simple test could be included with dumb circuitry (discretes, 555 etc) housed in a partially gutted LiFePO4 powered box too.

The breadboard section makes for great initial convenience, & you get 2 sections from a normal small breadboard, but Kiwi Board/Vero/PCB could be used for final soldered versions. A PCB "finger" would be needed for a single AA space in a 2 X AA battery box of course.

Quite aside from the LiFePO4 insights I've been motivated by the "potential" of the switched AA battery box, as discrete switches & suitable project cases can otherwise end up costing more than the internal electronics!
Stan.

 

[AllyCat]

total light cutoff by 2.5V -a near perfect matching!

Hi,

Yes indeed. These (AA) batteries are often used in (and sold specifically for) the better quality Garden Solar Lights, because there is no need for "intelligent" control, and the high cycle life. Two examples from this side of the globe, although the Maplin "noname" example is only a guess (I may have to buy one just to find out).

http://www.batteriesplus.co.uk/acatalog/3.2V_AA_14500_LiFePO4_600mAh_rechargeable_batteries_for_Solar_Lights.html
http://www.maplin.co.uk/replacement-lithium-battery-for-solar-lights-676623#overview

However, CALIBADC really is worthy of consideration, it's probably the nearest thing to a free lunch! No need for any additional components, not even a spare pin, and just a few lines of PICaxe code. The only tip I'd give is don't be too ambitious and try to use 10-bit A/D conversions; 8 bits will make the division/scaling easier and sufficiently accurate. You could start to signal a "Low battery" in your RF transmissions at (say) 3.1 volts and adopt "protection" methods (infrequent/no radio transmission, PICaxe naps, etc.) below 2.6 volts.

Cheers, Alan.

 

[AllyCat]

Hi,

Yes - sourced from FastTech in Hong Kong- refer their LiFePO4 page

Sadly, I have to report an email just received from FastTech. I was aware that our (UK) Royal Mail had taken a "strict" view about carrying Lithium-based cells (of ALL types) but it looks as if all shipments (using air transport) may now be affected:

Dear Alan,

We’re writing to tell you that due to current strict inspection of the aviation security, packages containing lithium battery will have high possibility of being return to us.

Considering the risk of being returned, we are compelled to take out the battery 1208603 and send the other parts of this order. We will arrange refund once gets your confirmation.

Sorry for being unable to arrange the shipment for the orders that included lithium batteries, because there were too many orders being affected. Your understanding would be highly appreciated.

Kind regards,

Jessica

Customer Advocate

FastTech - gadget and electronics
http://www.fasttech.com
support@fasttech.com

IMHO all rather daft when LiFePO4 cells are potentially safer than "normal" lithium cells. And that a LiFePO4 in a well-sealed insulating bag is banned, whilst higher-energy Li-Ion batteries inside "live" consumer items are still being carried on aircraft in their millions.

Cheers, Alan.

 

[neiltechspec]

Wasn't there a recent incident (within the past 6 months from memory) where a cargo plane came down somewhere near Dubai, said to have been caused by a fire started by Lithium Iron cells.

It was posted on the Aviation Herald website. (this website can make you a bit paranoid about air travel - beware !)

No doubt this kind of thing will have an impact.

 

[srnet]

Wasn't there a recent incident (within the past 6 months from memory) where a cargo plane came down somewhere near Dubai, said to have been caused by a fire started by Lithium Iron cells.

http://en.wikipedia.org/wiki/UPS_Airlines_Flight_6

 

[hippy]

IMHO all rather daft when LiFePO4 cells are potentially safer than "normal" lithium cells. And that a LiFePO4 in a well-sealed insulating bag is banned, whilst higher-energy Li-Ion batteries inside "live" consumer items are still being carried on aircraft in their millions.

I suspect one of the problems is that couriers and carriers don't have the resources nor skill to determine what any particular battery type is nor assess its safety or risk so the safest and easiest course is to prohibit all of generic type.

A phone carried by a passenger is likely to be considered less of a safety risk than a battery 'thrown in a cargo hold' which may suffer extreme environments and some rough handling.

 

[John West]

Eight years ago how many of us understood the ins and outs of the Euro, so very clear to us today in retrospect.

Actually, I know just about nothing regarding the Euro, other than that it's some sort of money. But thanks to these threads I'm learning a lot about new battery (cell) technology.

 

[neiltechspec]

That incident was longer ago than I thought. Must have been a final report update or something similar I read then.

 

[manuka]

Not entirely unexpected in light of the shipping cautions I'd expressed back at post #17... However NZ & China/HK have free trade and seamless shipping,so this alert may hence be country specific.

None of my previous FastTech deliveries have encountered this,nor have Kiwi model plane mates who regularly import fire breathing Li-Po battery packs from other Asian outlets (especially Deal Extreme ). Our experiences show China/HK sourced Li-ion/Li-Po/LiFePO4 cell/battery packaging is extremely protective and robust.

I need further items anyway so have hence just ordered assorted LiFePO4/Li-ion 14500 cells + dummies & chargers (as below) from FastTech & will advise of proceedings. Stan.

Update: It's less than 12 hours later, but FastTech have just emailed to say the entire order has been processed OK & is now already en route to me!

 

[manuka]

Sigh-many weeks later: The ordered items never arrived! But,just as I pondered a complaint,FastTech notified of a full refund. It transpires shipping regulations on Lithium cells may have indeed tightened (although I probably triggered this by including a Li-ion cell in my order). FastTech now say even most LiFePO4 cells need to be EXPRESS SHIPPING ONLY, at significant cost (~US$15 a package). Kiwi Li-poly mates now indicate similar issues.

However they've a new "CoolWorld" LiFePO4 14500 offering still showing free shipping. This may be an oversight, but I've ordered a couple (~US$5) as both a shipping & evalution trial. Given such colourful outer wrappers they'll be at least easy to identify!

EXTRA: Thanks to some arm twisting on my part Aus/NZ outlet Jaycar/Electus are finally about to stock both Li-ion & LiFePO4 "Powertech" cells. Although their prices are a tad electrifying,at least 14500/18650 will now be easily available here. They quote stock items -

SB2305 &#8211; 14500 LiFePO4 600mAh $9.95

SB2303 &#8211; 14500 Li-ION 750mAh $9.95

SB2307 &#8211; 18650 LiFePO4 1600mAh $17.95

SB2403 &#8211; 18650 Li_ION 2200mAh $19.95

I remain highly enthusiastic about LiFePO4 cells,and have found no signs of flagging in any I purchased earlier this year. One ran near completely flat in a torch left on for some days, yet seems to have recovered well. Li-ions however still concern me, both for their energetic contents & lower (100s) of charge/discharge cycles. Stan.

 

[Paix]

Hello Stan,

http://www.youtube.com/watch?v=1NU6LjenhKM&feature=em-subs_digest 5.46 or 7.10 to 8 minutes for the nitty gritty.

In essence Bruce Wilson, Tokora, NZ is reporting a problem with LiFeP04 batteries from Hobby King vie FedEx. There was apparently a work around, but not for the battery that he had ordered, a 6 cell 5000mAH, as this apparently exceeds the FedEx 100W/hour maximum limit.

I don't know what happens when they have a large consignment of individual orders for lesser sizes to load on an aircraft, but . . .

You seem to be well under the magic figure. So maybe mainly the RC and RCFPV guys getting hammered. One way or the other!

 

[AlastairD(Aus)]

Slightly off topic but I am in the process of building a house battery system for the off road 4wd motorhome that I am building on an Isuzu truck chassis.

I have 8 x 160AH LiFePO4 cells that are configured as 12v (actually nearer 14v most of the time) 320AH. I have spent many hours reading all of the available info, some of it poor, on the web. The charge/discharge curve is very flat with sharp voltage rises and drops at each end. When you log test data, which I have done using a home build Picaxe based logger storing data on a 8Gb micro SD, there is only a few % of capacity in each peak hence there is little benefit (energy) to be gained by charging or discharging to the limits and a lot of risk.

For my system I have a dedicated monitor board (discrete components) which will disconnect at a low 2.5v and a high of 3.7v. I am in the process of building a 28X2 based monitor which will log the individual cell voltages and temperature again to a micro SD. One risk with multi cell LiFePo4 systems is that individual cells can get out of balance with the others and lead to thermal disaster. My monitor will provide data over a long period for me to digest and learn but also shut down the system if high temps or significant voltage differential appears. At the moment I will use a differential of 0.05v as after several cycles from fully charged to 70% DOD I have not yet seen a differential great than 0.01v, admittedly my tests have been relatively benign and not high current discharge. I did make sure the cell bank was properly balanced at the start.

At this stage I am very impressed with LFP cells as by comparison to traditional AGM DC batteries they provide a more stable voltage during discharge, accept near full charge currents right up to near 100% and have no memory effect so multiple partial top ups do no harm - typical in real world situations. One of their positives is also a potential danger - they happily provide huge discharge currents and so real care must be exercised.

Sorry to be a bit off topic but I could not resist. I will write up my project once it is completed.

cheers

 

[manuka]

Paix: Thanks- akin to what I've been hearing.

AlastaitD(Aus): Well received! Compared with AGM lead acid you could also credit LiFePO4 as being light weight, more compact, longer lived, gas free, acid corosion free, abuse tolerant, spill proof and of course free of lead toxicity issues. This'd make a great SiChip article!

 

[Paix]

@AlistairD, This won't be much use to you, but may be something that is worth reading as you have a heavy investment in your LFP battery stack.
LT Journal

Basically Google "Linear Technology battery management" and read yourself silly. They certainly seem to be the semiconductor house that is well on top of their game with a lot of useful help for engineers of all sorts and a lot of free software tools to 'help you understand it'. If you want to stay abreast of the subject or just need something to put you to sleep at night, then this is probably just the job. Certainly a company to keep an eye on for answers. I can't vouch for the costs, but I think that the word 'premium' or 'hens teeth might come into play'.

Sounds like quite some undertaking and look forward to your project completion and write up. I am assuming also that you may have probably read the LT literature on the subject and you definitely seem to be headed in the right direction with single cell balancing across the board.

 

[premelec]

@AlastairD - I agree with Paix it's well worth looking at ICs made to monitor and balance batteries - some are quite clever in re-routing overcharge to where it's needed and so forth... I saw a non- Maxim or LT unit announced recently but can't find the manufacturer's part just now [I posted earlier on this forum]. With all the electric vehicles with multi KWHr energy storage & cells these ICs are becoming more common.