PDA

View Full Version : I see a Lithium battery dead end looming -- have you figured out the workarounds?



NonPolluter
09 February 2011, 1913
Sparky's saga mirrors mine. I'm going to Sealed Lead Acid "pure lead" high discharge rate AGMS's because I can see a total dead end with Lithiums of all types: Unless you have a cell-level battery analyzer (like the ones for lead acid), you can't measure Capacity. And, if you can't measure pack/cell capacity, you won't know which of the battery, or batteries, in the series is causing the entire battery bank to underperform.

For example, right now, our Native Z6/Z6000 electric scooter shows 55-57 volts. But the bloated cells (damaged, read earlier thread) won't move the bike.

I had a similar experience with my Montague electric bike, with SLA's two years ago: Full voltage on both batteries, but one had nearly zero capacity, as measured by a 12V battery analyzer. Although our investment in battery analyzers has been $1500, we've recouped some of that cost by charging $10 to test batteries in our town. Also, we're still riding electric 'scoots whereas those who don't have expensive battery analyzers have junked their bikes. Hence, I see a dead end for Lithium bike experimenters, or "paying Beta testers" aka some EMS electric motorcycle owners.
http://www.gnbsystems.com/testers_rc300.php
http://cpc.farnell.com/1/1/54466-sla-battery-tester-gold-ibt-act-meters.html


Questions for people in this forum:

1) How many miles have you gotten out of your Lithium pack, and please specify brand and model.

2) Are you running with/without a cell-level battery bank equalizer? What are your conclusions?

3) I read that the read-world range of a Zero electric motorcycle is only 26 miles; whereas, the Vectrix got 65 miles. Why would you buy a $10,000 Zero bike that is, according to some reviewers, not suitable for highways?

4) If "battery doping" by the battery genius Deafscooter allowed him to get 45MPH from a 24volt bike, why can't some small company buy his knowledge and sell products/kits/chemicals to every electric moto owner?

5) Does anybody know where I can get a 48V "bad boy" charger?

Thanks for all replies.

BaldBruce
09 February 2011, 1922
Pretty simple rules for Lithium. Design your pack to be larger than you normally use in AH so you don't push them and keep the C rate reasonable. Monitor the per cell voltage so your system shuts down before you hurt them.

DaveAK
09 February 2011, 2136
I'm not sure where you're getting your information, but I think you might be jumping to some wrong conclusions.

1) Zero. My bikes not on the road yet, but I'm hoping for 50ish miles, and expect that to be a reasonable estimation, give or take. CALB 60AH LiFePO4.

2) Not sure what you mean by battery bank equalizer, but I'll be using a MiniBMS for both cell level LVC and HVC. Not expecting any problems with it.

3) I've heard nothing about the Zero range, but they make more than one model and have new versions out for this model year. As for the Vectrix I heard it was more like 30 miles, and had more than its fair share of problems mostly battery related.

4) I bet I could get 45mph from a 24V bike without being any kind of genius, deaf, dumb or blind.

5) eBay?

A good BMS will solve your problems. A bad one will make things worse for you. You'll get plenty of opinion on what's good and what's bad, and that's where you're going to have to educate yourself. If you can't navigate that minefield then certainly SLAs are a safer option.

chef
09 February 2011, 2241
SLAs are inferior to LFPs in most aspects.

Leads are heavier for the same amount of stored energy (lower energy density).
Leads can only be discharged to ~50% before they suffer plate damage, so not only is the battery heavier to begin with, you can only use half its capacity. Deep discharge leads can go lower since they have thicker plates, but they will suffer damage with repeated deep discharges. The downside with deep discharge leads is they cannot deliver the high current of regular leads.
Leads have a cycle life of ~500 normal discharges (50% DoD). LFPs: ~2000 cycles at 80% DoD.

The only significant advantage leads offer is higher discharge capability. Combine low capacity with high discharge and the pack won't get you far.

There are a number of successful LFP packs in operation with many, many cycles on them. EMS doesn't have any despite their claims. They just don't know how to properly design packs for longevity. They push the cells too hard in every regard and it's no surprise your pack developed bad cells.
My GPR-S originally came with a lead pack. I got less than 100 cycles out of it before it dropped to less than 50% of original capacity. EMS configured the bike to draw way too much current with undersized cells. That should have been a warning to stay away from any of their battery packs. The LFP pack I got from them lasted a couple hundred cycles (hey that's twice as long as the lead pack :mad:). Hindsight 20/20...



A good BMS will solve your problems. A bad one will make things worse for you. You'll get plenty of opinion on what's good and what's bad, and that's where you're going to have to educate yourself. If you can't navigate that minefield then certainly SLAs are a safer option.Well said. Unlike the motto "even bad sex is good sex", a bad BMS is much worse than no BMS. The Mini BMS is designed well and does a good job. There are ways to get by without a BMS on LFPs if you're intimately familiar with the specific manufacturer's chemistry. That entails sacrificing capacity and operating in the 20-90% range and periodic manual checkups. It's tough to sacrifice capacity on space-constrained 2-wheelers, but if I had to do it over again I'd gladly make that compromise and still have healthy cells.

Richard230
10 February 2011, 0857
The only information that I can add is that my first GPR-S, with its 20-cell 50Ah Hi Power battery pack seemed to perform just fine for the first 300 miles - when the BMW melted down. The battery pack was able to get me the 35 miles home from my dealer with power to spare and voltage drop in the pack when I arrived home was still above the nominal 60 volts (it was 68 volts fully charged) for a 20-cell pack.

My next sepex GPR-S, with its 24-cell 50 Ah Hi Power battery pack also made that initial trip with no problems, using 33 Ah during the 30-mile ride. I later ran it down once to 40 Ah before the pack sagged just to get a feel for the maximum range. I then typically would run the pack for 25 Ah (maximum) before returning home and recharging. The pack lasted for 1300 miles and maybe was recharged as many as 80 times, before it started to act strange. Up until the time the only symptom was noticeably less performance than my original RT-motored GPR-S, even when new. However, I still suspect that some of my batteries were damaged before I even took possession of my bike. It had about 200 miles on the clock when I bought it (even though the manufactured date of January 2010 on the headstock is a week before I the purchase date of January 10, 2010) and the voltage of the fully-charged pack was only 78 volts (I now know it should have been 81 volts). Later, when I inspected the batteries, I noticed that they were all covered with dust and were discolored as if they had been sitting around for some time before being placed in service.

So my thoughts are that perhaps these batteries have lived a hard life traveling from the factory in China to the US distributor, to EMS and finally being installed into the vehicle. They may have been improperly stored and may also have been placed in service in some other application before being installed in my bike. I also note that the batteries were all built at the factory during the summer of 2008 and that the Hi Power importer says that the batteries are no longer available in the US any more. So my bottom line is that the problem may not be the LiFePo4 battery technology, but improper shipment, storage, initial charging and incorrect initial use of the batteries in the vehicles using these batteries.

HighlanderMWC
10 February 2011, 1218
6200 miles on a Brammo Enertia. So far no signs of any reduced capacity.

NonPolluter
11 February 2011, 1112
4) I bet I could get 45mph from a 24V bike without being any kind of genius, deaf, dumb or blind.





If you spend a few hours watching all of Deafscooter's achievements on many videos, you'll see that he is the "Dr. Werner Von Braun" (the rocket scientist) of lead acid battery doping.

NonPolluter
11 February 2011, 1118
LIFEPO4, like the Electric Motor Sports GPR-S and Native Z6000 scooter, look good on paper. Real world experience is a nightmare, compared to just buying even a cheap 50cc Chinese scooter, like the one I had two years ago.

Here is the real-world experience with Thundersky batteries, and the extremely time consuming data-logging, repairs, drilling bloated cells, hacksawing into battery trays, etc. necessary simply because consumers can't buy a cell-level battery analyzer:

guity's gpr-s experience (http://electricmotorcycleforum.com/boards/index.php?topic=512.0) 1 (http://electricmotorcycleforum.com/boards/index.php?topic=512.0) 2 (http://electricmotorcycleforum.com/boards/index.php?topic=512.15) ... 13 (http://electricmotorcycleforum.com/boards/index.php?topic=512.180) 14 (http://electricmotorcycleforum.com/boards/index.php?topic=512.195)

DaveAK
11 February 2011, 1124
You seem fixated on a cell-level battery analyzer. What exactly do you see as its function, and what makes it necessary? You must be talking about something other than a BMS that will give you LVC and HVC protection and/or a datalogging cell-log that will track voltages, all on a per cell basis.

NonPolluter
11 February 2011, 1318
You seem fixated on a cell-level battery analyzer. What exactly do you see as its function, and what makes it necessary? You must be talking about something other than a BMS that will give you LVC and HVC protection and/or a datalogging cell-log that will track voltages, all on a per cell basis.


A Battery Analyzer is a diagnostic tool. You connect them to the terminals, and the readout tells you the Ah capacity, regardless of the State of Charge. In other words, you get the answer you need in minutes, without having to even turn the electric bike on.

The drawback with most of them is that they need to rest, after doing three or four, because the tester gets a little hot, and inaccurate.

Here are inks video clips to two models that I do not own:
http://www.dailymotion.com/video/x6dei7_the-soctester-the-most-accurate-bat_news
http://www.youtube.com/watch?v=ezahuellCn0

Note that these video clip are informercial by the sellers.

chef
11 February 2011, 1356
LIFEPO4, like the Electric Motor Sports GPR-S and Native Z6000 scooter, look good on paper. Real world experience is a nightmare, compared to just buying even a cheap 50cc Chinese scooter, like the one I had two years ago.

Here is the real-world experience with Thundersky batteries, and the extremely time consuming data-logging, repairs, drilling bloated cells, hacksawing into battery trays, etc. necessary simply because consumers can't buy a cell-level battery analyzer:
You've taken one manufacturer's faulty implementation of LFP and generalized it to all LFPs. This in spite of the information presented to you here. No offense but if you've already made up your mind and won't accept any information contrary, why bother asking? It's turning into a Rickard fest.

If you're not going to be convinced by anyone else's experience, build both lead and LFP bikes (with properly configured packs of course). Run on leads for 1000 cycles... oh wait you'll get 500 or fewer full cycles. Run on LFP for 1000 cycles and you'll still have another 1000 to go.

BaldBruce
11 February 2011, 1400
The battery analyzer you are talking about is for a 12V lead acid battery ONLY. They rely on the linear relationship between voltage and state of charge in lead acid batteries ONLY. A load is applied and the resultant loaded voltage is measured over the period of the test to determine capacity/health. This method has no bearing whatsover on lithium batteries. Their load curve is almost flat and it is the knees of the curve we are looking for. We want to operate on the flat part of the curve and stay away from where the voltage rises like crazy or falls of the cliff. Damage to your cells is right around the corner if you allow them to see either area outside the flat part of the curve. Thus most people who are SUCCESSFULLY operating lithium systems are using a BMS to at least make sure the cells are always in their "safe " operating range. Whether or not a BMS should do cell balancing, logging, reporting, temperature sensing, etc is a whole nother discussion. (A whole raging debate on another forum if you really want to get bored quickly.)
(Also a whole seperate argument is also flying around concerning the definiition of BMS in terms of which function this term implies even. Clogging my inbox lately it's so bad......)

DaveAK
11 February 2011, 1429
Pretty much what Bruce said. Start with LVC and HVC and go from there depending on your needs and desires. I want the full gamut of what can be done, and will have it all when I'm finished. My MiniBMS will give me LVC and HVC to start with, and that's all I really need. I'll do the rest just 'cause I can. :D

larryrose11
12 February 2011, 0924
The battery analyzer you are talking about is for a 12V lead acid battery ONLY. They rely on the linear relationship between voltage and state of charge in lead acid batteries ONLY. A load is applied and the resultant loaded voltage is measured over the period of the test to determine capacity/health.

I totally agree with Bruce. Further, different SLA have different pukert effect's, so the load applied from the analyzer has a direct effect in its measurements. If the analyze ran different loads, the it will also have different results.


There are ways to get by without a BMS on LFPs if you're intimately familiar with the specific manufacturer's chemistry. That entails sacrificing capacity and operating in the 20-90% range and periodic manual checkups. It's tough to sacrifice capacity on space-constrained 2-wheelers, but if I had to do it over again I'd gladly make that compromise and still have healthy cells.

I have been saying this for years. I have gotten some good responses from some BMS manufactures that will let me program the HVC, LVC myself for each cell to stay in the healthy range



This method has no bearing whatsover on lithium batteries. Their load curve is almost flat and it is the knees of the curve we are looking for. We want to operate on the flat part of the curve and stay away from where the voltage rises like crazy or falls of the cliff. Damage to your cells is right around the corner if you allow them to see either area outside the flat part of the curve. Thus most people who are SUCCESSFULLY operating lithium systems are using a BMS to at least make sure the cells are always in their "safe " operating range.

Exactly. ALL the major OEMs that will warranty their packs, be it car or bike, will do this. The VOLT, has a 16 kW-hr pack, but limits the avalable to 8-10 kW-hrs for durability. The HEV packs in the Ford's limit the (small, 6Ah or so) pack usage to the 40-60% range, and last over 300K miles.

Anyone spotting the trend here? This horse AINT DEAD YET?!?
If you want good durability out of your expensive Li based batteries, stay in the 80-20% or so range of battery operation.

Minimally, have some sort of warning that lets you knw when you hit the LVC / HVC for the healthy range.
Better still, find a BMS that will do this for you, one that will communicate with your motor controller that will throttle back the current (either direction) when it hits these voltage ranges under operation, and will talk to the charger as well and keep this cells in this region

Skeezmour
12 February 2011, 1446
Geez I wish I know of a BMS that did all this....oh wait :)

Have some updates on the site about the new items in the line up. More to come very very soon.

Gene

Only2Jakes
12 February 2011, 1625
Nice Gene, I'm definately getting that and your BMS if I ever do an electric car.

I recently oredered some of the 2 headway cell mounts/brackets and bus bars, waiting on an email invoice :)Jake

EVcycle
12 February 2011, 1636
Questions for people in this forum:

1) How many miles have you gotten out of your Lithium pack, and please specify brand and model.

Over a 1000.

2) Are you running with/without a cell-level battery bank equalizer? What are your conclusions?

Individual changers to each paralleled bank. (I.E 26 chargers to 26 packs of 4 Lifepo4 batteries) plus Celllog 8 devices.

Works great!

SplinterOz
12 February 2011, 1650
I am building a headway pack for my bike and I am going to use Celllog 8 devices (http://www.hobbycity.com/hobbycity/store/uh_viewItem.asp?idProduct=10952) to keep track of my cells.
With these I am going to create a circuit from the alarm output to ensure that I avoid HVC and LVC problems.
Working from this thread on Endless-Sphere (http://endless-sphere.com/forums/viewtopic.php?f=14&t=19955) I should be able to kill my bulk charger if any cell is HV. I plan to run a balancing charger across the pack every weekend to ensure the cells are balanced. When the bike is operating the alarm circuit will either sound an alarm or drag the throttle down to reduce my total power drain.

EVcycle
12 February 2011, 1702
I am building a headway pack for my bike and I am going to use Celllog 8 devices (http://www.hobbycity.com/hobbycity/store/uh_viewItem.asp?idProduct=10952) to keep track of my cells.
With these I am going to create a circuit from the alarm output to ensure that I avoid HVC and LVC problems.
Working from this thread on Endless-Sphere (http://endless-sphere.com/forums/viewtopic.php?f=14&t=19955) I should be able to kill my bulk charger if any cell is HV. I plan to run a balancing charger across the pack every weekend to ensure the cells are balanced. When the bike is operating the alarm circuit will either sound an alarm or drag the throttle down to reduce my total power drain.

Sounds like a good plan. :)

BaldBruce
12 February 2011, 1817
I am building a headway pack for my bike and I am going to use Celllog 8 devices (http://www.hobbycity.com/hobbycity/store/uh_viewItem.asp?idProduct=10952) to keep track of my cells.
With these I am going to create a circuit from the alarm output to ensure that I avoid HVC and LVC problems.
Working from this thread on Endless-Sphere (http://endless-sphere.com/forums/viewtopic.php?f=14&t=19955) I should be able to kill my bulk charger if any cell is HV. I plan to run a balancing charger across the pack every weekend to ensure the cells are balanced. When the bike is operating the alarm circuit will either sound an alarm or drag the throttle down to reduce my total power drain.

I know you are aware of it, but want to make it clear for others who are interested in those great little cell loogers. If you want to use the alarm output of more than one, you MUST isolate the outputs. Smoke will ensue if you try to tie more than one of these outputs together. And also remeber they have a cool USB port to dump data from, but the same caveat applies. Do not tie more than one port together at the same time or you let the smoke out.

P.S. I built such an interface PCB for my own bike if someone wants a DIY project or advice on how to do it.....

EVcycle
12 February 2011, 1825
P.S. I built such an interface PCB for my own bike if someone wants a DIY project or advice on how to do it.....


Yes please!!!!

chef
12 February 2011, 1957
Me 2! I only have 1 CellLog but plan to add more later.

BaldBruce
12 February 2011, 2039
841842
Here is a picture of the front and back side of the "KISS my BMS" PCB. Works with up to 3 cell loggers. LED indictes when the alarm has tripped. Relay is for interfacing to charger and controller. Runs on 12V. Latchs relay until power is reset so charger can't cycle.

I'll post the schematic and layout on Monday. My files for that project are on a laptop at work. It could easily be enlarged to work with higher number of cell loggers or a larger relay for bigger chargers. (Mine is a 72V 4 amp charger system, so this size was great for me.)

EVcycle
12 February 2011, 2243
Very cool! I just want it for a LVC alarm.

Thanks!

SplinterOz
13 February 2011, 0136
Bruce, Yep I was aware but thanks for the reminder. The ones I ordered are the really cheap ones not usb data logging. I would like a copy of your arrangement as I would have had to create my own :)

I have the 8S model so I might have to modify your layout for the 9 wires. Do you have an overall schematic of your setup it would be really handy.

Thanks

larryrose11
13 February 2011, 0746
Geez I wish I know of a BMS that did all this....oh wait :)
Have some updates on the site about the new items in the line up. More to come very very soon.

Gene

Gene, Im in the dark here.
What web site? If your a business owner, perhaps put a link in your signature?

teddillard
13 February 2011, 0831
Gene's with Manzanita: http://www.manzanitamicro.com/

...gonna have to add him to the ranks of coy bitchdom with Mark and Tony. :D

Skeezmour
13 February 2011, 1130
Hey I'm low key thats all :)

mpipes
13 February 2011, 1311
sweet, now that I know Gene is with Manzanita, I can annoy him with questions before I place an order for 150 headway cells..... :D

Richard230
13 February 2011, 1620
Speaking of LiFePo4 prismatic cells, has anyone noticed that Electric Motorsport has stopped selling Thundersky batteries - which they were promoting as the latest and greatest thing to power electric vehicles not so long ago? Now they are offering GBS batteries, with LiFeMnPo4 chemistry - and they come with their own BMS system and matching charger. Not cheap though. Prices range from $2,290 for a 72V, 40Ah pack to $4,522 for the 100Ah pack. A 60Ah pack is somewhere in between.

So does that mean the TS batteries have followed in the footsteps of Hi Power batteries? Soon to become collector's items?

BaldBruce
13 February 2011, 1821
Nah, TS is still in bussiness. They recently changed their name to Winston Battery Limited after their chaiman and CEO Winston Chung. They did sign a distributor agreement with a US company, but this does not give that company exclusive rights to sell from what people in the industry are telling me. I still maintain that the large format prismatics are in their infancy and will undergo lots of growing pains as they move into commodity status. Still, Winston and CALB are the major chinese players in the game and probably will be for some time.....(Old names of ThunderSky[yellow] and SkyEnergy[Blue]).

BaldBruce
13 February 2011, 1851
KISS my BMS wiring diagram for my bike. 12V line to BMS missing. Internal DPDT relay shown swithching for both LVC and HVC functions.
852

cycleguy
13 February 2011, 1958
Nanocapacitors! 3000 wh/Kg, now we're talking. No BMS needed, almost unlimited cycle life, high rate charge and discharge, what more can you ask for other than, can we buy them within a year for the cost of Lead Acid?
This should be priority #1. Every effort should be made to bring these to market ASAP.

http://www.technologyreview.com/energy/22297/

SplinterOz
14 February 2011, 0145
KISS my BMS wiring diagram for my bike. 12V line to BMS missing. Internal DPDT relay shown swithching for both LVC and HVC functions.
852


Thanks very much... will compare to my plan. (and update mine as yours is implemented)

Richard230
14 February 2011, 0840
Returning to the subject matter (whatever that was - I have short-term memory problems), I see in the "New Products" section (page 37) of the February issue of the great British motorcycle publication Bike magazine, that LiFePo4 batteries have reared their ugly heads in the IC motorcycle battery market.

Under the heading "Racing Batteries LiFePO4 battery", these batteries are described as the "latest cutting-edge battery technology". They are said to weigh up to 5kg less and are much smaller than typical SLA motorcycle batteries. The article goes on to say: "LiFePO4 can provide more current from less material than lead/acid, which means that the cold cranking amps - what you need to get your bike started - are elqual to or more than the original fitment battery. But you will damage the battery if you allow it to fully discharge and the drain is faster if, for example you leave your lights on when the engine isn't running."

I guess that is why these batteries are for racing only. If you install them in a normal motorcycle, they will blow up like a blimp, stink and sizzle - to say nothing of damaging your BMS. :O

jazclrint
14 February 2011, 1643
Nanocapacitors! 3000 wh/Kg, now we're talking. No BMS needed, almost unlimited cycle life, high rate charge and discharge, what more can you ask for other than, can we buy them within a year for the cost of Lead Acid?
This should be priority #1. Every effort should be made to bring these to market ASAP.

http://www.technologyreview.com/energy/22297/

Isn't that what chip Yates said in one of his interviews with MotoPodcast.com, that he didn't want to hear about. Sounds like he might have some experience in that area.

Richard230
16 February 2011, 1640
It seems like LiFePO4 batteries continue to make inroads into the IC vehicle market. I wonder if they will replace SLA and flooded batteries in that application? Here is another example of this trend:

http://www.motorcycle.com/news/price-drop-on-ballistic-evo-2-batteries-90427.html