Seeking The "Most Cost Effective" Battery

[safe]

Seeking The "Most Cost Effective" Battery

We know that if you are planning to log a lot of miles that the more expensive batteries start to look very practical. However, if you aren't sure if you can make the higher mileage all that startup cost is a problem.

On the other hand, SLA has a problem in that the Peukert's Effect means that you have to configure things so that your bike doesn't use more than 1C or else you start to waste battery efficiency severely. (and hurt it's lifespan)

So here's the "question":

What might be the optimal cheap battery that exploited the advantages of SLA for bulk storage combined with some other chemistry like NiMh to assist in lowering the overall Peukert's Effect?

 

[safe]

My Current Theory

Using this logic what I'm thinking to do for my Project #002 is to use 80 NiCads or NiMh cells (to fill up the frame) to give about 300 watt hours worth of energy combined with four 12V 12Ah SLA cells that give another post-Peukert energy of 300 watt hours. The costs look like:

NiCad - $110
NiMh - $190 (costs more, but you get more Wh)
SLA - $100

...so you can see that a 600 watt hour pack could be put together for not that much money. (compared to something like the Ping LiFePO4 packs of 48V 15Ah that are selling for about $400 this is even less expensive, though heavier) Using NiCads and SLA you could get going for a little over $200.

The idea of using NiCad's and SLA seem's good because the NiCad's will last a long time while the SLA's will need occasional replacement. ($100) Such a setup could go for thousands and thousands of miles.

The technical questions in all this is:

"How do the discharge charactoristics of SLA and NiCad or NiMh compare?"

"Would there be some way to 'favor' the NiCads or NiMhs so that the SLA can be assured to not be overstressed?" (some way to keep the SLA's from dumping their load too rapidly)

 

[TylerDurden]

Miles are an inferior metric. Not very appropriate for batteries... tires, yes.

Discharge-Cycles are much better, that is why more knowledgeable people use them.

 

[OneEye]

Agreed, however many folks want to be able to create direct comparison to $ per mile they are familiar with through car living. Seems like some electric car companies are exploring battery rental by the mile or by the month to try to capitalize on that cost comparison.

 

[TylerDurden]

Agreed, however many folks want to be able to create direct comparison to $ per mile they are familiar with through car living. Seems like some electric car companies are exploring battery rental by the mile or by the month to try to capitalize on that cost comparison.

Ja, not to split hairs, but $/mi would need to be system dependent. Batts are a component or subsystem.

You wouldn't compare $/mi in cars by types of fuel and ignore the type of vehicle they're used in.

 

[safe]

The technical questions in all this is:

"How do the discharge charactoristics of SLA and NiCad or NiMh compare?"

"Would there be some way to 'favor' the NiCads or NiMhs so that the SLA can be assured to not be overstressed?" (some way to keep the SLA's from dumping their load too rapidly)

Metrics aside (I never really mentioned them) the concept involved mixing chemistries of SLA and either NiCad or NiMh (or I suppose LiFePO4 if you could get some cheap) in order to increase the yield of the SLA's.

The simple logic goes:

SLA's have a "C" rate weakness, have lot's of capacity and are cheap.

NiMh or NiCad's (in SubC form) come in high "C" rate offerings, but have low storage capacity and are cheap.

Blend the two together and you get the sort of "Yin and Yang" of batteries... for a solution that exceeds their qualities individually. (it's a little like the Supercapacitor plus SLA concept... getting the best of both worlds)

I have yet to hear of someone that has a lot of experience testing this sort of thing, so if I do it I'll be kind of exploring new possibilities...

 

[TylerDurden]

Metrics aside (I never really mentioned them) ...

It was your first premise, loonball:

We know that if you are planning to log a lot of miles that the more expensive batteries start to look very practical. However, if you aren't sure if you can make the higher mileage all that startup cost is a problem.

Second post:

The idea of using NiCad's and SLA seem's good because the NiCad's will last a long time while the SLA's will need occasional replacement. ($100) Such a setup could go for thousands and thousands of miles.

As for combining chemistries, hybrid batteries have been discussed here already.

 

[D-Man]

Let me know when you find out safe.

 

[safe]

Let me know when you find out safe.

All Battery is selling their NiCad SubC cells on eBay in matched sets of 96 at a time for $119.99 plus $13.95 shipping, so that comes to $1.40 per 2.4Ah cell.

http://cgi.ebay.com/96-NiCd-Sub-C-2400mAh-Batteries-for-PowerTools-Flat-Top_W0QQitemZ330198582341QQihZ014QQcategoryZ40975QQtcZphotoQQcmdZViewItemQQ_trksidZp1742.m153.l1262

That's really cheap...

In fact, at All Battery they normally sell them at from $2.69 to $2.20... so they are selling them at a real discount. I know people will say:

"Oh my god, those cells are going to be junk, they will not pull 10C for long before they go bad... etc... etc..."

...well, you know the logic, don't try to run 10C and you will probably be fine. Design for something more like 3C and you would be okay. NiCads last a long time too.

So in an unexpected irony the NiCad's might actually be in a very short period of time cheaper than Lead.

The SLA works out to: 35 cents per Wh.

The NiCads are: 48 cents per Wh.

...but if you think in terms of lifecycle (both calender and cycle) then the NiCads will pay off in the long run. (at that cross over point it might only be a few thousand miles rather than the 5,000 miles or so that the expensive stuff requires)

LiFePO4 tends to run upwards of a $1.00 a Wh.

These seem to be selling at a very fair price and even slightly aged NiCads are not an issue because they last a long time in storage. (unlike others)

So I might just go full NiCad... at this price the balance tips...

 

[safe]

Let Me Repeat...

This is the "Most Cost Effective" Battery thread... (I'm sure we've had many similiar ones going back in time) If people have examples and ideas that push the envelop of low price over time AND low price for startup then post here.

If it takes over 5,000 - 10,000 miles to begin to pay it back then that's not exactly "Most Cost Effective" because the startup cost is too high.

So it's all about low startup cost and low long term cost. Whatever seems to give the best blend of the two. Obviously there's some subjective judgement involved, but that's just how things are.

A123 cells might be the leader in performance, but they just aren't cheap... you pay a lot for the privilege of being "leading edge".

 

[deardancer3]

When calculating Most Cost Effective Battery, some factors also have to be analyzed:

First the technical sophistication of the user, as some of these battery systems can be complicated to charge, you may need two differnt charges, have to do some wiring stuff for every recharge cycle. So the cheapest system may not be for people that just want "plug n play". Add in the cost of a second charger, the time to do this stuff, and the risk of errors.

Second, While searching for most cost effective battery, one is dealing with postings, listings and advertisements touting lower prices and certain specifications. The market is riddled with low cost providers whose specifications are "Best Case" and not real world. Those prices and specifications usually can't be counted on for long. Again, the less sophisticatd user is the one that is "taken" with the advertisements, then gets "taken " with poor performance and even worse warrranties.

I am as frugal as the next guy and I like your search for best battery for the dollar. But we always caution folks to not solely buy on the basis of price unless you are an expert in the field; there are shady dealers in every market place looking to scam someone looking for a deal.

(I was an advisor to the jewelry trade on fake materials; Fake sellers have a very sophisticated marketing strategy with carefully word- smithed messages, while contending they dont know the Engish language that well. Some of this battery stuff smells the same as the guys sellling fake stone)

 

[Q-Bike]

Seeking The "Most Cost Effective" Battery

We know that if you are planning to log a lot of miles that the more expensive batteries start to look very practical. However, if you aren't sure if you can make the higher mileage all that startup cost is a problem.

On the other hand, SLA has a problem in that the Peukert's Effect means that you have to configure things so that your bike doesn't use more than 1C or else you start to waste battery efficiency severely. (and hurt it's lifespan)

So here's the "question":

What might be the optimal cheap battery that exploited the advantages of SLA for bulk storage combined with some other chemistry like NiMh to assist in lowering the overall Peukert's Effect?

Safe, a fundamental variable to do this kind of cost analysis is the needed range. If you ride to your work 10 km away all the year and can charge the battery in your office, a single 36V Dewalt (at about $150, with charger) can be economically interesting, even if your 'mileage' is not very high. But if you need more than 100Km range for a few weekends a year, that will be another history...

 

[TylerDurden]

So it's all about low startup cost and low long term cost. Whatever seems to give the best blend of the two. Obviously there's some subjective judgement involved, but that's just how things are.

Newer, better stuff usually costs more up-front, "but that's just how things are".

Get over it, already.

 

[CGameProgrammer]

Don't dismiss it too quickly, TD. Think of cars or motorcycles. If you browse EVAlbum, the overwhelming majority of vehicles use lead-acid batteries. Flooded batteries, usually. There's a reason for that. Even sealed lead-acid can be had for $0.25 per watt-hour and it comes in massive sizes which simplifies wiring. Also, there's no need to balance it. Thundersky LiFePO4 can be had for $0.75 per watt-hour and those also come in large amp-hour sizes (3.2V 90Ah). A123s at current ebay prices would be around $1.58 per watt-hour, and each cell is tiny.

This isn't a huge deal for e-bikes but it really matters when you're dealing with 72V 50Ah configurations and things like that. $900 vs $2700 vs $5688.

 

[EMF]

The thing Safe can't get through his thick head is there is no answer to this thread or 90% of the ones he starts. So he just keeps rephrasing the same old tired subjects over and over and over...

Ebikes are custom made to suit the rider and each rider has goals. No one solution will fit all. Another worthless thread

What he ought to do is design one for his bike, that meets his "cheapness goal" . Actually buy the stuff, then put it together then report about it in here- along with his collected data from his project. Thereby posting useful information that others can learn from, that might decide to create a similar ebike, rather than pollute the forums with these silly musings.

 

[safe]

A Visual Description of the Problem Space

A lot of times we talk about things and so many exceptions to the rule get thrown in and other tangents are argued over that we never get that high level overview of the problem. When it gets to that point you need a visual representation to break through the fog and bring clarity.

This chart should do that. (I hope)

The black line is the "baseline" which is the entry level price for an SLA pack after figuring it's Peukert's Effect losses of 50%. The other chemistries are then approximated based on typical costs and cycle life. (so this represents equalized total Watt Hour packs... same energy delivered)

From this chart you get a sort of visual description of exactly how long it takes various chemistries to achieve the "baseline" value. That's sort of your "battery freedom date"... the time that your more expensive battery becomes "free" relative to the baseline of SLA. The idea is to think in terms of "price per cycle" from the start rather than looking at it from the end which you might never get to.

You can see that LiFePO4 (at current prices) does not really break even until about 1500 cycles... but NiCads break even almost immediately... NiMh breaks even at about 300 cycles.

Another visual to keep in mind is the slope. Obviously if we want to compare the cost per cycle verses the baseline of SLA then the slope of the lines will all be the same as the depreciation of the asset occurs. All chemistries are then being depreciated at the same rate. NiCads depreciate to the baseline almost immediately and after that (all the way to 1000 cycles) becomes pure profit relative to SLA. The chart doesn't display negative numbers (which represent profit on your battery choice) but it could have been presented that way to show the full picture.

(okay, sounds like another chart needs making)

 

[safe]

So this represents the full life cycle of the chemistries.

The final score becomes:

SLA Baseline - $0 difference

NiMh - $33 more expensive than SLA potentially.

LiFePO4 - $68 less expensive after 2000 cycles

NiCad - The big cost effectiveness winner at $214.

Note: The NiCad price is the one on "special" so that might not be a typical price. Even so, NiCad's seem to be holding their value pretty well.

Also, this is "raw cell" pricing, not including all the other stuff that ends up being needed.

 

[fitek]

Safe, I only had one go at testing that all-battery SubC, and it stank. I believe it may have been charged incompletely. Unfortunately I'm not going to be home at all to retest until next week.

I read the all-battery forum and there were some complaints about that battery. Apparently the last half year they have sucked, and some people have been shipped duds.

 

[safe]

Safe, I only had one go at testing that all-battery SubC, and it stank.

I read the all-battery forum....

Now you do know I'm talking about the 2400mAh NiCad SubC's and NOT the 3500mAh NiMh SubC:

Yes, it's the NiCad:

http://ngvi.ebay.com/vi?ngvi&backtoid=0&ppsp=1&li=0&pdp=0&dsr=0&item=330198582341&backto=http%3A%2F%2Fwww.ebay.com&pass=3vD9p4M6Qa8O%2BaoWF%2FcpILQ6jMY%3D

No, NOT the NiMh:

http://ngvi.ebay.com/vi?ngvi&backtoid=0&ppsp=1&li=0&pdp=0&dsr=0&item=260204452094&backto=http%3A%2F%2Fwww.ebay.com&pass=TQF54ZH3oFGX9XLxhH6Z7Stlrsg%3D

...and I went to the All Battery Forum and searched for NiCad SubC and didn't find anything. So you need to show me where you read that. NiCad's are a more proven chemistry than NiMh and the expectations are low for that price point. (they only need to deliver 2400 mAh)

Plus, when I go to the Tenergy website they don't even list this product there which means it might be new or it might be some special order batch that got dropped for whatever reason. Hopefully there's some deal that went south on them and they now just want to get rid of them.

They are selling them at All Battery (their direct buy website) for a full dollar more!

So maybe they are old? Repackaged?

http://www.all-battery.com/index.asp?PageAction=VIEWPROD&ProdID=418

 

[Link]

IIRC, NONE of the Tenergy batts delivered their rated capacity. I did a fair bit of searching after I found their D cells. Their sub-c cells did okay when it came to discharge rates, but they didn't live up to near what they should have. I think the 3500mAh ones delivered something like 2200mAh actual.

Another thing to keep in mind is that Tenergy used to have shills on various battery forums and in their review sections, giving glowing reviews about how the batteries performed even better than expected, how great the price was for top-quality cells, etc.

 

[safe]

I think the 3500mAh ones delivered something like 2200mAh actual.

Again... it's the NiCad's (2400mAh) that I'm interested in.

The NiMh are a fundamentally more fragile product.

NiCad's have been around a long time and 2400mAh is not an exceptionally high bar to clear. At the price they are selling them if they delivered 1800mAh they would still be a good deal.

They are selling for only $1.40 each. (96 in a set)

Plus I wouldn't overstress them with high "C" rates either in charge or discharge prefering to stick to about half their ratings. (that tends to be gentle on them)

 

[TylerDurden]

Don't dismiss it too quickly, TD. Think of cars or motorcycles. If you browse EVAlbum, the overwhelming majority of vehicles use lead-acid batteries. Flooded batteries, usually. There's a reason for that.

Car-size, safe lithium chemistry is too new for most EVers to have.

I can tell you it is worth paying the money up front for better batteries. It's a better value, plain and simple.

Like welding gas; if you buy/rent the big tanks from the beginning, you save the most money.

 

[Link]

I think the 3500mAh ones delivered something like 2200mAh actual.

Again... it's the NiCad's (2400mAh) that I'm interested in.

The NiMh are a fundamentally more fragile product.

NiCad's have been around a long time and 2400mAh is not an exceptionally high bar to clear. At the price they are selling them if they delivered 1800mAh they would still be a good deal.

I was giving an example of the brand in general.

The NiMH cells are indeed bad.

Funnily enough, I think that's about what you actually get out of them.

 

[safe]

Like welding gas; if you buy/rent the big tanks from the beginning, you save the most money.

A good point.

There's no doubt that eventually some more expensive batteries to buy "up front" end up delivering more in the end.

It's the "3,000 Mile Rule" in practice... if you can surpass 3,000 miles then the options open up.

If you only ride 500 miles per year (like some kids around here with their electric scooters) then the low entry price of a couple SLA's is a lot easier on the wallet than some sophisticated LiFePO4 with BMS and fancy charger. Plus, for kids that lose interest in things it's not such a loss for the thing to eventually wind up thrown in the garbage.

You would be amazed at the junk I've been able to pick up from the curb. I've gotten three full bicycles that would have been taken away (as trash) for use as spare parts.

There's nothing more "cost effective" than free... DoctorBass seems to have figured that out...

 

[stew007]

Have any of you considered using Lipo's? for a 12s 44.4v 10ah battery, soldered up yourself costs $420, thats with copper tab cells from http://www.maxamps.com/Lipo-10000-Cell.htm very easy to make including soldering on a balance plug for safe charging. They are light weight and not at all bulky.

You can balance charge them safely in 30-45 mins easy, I use the Hyperion Duo charger that can balance charge 2x 22.2v packs at 8 amps. These cells should last a long time too since the demand from them are really low, they are rated at continuous 100 amps, 300 amps burst! also some good deals from here http://home.comcast.net/~truerc/custom/custompack.htm Ive used Lipos for a few years now, never had a problem

Im using the maxamps 10ah cells on my new bike build with a Crye 408 front 26" hub cant wait to see how it goes, should be a lightweight efficient combo.