PDA

View Full Version : LIPO Batteries



cycleguy
26 December 2010, 2202
It's getting very hard not to take a serious look at Lipos now that prices are close to that of LiFePo4's. I'm not talking about the large 10ah plus single cells but the RC packs from hobby supply shops. You can build a lipo pack that is 25% lighter and 25% smaller with much higher c ratings for the price of Headways.

The down side is that lipos are much more volatile when misused which makes it even more important to have a very good BMS system to insure proper charging and discharging. That's were the problem lies, all of these RC packs are wired in series, I have yet to see a 3.7V/100Ah pack, they are all 5ah or less with varying voltages depending on pack size, which makes it impossible to use a standard BMS unless you use one module per cell, not practical as there would be hundreds of them, or rewire the packs into a parallel string, also not a practical solution.

One neat thing these packs do have is a separate set of wires from each individual cell that's used for charging/balancing. The chargers have multiple outputs to charge and balance these cells individually at the same time. However, that wont do much good when trying to charge a large pack, as you would need one charger for each pack which can be as much as 80-100 for a motorcycle sized pack.

My thoughts would be to parallel the cells together into groups using the charge wires. The charger and BMS would be hooked up through the charge wires which would be configured exactly the same as a typical LiFePo4 pack. It wouldn't matter that each pack is created from series strings as long as they are paralleled on the BMS and charger side.

I'm wondering if anyone has tried doing this and if they can share their experience.

Tony Coiro
26 December 2010, 2232
I've been planning to do this as you described, I'll let you know how it goes.

larryrose11
27 December 2010, 0537
taking the packs apart and re-assembling them into a traditional parallel module arrangement would yield the best results. Some of the RC guys will parallel the available packs, but they are only using 4 or 5 cells, and are charging them individually.

Durability is another issue. Has anyone used them for over a few years?
If the RC guys get 1-2 seasons out of a pack, they are happy. I just got back into RC last summer (boats) and I have 3 LIPO batteries, well see how they do next summer.

billmi
27 December 2010, 0657
I definitely think re-building the packs to parallel the cells would be the way to go. The individual cell charging/monitoring wires are pretty thin, and I expect they'd need to be replaced to handle higher C rate discharging.

On durability, numbers I've seen so far are 500 cycles. For racing that's probably good to get through a season, for a commuter bike, compared to 1000 cycles for cylindrical cells or 3000 for prismatics, they've got a way to go on price to be a comparable value.

cycleguy
27 December 2010, 0819
I definitely think re-building the packs to parallel the cells would be the way to go. The individual cell charging/monitoring wires are pretty thin, and I expect they'd need to be replaced to handle higher C rate discharging.

On durability, numbers I've seen so far are 500 cycles. For racing that's probably good to get through a season, for a commuter bike, compared to 1000 cycles for cylindrical cells or 3000 for prismatics, they've got a way to go on price to be a comparable value.

The batteries don't discharge through the monitor wires, they just need to be able to handle charging currents, perhaps 10 amps max.

I agree, rewiring each pack into a parallel string would be best, but that adds a bunch of labor, which I'm trying to avoid.

I've heard that charging these lipo's no higher than 4.0V and never discharging them below 3.35V greatly increases their cycle life, approaching that of LiFePo4.

Tony Coiro
27 December 2010, 0826
On durability, numbers I've seen so far are 500 cycles. For racing that's probably good to get through a season, for a commuter bike, compared to 1000 cycles for cylindrical cells or 3000 for prismatics, they've got a way to go on price to be a comparable value.

I don't wanna be Mr. Debbie Downer here but I really don't think you can believe that cycle life, the numbers I've seen from companies who only test batteries are MUCH different the cycle life quoted by CALB and TS.

Barron
27 December 2010, 1129
If you parallel the cells only using the balance wires is the following scenario a consideration:

Two cells in parallel, one of them weak. On discharge, the weak cell drops to e.g.1V below the strong cell.
The cells will equalize because they are connected in parallel through the balance wires, but a potentially large current flows, even if briefly, while that equalization occurs.
Lets say there was 1 foot of 20 gauge wire between the paralleled cells with a resistance therefore of around 10 milliohms. The equalization current could therefore be initially high as 100 amps or 100W of power dissipation in the wire which could melt it...

Whadya think?

cycleguy
27 December 2010, 1237
If you parallel the cells only using the balance wires is the following scenario a consideration:

Two cells in parallel, one of them weak. On discharge, the weak cell drops to e.g.1V below the strong cell.
The cells will equalize because they are connected in parallel through the balance wires, but a potentially large current flows, even if briefly, while that equalization occurs.
Lets say there was 1 foot of 20 gauge wire between the paralleled cells with a resistance therefore of around 10 milliohms. The equalization current could therefore be initially high as 100 amps or 100W of power dissipation in the wire which could melt it...

Whadya think?

Yes, good point, I think this is a definite possibility. Although the same thing would happen inside a Thundersky battery, it wont be an issue because the cells are balancing through the much larger internal power bus.
This issue looks like a major stumbling block with this strategy.

billmi
28 December 2010, 0558
Tony,
I simply have taken it for granted that LiPO manufacturers probably stretch/best case scenario their life cycle estimates just like LiFePO manufacturers meaning the disparity in battery life is likely still there.

markcycle
28 December 2010, 0726
I've been using Lipo for a year now and have fallen in love with them. Nothing puts out amps like Lipo
I am using the cheap stuff from Hobby King 20C cont. 30C peak

My fist pack based on 6S 1P 5AH cells
built to 24S 3P
Not pretty but protected this is a cad drawing of the battery housing I built
http://dl.dropbox.com/u/3803661/battery%20frame.jpg

Second pack
25S 2P wrapped in the fireproof bag material then tapped up
http://dl.dropbox.com/u/3803661/IMG_0186.jpg

Second pack mounted on the bike as a booster pack
http://dl.dropbox.com/u/3803661/IMG_0198.jpg

Bottom line with Lipo is this

Pre test all cell before paralleling then test again
Never overcharge (at the cell level) and balance the cells, for longer life charge to 4.1 volts per cell
never undercharge

and enjoy the power

Nuts & Volts
28 December 2010, 0814
I have heard from endlessphere that the new nanotech cells have a longer cycle life, but I have seen no hard data. But one of the guys did pull 200C out of there 45-90C nanotech. Yes you read that correctly!
http://www.hobbycity.com/hobbyking/store/uh_viewItem.asp?idProduct=14616

Also I think my next motorcycle will definitely be LiPo. The thing I just learned is that you can parallel the balancing tabs so that you can balance the cells even if you dont parallel on the cell level first.

Kyle

cycleguy
28 December 2010, 0821
How are you charging this pack and are you using a BMS?

cycleguy
28 December 2010, 0845
The thing I just learned is that you can parallel the balancing tabs so that you can balance the cells even if you dont parallel on the cell level first.


This is exactly what we are talking about, but Barron brought up a potential scenario that may rule out the possibility of paralleling the cells through the balance wires.

What will prevent a momentarily low or failing cell from drawing high current through the thin gauged balancing wires?

Nuts & Volts
28 December 2010, 0904
This is exactly what we are talking about, but Barron brought up a potential scenario that may rule out the possibility of paralleling the cells through the balance wires.

What will prevent a momentarily low or failing cell from drawing high current through the thin gauged balancing wires?

My bad i didnt refresh myself on all posts. I would say two things could happen. Either the small lead melts/burns up or the cell recieving the current is destroyed. If that cell is momentarily low then that means it had a manufacturing error and isnt up to spec because all the cells were balanced on the charge. This means the cell should be removed from the pack regardless if you destroy or it destroys the wire. These high C LiPo should not fall out of balance unless you take an absurd amount of current out or you discharge below 15% or so. I dont have hands data, but I have some LiPo for an ebike that I will using in the near future.

My 2 cents
Kyle

cycleguy
28 December 2010, 0931
My bad i didnt refresh myself on all posts. I would say two things could happen. Either the small lead melts/burns up or the cell recieving the current is destroyed. If that cell is momentarily low then that means it had a manufacturing error and isnt up to spec because all the cells were balanced on the charge. This means the cell should be removed from the pack regardless if you destroy or it destroys the wire. These high C LiPo should not fall out of balance unless you take an absurd amount of current out or you discharge below 15% or so. I dont have hands data, but I have some LiPo for an ebike that I will using in the near future.

My 2 cents
Kyle

I agree, this would only happen in the event of a cell failure, and that cell would need to be replaced, however, the last thing you would want is a fire hazard occurring in the mean time.
Adding a 5 amp fuse to each balance wire will eliminate that hazard, and trigger a BMS LVC warning, but once again, this will add more complexity, cost and labor.
It would be much easier to bus the cells in parallel to begin with.

frodus
28 December 2010, 0949
It would be much easier to bus the cells in parallel to begin with.

I second that.

parallel first, then series. You save time, money and frustration.

If you parallel packs, then series the groups of packs, you'd have to attach the balance wires, which could actually transfer quite a bit of current if the cells are not matched. It won't neccessarily ONLY happen if a cell fails (it'l definately happen if a cell fails), but if cells are unmatched (like many hobby cells are), then you risk a higher IR in one cell and it'l shuttle current to make up for the discrepancy.

Remember, its a node. It could potentially see full current of that cell. If you do fuse it, and it blows due to overcurrent through the sense wire, you no longer balance that cell. LVC on the cell is no longer sensed by the BMS and you now have an unbalanced pack.

cycleguy
28 December 2010, 1034
I second that.

parallel first, then series. You save time, money and frustration.

If you parallel packs, then series the groups of packs, you'd have to attach the balance wires, which could actually transfer quite a bit of current if the cells are not matched. It won't neccessarily ONLY happen if a cell fails (it'l definately happen if a cell fails), but if cells are unmatched (like many hobby cells are), then you risk a higher IR in one cell and it'l shuttle current to make up for the discrepancy.

Remember, its a node. It could potentially see full current of that cell. If you do fuse it, and it blows due to overcurrent through the sense wire, you no longer balance that cell. LVC on the cell is no longer sensed by the BMS and you now have an unbalanced pack.

Merry Christmas Travis, I think we can close the debate on this subject. It's pretty clear that using these RC lipo packs as supplied is not a good solution.
I'm wondering if similar packs are available in parallel configurations, I've never seen them.
If they are available at the same cost as the series configurations, I would seriously consider using them, the weight, size and power density per cost ratio is too good to ignore.

Nuts & Volts
28 December 2010, 1100
agreed, much better to parallel first and then series

frodus
28 December 2010, 1104
Wonder if these hobby places could put in an order for a pack in different configurations....

the issue would be larger wires coming out.

Wonder if they sell individual cells :)

BaldBruce
28 December 2010, 1226
Check out places like all-battery (http://www.all-battery.com/37volt-5000mah20cli-polycell.aspx) They make custom packs or you can buy the individual cells. Probably more expensive than other Li, but you could definetly put together a killer small pack your self this way. The 5AH pack I linked to in a 19s2p would be 72V, 10Ah, 250A Peak, $1100, and only weigh 11 Lbs!!!!

I have had good luck from them with prior purchases of NiMH and NiCad cells. Previously bought an excellant flashlight from them also.

cycleguy
28 December 2010, 1234
Check out places like all-battery (http://www.all-battery.com/37volt-5000mah20cli-polycell.aspx) They make custom packs or you can buy the individual cells. Probably more expensive than other Li, but you could definetly put together a killer small pack your self this way. The 5AH pack I linked to in a 19s2p would be 72V, 10Ah, 250A Peak, $1100, and only weigh 11 Lbs!!!!

I have had good luck from them with prior purchases of NiMH and NiCad cells. Previously bought an excellant flashlight from them also.

I've looked at these before, way too expensive compared to the Zippy packs at Hobby King. A 6s1p 5ah Zippy pack goes for $45, compared to 6 of these at $159.00

markcycle
28 December 2010, 1335
I'm not one for debate
I'm a realest and the reality is we aren't going to get individual cells for cheap, or even a fair price. Now if anyone here has tried to take apart an RC Lipo pack then you know it's far more dangerous and risky taking apart a pack and risk puncturing the foil then it is paralleling the cells through the balance leads.

The key is to test the cells, test them twice test them under load, this is what I do. Then place them in parallel.
You'll mitigate the risk but not eliminate it, I'll be the first to say this.

Now for monitoring

I monitor every parallel string and if a string has a bad cell or a under performing cell I'll see the voltage sag on that string and at some point I'll get a low voltage alarm. More than likely a cell will under perform before it goes dead and internally shorted. Remember we are using these cells way under there capabilities, at least i am. RC guys will fully discharge these cell in ten minutes then turn around and charge them in ten minutes.
This is my crazy array of monitoring in reality I never have to look at it just listen for a alarm.
http://dl.dropbox.com/u/3803661/Cell%20monitoring.jpg

I'm even doing a interesting mixed chemistry setup that uses the best of LIFEPO4 and Lipo in a diode summed system. You'll see the CA meter and a second meter up top. The top meter monitors the Lipo pack current. Up to 1.5C or 60 amps, the LIFEPO4 batteries run the bike beyond that the LIPO kick in at 2C the LIFEPO4 stay at 1.5C and the Lipo do .5C. My 40AH LIFEPO4 cells now never see much over 60 amps while my 20C Lipo see as much as 200 amps peak.
Mixed chemistry battery sharing is really is a subject for another thread.
But I'll leave you with this.
http://dl.dropbox.com/u/3803661/voltage%20Booster.jpg

I'm going to add if you aren't sure of what you are doing don't do it. Even if you are sure, sleep on it, review it, do a risk analysis, because the risks are real but if you know what you are doing then so are the rewards.
Mark

BaldBruce
28 December 2010, 1354
Please start another thread Mark, I would love to hear more on your dual chemistry setup. I have considered this, but come up with no practical way to make this work other than isolation diodes. The addition of a second contactor looks interesting.
Bruce

chef
28 December 2010, 1519
Is cost the motivation for using small LiPo cells instead of larger ones? Harlan @ Hollywood Electrics has built a large-cell (relatively) LiPo pack for his GPR-S and was very happy with it. Jeff K's LiPo GPR-S is on its way to me and I'm looking forward to see how it performs.

I agree with Bruce, would like to read more about Mark's dual chemistry setup. Looks really cool.

lugnut
28 December 2010, 1543
I'm not one for debate...
I'm even doing a interesting mixed chemistry setup that uses the best of LIFEPO4 and Lipo in a diode summed system. You'll see the CA meter and a second meter up top. The top meter monitors the Lipo pack current. Up to 1.5C or 60 amps, the LIFEPO4 batteries run the bike beyond that the LIPO kick in at 2C the LIFEPO4 stay at 1.5C and the Lipo do .5C. My 40AH LIFEPO4 cells now never see much over 60 amps while my 20C Lipo see as much as 200 amps peak.
Mixed chemistry battery sharing is really is a subject for another thread.

Hi Mark,

I find this interesting. Please do start a thread, if you care to. And I don't want to start a debate. But can you tell me the motivation for keeping the LiFePo4 below 1.5C? I thought most were good to 3-4C and 6-10C peaks.

Also, does this require two separate chargers? And do you have a BMS on each of the 2 batteries?

Cool work :-)

frodus
28 December 2010, 1555
Cost is getting there, although just looking at online prices they're still slightly more than LiFePo4.

Lets say for the 30C LiPo from Turnigy:
http://hobbycity.com/hobbyking/store/uh_viewItem.asp?idProduct=14700

Its ~$70, 5Ah and 22.2V or 111Wh Which works out to be about ~$0.631/Wh

Headway from Manzanita Micro for instance:
http://www.manzanitamicro.com/index.php?page=shop.product_details&flypage=flypage.tpl&product_id=62&category_id=29&option=com_virtuemart&Itemid=64

is $19 a cell, 10Ah and 3.2V or 32Wh which works out to be about $0.594/Wh

But as quantity goes up (for either one) the price drops. So for 5kwh, LiPo would cost $3155. Headway (for instance) would cost $2970, but you can't do the high C-ratings.

Pro's for lipo:
More power
higher current

cons:
slightly more expensive
sensitive to temperature
volitile
balancing absolutely required
its a PITA to assemble a pack (lots of solder, lots of connections)

DaveAK
28 December 2010, 1642
I just priced LiPo at less than Headway, (using a Turnigy 8S1P 5.8Ah 25C pack), so depending on the pack configuration and specs you go for I don't think price (of the cells) is an issue any more when choosing between the two. Turnigy 4.7kWh pack for $2354 vs. Headway 4.6kWh pack for $2736. LiPo probably represents more work in pack building though, and so more money there. Also have to consider the quality as I hear that the Turnigy cells as they come require significant balancing. Not sure of their reliability in the long run though.

The rest of the cons are still there of course. I think I'm going to add some LiPo to my battery testing experiments and see what I can come up with.

Edit: And a LiPo pack is 75% the weight of an equivalent Headway pack by my estimates.

EVcycle
28 December 2010, 1644
"Headway (for instance) would cost $2970, but you can't do the high C-ratings."


Wanna Bet?



Haven't you tested these cells? 20C is not enough to move you down the road?

How much does 4S give you in amps?

frodus
28 December 2010, 1829
Relax ed, you forgot to read everything again *smacks ya upside the head*. 10Ah headways won't do 30C (like the lipo I chose in my example) without a HUGE voltage drop and HUGE temp rise. So if you mean can it do it, yes. Can it do it without problems? No.

And yes, we've tested these headways to 20C, and the temp rise after a continuous discharge is a bit more than I'd like, and the voltage drop is pretty high. They're not high-C rated batteries. They're more like medium.

so 30C lipo, running at 5Ah, 4 in parallel (I assume you meant parallel, not series) would be what, 600A

Just try to run headways at 30C without a voltage drop near 2.0V. The max I'd run them is 15C for short bursts. So for an equivalent Ah pack (20Ah), it'd be 2 in parallel, and it'd get you 300A at 15C, but the voltage drop would be considerable.


Anyone want to get a 10Ah 3.7V pack for me and I can throw on the tester and compare results with a 10Ah headway? I know the nominal voltage is different, but we can see which one fairs better.

cycleguy
28 December 2010, 2002
Is cost the motivation for using small LiPo cells instead of larger ones? Harlan @ Hollywood Electrics has built a large-cell (relatively) LiPo pack for his GPR-S and was very happy with it. Jeff K's LiPo GPR-S is on its way to me and I'm looking forward to see how it performs.

I agree with Bruce, would like to read more about Mark's dual chemistry setup. Looks really cool.

I recently got some quotes on those EIG batteries you are referring to. A 20ah cell was $73, if you purchased $100K worth, the price dropped to $43. Not even remotely affordable in my opinion.

EVcycle
29 December 2010, 0425
Remember Travis (Ed sticks his tongue out) I have the 8AH batteries.

I have been watching the post waiting for someone to actually do a apples to apples comparison, on a tester.

I looked at these type of batteries before too and also found them to be too expensive and not worth all the hassle compared
to the headways or larger TS type cells.


It is still a good subject to review as the technology advances.

teddillard
29 December 2010, 0505
I thought I came up with half the weight a while back, not sure how you get 75% less Dave. It was 75% the size, and half the weight... crap, now I gotta go look for that again. (edit: crap, I think it was 75% weight and half the size... I hate math.) But from where I sit, everything else given, that alone makes them without question the choice, and worth the effort. It all comes down to the usual saw- it depends on what you're building for I guess.

On the 8ah Headway 20C claims, Ed, I remember that and haven't been able to find the testing that backs that up since the ElMoto crash. Any idea where that was? I too would like to see some reliable and standardized testing...

EVcycle
29 December 2010, 0532
It was on their web page (20C) and from the amp reading on our battery only drag runs I have no doubt it was putting out 20C+.

As far a "laboratory" testing, I am not sure where they are at.

DaveAK
29 December 2010, 0929
I have no reason to doubt that Headways can do 20C, but is that peak or continuous? The LiPo's we're talking about are good for 20C+ continuous.

Ted, I'll post my calculations when I get to work, (that's where they're written down), and do a volume comparison as well.

frodus
29 December 2010, 0933
Remember Travis (Ed sticks his tongue out) I have the 8AH batteries.


Remember my example was using the 10Ah cells..... *sticks tongue out*

8Ah will do 20C..... but its not as cheap as LiPo

From before, the cells I used were 30C 3.7V 5Ah cells and cost ~$0.631/Wh

Then there's the 8Ah 20C+ Headway cell:
$19 a cell, 8Ah and 3.2V or 25.6Wh which works out to be about $0.742/Wh

So comparing cost of the 8Ah and the 5Ah cells, there's a difference of what, 11 cents per Wh? That is a pretty huge jump.

frodus
29 December 2010, 0936
I have no reason to doubt that Headways can do 20C, but is that peak or continuous? The LiPo's we're talking about are good for 20C+ continuous.

Ted, I'll post my calculations when I get to work, (that's where they're written down), and do a volume comparison as well.

The 8Ah cells are more expensive than lipo when you look at $/Wh.


If someone wants to send me a cell to test, I can test batteries up to ~150A. Temp and Voltage.

frodus
29 December 2010, 0945
On the 8ah Headway 20C claims, Ed, I remember that and haven't been able to find the testing that backs that up since the ElMoto crash. Any idea where that was? I too would like to see some reliable and standardized testing...

I have a 200A and 40A test done by someone independant of Headway for both 10Ah and 8Ah cells. At 200A on the 8Ah cell, voltage dropped to 2.3V. The test was 11 seconds before it started to drop too quickly and the test was stopped. The 10Ah cell droped to about 2.08V and was stopped after 11 seconds. For 40A, the 8Ah cell was at 5C and dropped to ~3V and went for 12.5 minutes before it hit 2V and the temperature rose from 37C to 57C. The 10Ah was at 4C and dropped to ~3V as well and went for 13 minutes before hitting 2V and the temperature rose from 38C to 64C.

These tests are over a year old though, so I don't know how accurate they are now with the newer cells with improved manufacturing processes.

cycleguy
29 December 2010, 0949
I think it's pretty safe to conclude that the current RC Lipos, especially the 15C ones are a much better value than anything currently available. The challenge is to figure out how we can use these safely for motorcycle applications, without adding too much complexity and additional cost.

EVcycle
29 December 2010, 0959
I will respectfully disagree until I see a similar test done on those cells.

I am not disagreeing due to my liking of Headway cells or think they are better. I would rather not have anyone spend a chuck of change
on a load of these batteries unless we have a independence lab do a similar test.


:)

calvin2
29 December 2010, 1015
I recently got some quotes on those EIG batteries you are referring to. A 20ah cell was $73, if you purchased $100K worth, the price dropped to $43. Not even remotely affordable in my opinion.
Where did you get the EIG quote from? I tried to get a quote from EIG the other day and they stated that they only sell to businesses and do not sell retail to DIY/hobbists. Then they pointed me to Mission Motors in SF.

larryrose11
29 December 2010, 1025
Lets say for the 30C LiPo from Turnigy:
http://hobbycity.com/hobbyking/store/uh_viewItem.asp?idProduct=14700

Its ~$70, 5Ah and 22.2V or 111Wh Which works out to be about ~$0.631/Wh

Headway from Manzanita Micro for instance:
http://www.manzanitamicro.com/index.php?page=shop.product_details&flypage=flypage.tpl&product_id=62&category_id=29&option=com_virtuemart&Itemid=64

is $19 a cell, 10Ah and 3.2V or 32Wh which works out to be about $0.594/Wh

But as quantity goes up (for either one) the price drops. So for 5kwh, LiPo would cost $3155. Headway (for instance) would cost $2970, but you can't do the high C-ratings.


On energy density:
using Frodus's example, the Turnegy cells come out to:
111Wh / 0.805 Kg = 137.8 Wh/Kg

For the Headway 10AH 38120 S Cell:
32Wh / 0.307 Kg = 104.2 Wh/Kg

32% more energy dense. Not a small difference. In this example 5kw-hr pack,
they would weigh:
5000
In a drag application, you could take advantage of the LI-Polly power density, and size the back to meet your power requirements. The pack in the link is rated @ 40C for 10 sec, which would need to b confirmed in testing. That's 200A for each of these modules. You could parallel them to get to 400A, have a 10Ah equivalent cell, and come out better on weight.

All of us are fighting the weight of the bikes. I think that the results speak for themselves.

DaveAK
29 December 2010, 1041
The 8Ah cells are more expensive than lipo when you look at $/Wh.
So are the 10Ah ones by my reckoning.

From the spot check I've done on prices and specs based on cells only with no regard to BMS or pack construction, LiPoly is cheaper, lighter and takes up less space than either style of Headway. (15% cheaper vs. 10Ah, 25% lighter and 22.5% smaller.)

frodus
29 December 2010, 1042
Where did you get the EIG quote from? I tried to get a quote from EIG the other day and they stated that they only sell to businesses and do not sell retail to DIY/hobbists. Then they pointed me to Mission Motors in SF.

cycleguy is Roehr motorcycles, so that probably has something to do with it.

frodus
29 December 2010, 1046
So are the 10Ah ones by my reckoning.

From the spot check I've done on prices and specs based on cells only with no regard to BMS or pack construction, LiPoly is cheaper, lighter and takes up less space than either style of Headway. (15% cheaper vs. 10Ah, 25% lighter and 22.5% smaller.)

with higher risk of fire/explosion. Lipo's are very volitile if not watched carefully (temp, voltage).

Long term tests on lifecycle have not been performed. They don't care if one pack only lasts a season or two. They're not designed with lifecycle in mind, only power and energy density. They also need different chargers and BMS.

And for these hobby cells, consistency in IR is not as important as it is with us in our bikes...... and they're still a PITA to assemble correctly.

DaveAK
29 December 2010, 1059
with higher risk of fire/explosion. Lipo's are very volitile if not watched carefully (temp, voltage).

Long term tests on lifecycle have not been performed. They don't care if one pack only lasts a season or two. They're not designed with lifecycle in mind, only power and energy density. They also need different chargers and BMS.

And for these hobby cells, consistency in IR is not as important as it is with us in our bikes...... and they're still a PITA to assemble correctly.
Never argued against any of these points, but the higher risk is what makes it all the more fun! :D

(The different charger/BMS requirements don't necessarily change the over all cost though.)

frodus
29 December 2010, 1104
They change the cost if you are upgrading and need to get a new charger and can't sell the old one for what you paid... :)

DaveAK
29 December 2010, 1111
They change the cost if you are upgrading and need to get a new charger and can't sell the old one for what you paid... :)
Same argument if you're changing from a DC to an AC motor. Let's try to keep this simple and say we're working from scratch. :)

Now here's a question about lifecycle. Let's say you've got a pack that will do 20C continuous, but it's a street bike and you normally run at a fraction of that, (2 or 3C maybe?) How would that effect lifecycle? Also add in the scenario that typically you only use half the useable capacity of the pack. What then? These RC guys are going flat out from full to empty every cycle aren't they?

BaldBruce
29 December 2010, 1127
682
One picture says it all about buying low cost LiPo.....
Seen some pretty spectacular fires at the local RC races. (Not me in the picture:cool:)

teddillard
29 December 2010, 1129
Funny. Every issue you hit on building a fast ICE motor hinges on life-cycle, too, doesn't it? Everything you do to an ICE motor to make it go faster always makes it blow up sooner. Why should batteries be any different? They have more energy and can release it faster, they're not going to last as long and are going to be touchy... and, like everything else, this just comes down to what we always hear- build it for what you want it to do.

BaldBruce
29 December 2010, 1132
Couldn't agree more Ted. Just need to educate people so they know the risk/reward they are strapping under their seat!

DaveAK
29 December 2010, 1149
Maybe I should just go and wrap myself up in bubble wrap then. Being careful not to suffocate myself of course. :)

frodus
29 December 2010, 1154
Now here's a question about lifecycle. Let's say you've got a pack that will do 20C continuous, but it's a street bike and you normally run at a fraction of that, (2 or 3C maybe?) How would that effect lifecycle? Also add in the scenario that typically you only use half the useable capacity of the pack. What then? These RC guys are going flat out from full to empty every cycle aren't they?

Discharging at a lower C-rate and to a lesser depth of discharge will greatly increase the life expectancy of a battery. Draw more amps and discharge longer, and you get less life.

DaveAK
29 December 2010, 1202
Discharging at a lower C-rate and to a lesser depth of discharge will greatly increase the life expectancy of a battery. Draw more amps and discharge longer, and you get less life.
It seemed like an obvious answer to me, but thought I should probably ask anyway. So if I replaced my pack with LiPo the max C I'd likely hit would be 7C or there abouts as it would be the limit of the controller. Considerably within the limits of the cells, thus extending lifecycle and reducing risk.

larryrose11
29 December 2010, 1255
Sorry about the incomplete post earlier, stopping short on the actual weight of each 5 kW-hr pack, but you get the point.

As far as the life issue, there is another component battery life, and that is calndar life. a Battery can go bad sitting on the shelf, even if it is maintained. The effect is the same, increased internal resistance as a function of calendar life. Perhaps Markcycle or some other frequenters of E.S can comment about calendar life.

As a battery ages with use, this also will manifest itself as increased internal resistance. If you are using a high c-rate battery, this resistance is all ready quite low. The net effect on voltage sag under load will be minimized, compared to a lower c-rated cell, all else being equal

As for safety, Charging is where most of the danger is. Limiting the charge to 80% full and avoiding high temperatures and low cells would go a long way.

A side note: If you use the existing cell balance charge leads and fuse each one, a blown fue is an indication of a weak cell

chef
29 December 2010, 1259
According to Harlan's experience with the EIG LiPo cells, they don't need a BMS for individual cell balancing, just pack HVC & LVC (charger & controller respectively). The data he's collected cells has shown the cells stay in a very tight voltage range:
http://www.elmoto.net/showthread.php?644-2010-Native-Electric-Motorcycle-For-Sale8-0V-Nominal-Lithium!&p=4516&viewfull=1#post4516

I wonder if LiPo inherently has more consistent IR or if the EIG cells are just that good. I'll take some measurements as time goes by to see how close the cells stay to each other.

EVcycle
29 December 2010, 1334
Funny. Every issue you hit on building a fast ICE motor hinges on life-cycle, too, doesn't it? Everything you do to an ICE motor to make it go faster always makes it blow up sooner. Why should batteries be any different? They have more energy and can release it faster, they're not going to last as long and are going to be touchy... and, like everything else, this just comes down to what we always hear- build it for what you want it to do.

Very good point, but that is pushing the Speed/life cycle/not kill yourself ( or wallet) is all about!!!


:)

teddillard
29 December 2010, 1338
Very good point, but that is pushing the Speed/life cycle/not kill yourself ( or wallet) is all about!!!


:)

yep. don't take long to get right back to the GO FAST = MONEY equation.

BaldBruce
29 December 2010, 1344
Risk to LiPo cells is many fold.
1) Over heat and they vent. Unbound Li plus moisture equals fire.
2) Over voltage and they vent. Unbound Li plus moisture equals fire. (remind me sometime to tell you the story of what happens to 128 cells when the charger they are on decides to go to 12V instead of 4.2V
3) Over discharge and they vent. Unbound Li plus moisture equals fire.

BUT, there is very good reason why almost every laptop made in the world and most cell phones take on that risk. They can't be beat for energy density. Just have to go that extra length with your BMS and charging/disharging systems to make sure you take care of them.

Mark pointed it out earlier and I'd like to stress again. If you are going to use packs assembled by a questionable source: Test and then test again to make sure a bad connection or bad cell doesn't bring your ride to a very hot ending......

teddillard
29 December 2010, 1348
Risk to LiPo cells is many fold.
1) Over heat and they vent. Unbound Li plus moisture equals fire.
2) Over voltage and they vent. Unbound Li plus moisture equals fire. (remind me sometime to tell you the story of what happens to 128 cells when the charger they are on decides to go to 12V instead of 4.2V
3) Over discharge and they vent. Unbound Li plus moisture equals fire.



Many? I see three. :D

...but point taken.

DaveAK
29 December 2010, 1352
Lithium and water in chemistry classes was always fun though. :D

DaveAK
29 December 2010, 1353
Many? I see three. :D

...but point taken.
I only see one.

BaldBruce
29 December 2010, 1357
Go play with your bubble wrap Dave.....

BaldBruce
29 December 2010, 1402
I'm a big fan of Li batteries! Just wanted everyone to remember why switching to LIFePO4 was such a big improvement in safety. (When I'm rich and famous, I'm gonna be driving a vehicle with over 6000 little bombs in it.!)

DaveAK
29 December 2010, 1420
I'm a big fan of Li batteries! Just wanted everyone to remember why switching to LIFePO4 was such a big improvement in safety. (When I'm rich and famous, I'm gonna be driving a vehicle with over 6000 little bombs in it.!)
I don't think anyone doubts that LiFePO4 is inherrently safer than LiPo, or better than lead acid which was the staple. It'll be a year or two before I can make the switch, and who knows what will be available by then? I'm happy that I stretched my budget to LiFePO4 and don't regret it at all.

Right now my plan is to use this bike as a test bed. I've become more interseted in exploration of the technology rather than having a daily commuter. This stuff is fun. :D

Warren
29 December 2010, 1506
Doctorbass' lipo bicycle...66mph max....698 miles...no fire so far.

http://www.diyelectriccar.com/forums/showthread.php/106km-h-electric-giant-dh-bike-52924.html

Warren

frodus
29 December 2010, 1525
Doctorbass is one of the most knowledgable people on Lithium batteries. He's built many packs, built his own BMS, tested many batteries and I suspect he's matched all of his cells.

He's one person I would completely trust to put together a great pack, and he's put a lot of time and effort into making sure he's got his bases covered.

billmi
29 December 2010, 1610
Doctorbass is one of the most knowledgable people on Lithium batteries. He's built many packs, built his own BMS, tested many batteries and I suspect he's matched all of his cells.


http://www.ri.cmu.edu/video_view.html?video_id=60&menu_id=387
In this lecture, Dr. Whitacre recommended cell matching as far more practical from a manufacturing cost and complexity standpoint than individual cell BMS. The way he described it makes a lot of sense, but to pull it off requires having large lots of cells to pick and choose from for exact matches. For the typical enthusiast builder that may not be so practical, unless you're someone like Doctorbass who's building lots of packs.


he's got his bases covered.
Or basses, as the case may be. :-)

teddillard
29 December 2010, 1648
dead link, Bill

Only2Jakes
29 December 2010, 1656
I say Lipo's are more for smaller applications because of the cheap pricing and the amp draw wouldnt be so high compared to a motorbike or car.
They can easily be used on motorbikes and cars but the thing is, are they as reliable as head way or any lifpo4 battery? Because this is what you would want in a commuter bike/car. If it's for offroad use or racing I say get stuck in with the nano-techs.

DaveAK
29 December 2010, 1717
http://www.ri.cmu.edu/video_view.htm...60&menu_id=387
In this lecture, Dr. Whitacre recommended cell matching as far more practical from a manufacturing cost and complexity standpoint than individual cell BMS. The way he described it makes a lot of sense, but to pull it off requires having large lots of cells to pick and choose from for exact matches. For the typical enthusiast builder that may not be so practical, unless you're someone like Doctorbass who's building lots of packs.
This all comes down to manufacturing tolerances and an ICE analogy might be blueprinting. (I'm not very good with analogies, so laugh all you want to. :))

Even assuming closely matched cells with a life expectancy of 500 cycles what's the tolerance? 500 plus or minus 10%, 5%, 0.1%? One of those cells is still going to start dying before the others. A BMS is probably unnecessary at the beginning of a well constructed pack's life, but it might be useful towards the end of its life. And if the pack isn't well constructed, (wider range of tolerances on the individual cells), then it becomes more useful earlier in the packs life cycle.

To me it seems a big unknown with a lot of good theories.

billmi
29 December 2010, 1813
This all comes down to manufacturing tolerances and an ICE analogy might be blueprinting. (I'm not very good with analogies, so laugh all you want to. :))

Pretty good analogy, I think - except instead of machining the parts to tighter tolerances, it would be buying dozens of engines, stripping them down and putting together the parts that matched the best. According to Dr. Whitacre, there are companies that do this already for LiPo used in space applications where a BMS is impractical due to weight limitations. He goes on to say later that performance wise individual cell monitoring and management is ideal. My interpretation of what he said is that for folks like us, buying a set of batteries, living with what we've got, and not worrying about paring down the cost for building 20,000 units, a BMS is a good idea, it's just not practical for the applications he was expecting his students to design for.

Tony Coiro
29 December 2010, 1815
For cell life, keep SOC between 30% and 90%, I believe Tesla operates within this same range. I know the Roadster has an extended range mode where it will charge up the last 10%. If you keep your SOC between those levels, that tolerance becomes less of an issue. I don't want to promise this, because I am currently unsure of the time frame but I will be getting a (probably Turnigy) LiPO pack tested by a company that just does battery testing before I start building my pack. I'll keep you guys posted.

markcycle
29 December 2010, 1847
If you put a lot of cells in parallel you get cell averaging. Which in practical terms is a form of cell balancing. Assuming a batch of cells has a random spread within a bounded range, and you weed out the bad ones. This I think is Tesla approach.

Mark

chef
30 December 2010, 0024
Doctorbass' lipo bicycle...66mph max....698 miles...no fire so far.

http://www.diyelectriccar.com/forums/showthread.php/106km-h-electric-giant-dh-bike-52924.html

Warren66 mph on a bicycle :eek:

That kid has some big hairy balls

DaveAK
31 December 2010, 1906
I've gone ahead and ordered an 8S1P 5.8Ah Turnigy pack to experiment with. Got some ideas rattling around in my head of how to do this, but it's got me thinking that to package them safely could really eat in to those weight/volume/cost savings. Haven't calculated out what my design would add yet but I'm reasonably confident that I can get a cost effective LiPo package that will stack up nicely against a LiFePO4 pack.