Custom A123's almost here!

[CGameProgrammer]

It wasn't GM; Chevron bought the patent for using NiMH in pure-electric cars, specifically so they can prevent anyone from doing so.

 

[usatracy]

GM used faulty Delco and much better Panasonic lead-acid batteries weighing 1300 lbs. to give the 1997 EV1 a range of up to 110 miles on a charge. GM also used a 700-lb. NiMH battery pack on the 1999 EV1 to give it a range of up to 160 miles.

The NiMH battery is recyclable, lasts longer than the life of the vehicle, is recyclable, has adequate power, is cheaper than Lithium, is safe, non-toxic and is capable of deep cycling on a daily basis.

The real reporter would inquire as to why GM is failing to produce the VOLT with existing NiMH batteries; it's not as if GM is not aware of them. GM even, at one time, owned the worldwide patent rights to the NiMH batteries, which it sold to Texaco (now part of Chevron).

----------------------
Perhaps I was not clear, GM owned the patent, sold it to petrol concerns which now licenses the technology with the right for use in mobile devices reserved and restricted, ie cant be used to power the drive system in land based vehicles, can be used in aviation, ocean and space.

 

[mcharles13]

Ok you guys have completely lost me. So i'm going back to the original question.

Hi,

I ride an XM-2000 EV Motor Scooter.

It comes with 38Ah / 60V (5 x 12V 38Ah (20hr rate)). So, by my reckoning I'd need 20s5p (possibly 19s5p) or 100 cells. This would give me 11.5Ah. Is the 2.3Ah per cell a 20hr rate or a 2hr rate? I'm hoping it's a 2hr rate! I'm hoping that 11.5Ah of lithium will give me a reliable 12 mile range (might be close!)
...

I don't understand what you mean by 20hr rate vs 2hr rate. That is not a term we use in the RC arena. We run these cells flat all the time. On many occasions i've pulled the full 2.3Ah out of the cells and put 2.3Ah back in. I know of no other battery that can do that w/o dieing or losing SERIOUS life cycles. The a123's take it, keep going and it doesn't affect their longevity. Keep in mind even if you run them flat as soon as the load is gone they come up to 2v/cell. 2V is a safe cutoff if you feel you must have a cutoff. I use a timer for air based vehicles and a simple watt meter for land based. i just periodically glance at the watt meter and when it hit's 2.3Ah (or whatever multiple for parallel) it stops.

If you have an 11.5Ah battery (5p) then regardless of how long you run it (20hrs or 2hrs) your done when you use 11.5Ah. Voltage after the initial dip (depending on load) will hold solid till that last drop then boom you have nothing. Again keep an eye on the watt meter's elapsed Ah usage display so you don't get stranded.

Now Regarding voltage dip, the more parallel you have the lesser the dip. For example in my e-maxx 4x4 monster truck i run a 5s1p setup and pull 900W. When i double that to a 5s2p setup I pull 1300W simply because the voltage dips less. With each increase in Ah you will have less and less v drop. On an 11.5Ah pack the dip will be nearly negligible.

Since you already have the bike. Measure your Ah usage on your next trip. Based on that number it will give you an idea of what size pack to make. Also watch the voltage drop as you move along. If it dips hard and continues to dip along the way keep in mind that the a123 equivalent will not dip over time. Use that extra power to your advantage. In RC many of us have learned that is is always best to get your desired wattage from voltage not amps. It's a much more efficient system that way. Gear up or go with a lower KV motor. Your motor will work less and you'll get more out of it.

,mike

 

[mcharles13]

On another note. FWIW I have tried my butt off to get the bigger 32series (that's what it's called not 32cells in series) cells that were designed for cars just so i can use them for these e-bikes but A123 won't give out ANY information regarding them other than what's on their website. It's really annoying because i am already an established vendor with them. They must be under an NDA at the moment OR are reserving that for only car usage. Who knows....

a123 really needs to come up with a bigger cell but not unattainable big.

,mike

 

[usatracy]

Rate, a battery that can provide 38 amps over 20 hours by delivering 1.9 amps per hour for 20 hours can't necessarily deliver all 38 amps in only one hour.

If his bike draws 50 amps, and he wants to ride it at 50 amps for one hour, a battery that is rated for 50 amp hours may not work if it can only deliver 30 amps for one hour and all 50 amps in 1 hour and 45 minutes

He will have to either slow down or pick a bigger or better battery

Thats what the 20 hour question is about.

 

[CGameProgrammer]

The NiMH battery is recyclable, lasts longer than the life of the vehicle, is recyclable, has adequate power, is cheaper than Lithium, is safe, non-toxic and is capable of deep cycling on a daily basis.

NiMH does not last long at all if you deep-discharge it. That's why hybrids always keep state of charge between 40% and 60%; they never stray much from half-full. That is what allows the packs to last for at least 200,000 miles. An EV cannot afford the luxury of only using 20% of its pack, so it would not last nearly as long with one.

 

[knightmb]

The NiMH battery is recyclable, lasts longer than the life of the vehicle, is recyclable, has adequate power, is cheaper than Lithium, is safe, non-toxic and is capable of deep cycling on a daily basis.

NiMH does not last long at all if you deep-discharge it. That's why hybrids always keep state of charge between 40% and 60%; they never stray much from half-full. That is what allows the packs to last for at least 200,000 miles. An EV cannot afford the luxury of only using 20% of its pack, so it would not last nearly as long with one.

To echo that, even the best battery technology isn't invincible. I've been using NiMH on my e-bike for years, but I don't run them to dead (well maybe only 6 times in all these years) and I try to keep them charged and topped off before each ride using a smart charger. So while I probably do push them a lot harder than what the EV cars did, I also take good care of them to help extend the life.

Everything has it's place, that's the great thing about battery technology. If you need some cheap "off the grid" power storage, SLA works great because it's cheap. For the mobile power user, SLA works, but it's weight and "death at discharge" state makes for issues that shorten the life. Enter NiCa, then NiMH, then finally Lithium-X. If I was going to build an EV from scratch and wanted it to be more affordable, then I would probably start with SLA. If I wanted some more power and range, move up to NiMH, then finally if you want a sports car and willing to pay the high price tag, enter the Lithium cells.

One day, Lithium cells might be as dirt cheap as alkaline , but until then we have to make due with what we have. NiMH has taken a hit in price increase pushing us more towards Lithium, figuring if you are going to spend more, why not get something better for slightly more?

That's the great thing about EV vehicles, things get better and you can "upgrade" them. When is the last time you said "you know, my car only gets 30 mpg, but this new fuel gives me 50 mpg, so I just starting buying it for the same price" with your gas vehicles? Electric is where it's at and it's just as fun to watch this "repeat history in action" as it was to watch computers really take off with the world and consumers.

One day our grandchildren will ask what it was like before IM and Cellphones, The Internet, and probably "Electric Cars".

 

[xyster]

If I wanted some more power and range, move up to NiMH, then finally if you want a sports car and willing to pay the high price tag, enter the Lithium cells.

One day, Lithium cells might be as dirt cheap as alkaline , but until then we have to make due with what we have. NiMH has taken a hit in price increase pushing us more towards Lithium, figuring if you are going to spend more, why not get something better for slightly more?

Decent lithium cobalt is cheaper and easier to find now than NiMH. It won't be long before even better LiMn and LiFePO4 cells are cheaper than NiMH too. Due to its inclusion in steel, now in much greater demand in China and India, nickel prices are not expected to decrease anytime soon. Lithium and the other constituents of lithium-X cells are far more plentiful than nickel. I'm very happy I skipped the nickel batteries and went straight from lead to lithium.
I see no compelling reason to choose NiMH now. Small 18v1.8ah Nicad powertools packs can still be had for $10 each. So those are a sensible option IF the EV'er is on a super-tight budget, and doesn't mind jumping through the extra hoops necessary to charge parallel strings of nickel batteries.

 

[jdh2550]

mcharles13 - the effective capacity of a battery changes dependent on the rate you discharge it. So, by just stating 2.3 Ah is incomplete. Because (even with Lithiums) if I draw 2.3 amps it will be at zero before 60 minutes have elapsed (unless it's the 1 hour rate that has been quoted). Because the effective capacity increases with decreased amp draw manufacturers like to quote the 20 hr rate. I believe it's a defacto standard (i.e. not strongly enforced). So, a 2.3Ah at 20hr rate battery discharge at 2.3 amps might discharge in say 55 minutes (wild ass guess!) - however, the relationship is exponential so the impact is much greater when you're looking at drawing 50 amps from a battery.

In the RC world what amps do people normally draw?

If you want to learn more google Peukert's effect, the first entry is the http://en.wikipedia.org/wiki/Peukert's_law <--- anyone know why that url isn't working?

Does that help explain my question / concern?

 

[TylerDurden]

the first entry is the Wikipedia entry <--- anyone know why that url isn't working?

Most likely the BBcode is choking on the apostrophe. Folks will just have to copy and paste the link.

:?

 

[jdh2550]

If people are interested in continuing the discussion of the politics of NiMH which CGameProgrammer and usatracy both alluded to - I've started another thread (because this topic REALLY interests me):

http://endless-sphere.com/forums/viewtopic.php?t=2342

 

[chuck]

jdh,

To the best of my knowledge, only lead based batteries have peukerts effect.

Your battery pack "lead" is rated at 2280 watt hours at the 20 hr rate. Throw in peukerts, you only get about 1/2 that if you discharge to 80%, or about 1140 watt hours. Using the gel cells, "lead silicone". These silicones probably have the worst voltage drop under load of any lead battery made. Probably have one of the highest peukerts of any lead battery made.

If you switched to the dewalts, from what I am reading here, a 1140 watt hour pack using the dewalts will give you the same range and speed that you are accustomed to. The dewalts don't care if you discharge them at the 20 hour or the 1/2 rate, they give the same total watt hours.

To design your theoritical pack using dewalts, I will defer to the experts here, they need the controller high and low voltage cutoff points, max amp draw, and the range required.

From what I have read here, in the long run, the dewalts will be much less expensive than lead. Not sure on the dewalts shelf life, but, Xyster has posted that it is not necessary to buy more than you need in pack size, as all batteries have a shelf life, they are going to degrade whether or not you use them, I think about 10% a year.

I started following the Lithium technology a year ago. Things have really changed. A year and a half ago, the latest and greatest was nimh, dumping the nicad was in vogue. This is a brand new technology, lipoh, in packs available to ebikers, just became available in the last few months.

Look at all the video here of Lowell, Knoxie, knightmb, very impressive. Knoxie's famous qoute, "dump the lead" anything else is better.

My battery pack of 864 watt hours of the best lead made, gets me about 12 miles at 25 mph, can get a lithium pack from electric rider, about 1/2 the weight, 35 miles at 35 mph. Availability is still in question.

I would like to see you start a thread and see what the experienced non lead users here can come up with regarding a lithium pack that will give you the amp draw, voltage parameters, and range necessary for your xm-2000. These guys here know what they are talking about. They will not let you down.

All the best,

chuck

 

[jdh2550]

only lead based batteries have peukerts effect

I haven't read all your post yet - but this statement is dead wrong. I suppose you're gonna make me work for the right to say that, eh? Read the wikipedia entry and you'll see that Peukert constants for lithium batteries are smaller (see example 2). But they do exist.

Also bear in mind that the constant is used as an exponent - so when we start looking at high discharge rates (say 20C or up) that constant has more significance. On the wikipedia page they're only looking at moderate discharge rates. This is important in this context because we're taking lots of little cells and asking them to perform at high C rates (21.7C for a 2.3Ah cell for my up to 50 amps scenario).

BTW, my assumption is that one has to look at the individual cell characteristic and not the combined battery pack characteristic. Is that a correct assumption?

Now, having responded to your opening line I'll go ahead and read the rest of your post... Thanks.

BTW - and this is from memory flooded lead acid has peukert constant of around 1.3 and lithium around 1.003. So Li is definitely way better (no dispute there).

 

[OneEye]

Peukerts effect holds for all battery types, as it is a different way of accounting for the battery's internal resistance. The exponent will be different for different types of batteries. A "perfect" battery would have an exponent of 1.00. For flooded SLA it is somewhere around 1.3. For SLA/AGM it is a bit lower. A123's are probably in the sub 1.1 range, which makes the effect much less visible.

[Looks like we were posting similar information at the same time. sorry for the duplicate info]

 

[Doctorbass]

Acording to battery University and other good sources, I understand that Peukerts effect is in part, due to a temperature increase when drawing more current. When temp is high, the electrochemical reaction inside the battery is less efficient.

Heat IS the first enemy of cells.. that's what is limiting the max current you can draw from. In relation to internal resistor.

So the lower is the internal resistor the lower the battery would loose energy in heat and the lower the peukert effect would affect it.. but the lead acid battery seems to me a bit more complicated with this relationship..

I consider that A123 cells have no problem with that.... maybe 0.xx %

 

[jdh2550]

mcharles13 - you say you're an A123 vendor - do you have access to a technical resource there? Ask them what the Peukert constant is for the cells you use.

doctorbass - what max amperage are you drawing? My point is that it might not be a big effect in your scenario but it might be a big effect in my 21.7C scenario.

 

[Doctorbass]

My senario already was 55A 20s1p A123 that is similar to yours

The battery heated up to around 40 degree Celsius. I lost only 3Wh out of 140Wh normally.

Doc

 

[jdh2550]

But how long do you draw 55Amps?

I draw 48amps while cruising at 42mph (WOT) on a flat road. Given that about 5 miles out of my 12 mile commute has either a 45mph or 50mph speed limit I am drawing that amperage for around 10 minutes.

For the rest of the ride I pull up to 48A when accelerating in stop and go traffic.

Does your scenario include such a situation? I'd be surprised if it did. I did see your stats where you state a 50A limit - which is indeed similar. However, I would imagine the amp draw of typical usage scenario is very different.

 

[mcharles13]

mcharles13 - you say you're an A123 vendor - do you have access to a technical resource there? Ask them what the Peukert constant is for the cells you use.

doctorbass - what max amperage are you drawing? My point is that it might not be a big effect in your scenario but it might be a big effect in my 21.7C scenario.

Good point. I'll shoot them an email today.

 

[OneEye]

A 21.7C draw can't last for 3 minutes even with a perfect 1.00 peukert exponent. This is not something that can provide you with 10 minutes of travel. When you add enough parallel strings to last more than 10 minutes you be pulling 1/4 of that from each cell, so only ~5.4C.

The ability of the cells to survive 20C+ discharge rates is great for those short-term power spikes, not for long term cruising. If you want to ride for 20 minutes you need enough parallel cells to make your overall average draw less than 3C.

 

[jdh2550]

BTW, my assumption is that one has to look at the individual cell characteristic and not the combined battery pack characteristic. Is that a correct assumption?

To clarify my quote - I mean with regards to Peukert effect. Whereas

When you add enough parallel strings to last more than 10 minutes you be pulling 1/4 of that from each cell, so only ~5.4C.

So, OneEye, you're saying that my assumption is incorrect? I.e. adding packs in parallel reduces the effective discharge rate, which in turn negates the high draw issues?

I thought that basically in a parallel circuit the amps flows through all part of the circuit at the same time (that's probably a lousy way of saying it - I'm a software engineer not an electrical engineer sorry). So, I thought that all those little cells were still working harder than they could for extended periods of time? I'm starting to think I'm over thinking this I can certainly buy your explanation. However, I'm not ready to plunk down a lot of cash yet...

Does anyone know of someone who has built a 20s5p pack for use with extended 45A draw? If not, then what's the closest someone has come to such a config?

 

[OneEye]

The current will be divided among the different parallel packs. The A123's play together nicely in parallel configurations.

GGoodrum is running with a 20s4p setup with his bike now. He has 8 dewalt packs wired for 72 volts (nominal). I think he is currently pulling power from the unswitched terminal, so there is no BMS involved. I don't know that he is pulling a high current from the setup though. I think Doctorbass' setup is something like 20s4p, so he would be another good reference. His peak current of 55A is in the range you are looking for, I don't know how long he runs that current. He does chime in on the 40mph club however, so you might want to ask him about distances at speed.

 

[mcharles13]

Hey guys. I'm not sure if this helps you but here is some data that might help regarding peukert effect.

http://www.rcgroups.com/forums/attachment.php?attachmentid=1100572

Here is one regarding longevity while abusing cells.

http://www.rcgroups.com/forums/showthread.php?t=672512

 

[Jozzer]

I've used both 20s3p and 20s4p at 40A with no sign of excess heat from the cells. At 10c (24A per cell) things begin to warm up, but since your down to 6 mins runtime at 10C I don't think this would become a problem in every day EV life. The same goes for voltage sag, at 5 or 6C pack voltage drops just a couple of V, at 10C it starts to get more noticable. (this equates to lost WHours from the pack, and risk of shortening pack life)

 

[GGoodrum]

45A sustained is not a problem for the a123 cells themselves. I have personally run lots of a123 setups in RC helicopters that will drain a 10s a123 pack in about 4 minutes, which is an average of about 34A, or 15C. The cells are good for at least 30C continuous and up to 50C bursts, so even in a "single-p" setup, the cells would easlly handle 45A. Using the DeWalt pack, however, means if you don't use the BMS, you are limited to 15A, or the internal fuse in the BMS will blow. That means to get 45A, you would need to run a 3p configuration (i.e. - 20s3p...), but probably 4p, just to be safe. My current setup is two packs in series, and four of these series pairs in parallel, or 20s4p. This actually gives you 66V nominal, not 72V. The packs are marked 36V, but they really aren't that high, except the first few minutes off the charger. We use 3.3V per cell for the nominal voltage rating so to get an equivalent to a 36V SLA, for instance, you need to use a 12s a123 pack configuration.

On my other two ebikes, I use a 16s4p (sometimes 16s5p...) setup on one, and an 18s6p (108 cells total...) configuration on the other. I usually see peaks of 2400-2500W on the 18s and 20s configurations, and peak currents of about 38-40A on the 18s setup and about 35-37A on the 20s setup.

-- Gary