• Log in with Facebook Log in with Twitter Log In with Google      Sign In    
  • Create Account
  LongeCity
              Advocacy & Research for Unlimited Lifespans


Adverts help to support the work of this non-profit organisation. To go ad-free join as a Member.


- - - - -

1200 mile electric cars?


  • Please log in to reply
36 replies to this topic

#31 nefastor

  • Guest
  • 304 posts
  • 0
  • Location:France

Posted 11 March 2008 - 08:21 PM

Thanks Nefastor. I would be very interested in your follow up on this statement:

Naked cells can be combined in series and parallel like any batteries HOWEVER this is serious, potentially harmful business. If you're interested, I can post how to do it safely. If you make your own battery you'll also need to add battery supervision chips to it for safe operation (i.e. avoiding over- and under-charge). A bunch of these chips can be obtained as free engineering samples from companies like Texas Instruments. If you can make a high-current motor controller, you will have no trouble understanding how they work.


I thought you might :|o

Okay, naked LiPo cells such as the ones sold for R/C modeling are the real deal, but without any form of protection whatsoever except a passive anti-explosion feature : a high tensile-strength internal net that will cause them to swell A LOT when mis-used, giving you enough time to disconnect them before they explode. I've seen it work, it's amazing. Problem is, this is not an elastic net, and a swollen cell is as good as dead. I've tried recharging one, and it wouldn't hold more than 10% its original capacity.

The main issue you face with interconnecting battery cells is balancing : when they are delivered to you, each cell will carry a different charge. This MUST ABSOLUTELY be taken into account ! Also, because of rather small voltage difference over the charge range of the cells, you CANNOT consider as equally charges two batteries that show the same voltage on the multimeter, unless your multimeter is a well calibrated unit with at least 20,000 points.

If you connect in parallel two cells at different charge levels, one will start charging the other. However, because there is no current limitation and the cells have a low internal resistance, you run the very real risks of exceeding the cells charge and discharge current (resulting in explosion) and turning your connections into molten metal.

If you connect in series two cells at different charge levels, it will be impossible for your charger to reliably measure the actual voltage on each cell or the battery, resulting in the most charged cell in the battery to be overcharged. That means KA-BOOM.

Therefore it is essential to manually balance the charge of cells (or batteries of cells) you wish to interconnect. There is NO easy, practical way to do that. The method I use, however, is the cheapest and least likely to kill you, because it's too simple you can't get it wrong.

The basic idea is to let the cells balance each other, in pairs, by letting the most charged one discharge into the least charged one. But what you want is to make sure you do not exceed the rated currents for your cells.

You do that by using a resistor :

- Connect the negative poles of two cells together

- Connect the resistor between the positive poles of the two cells

The resistor's value is the voltage difference between the two cells divided by the charge current for your cells. Keep in mind that, for large AND unbalanced cells, a regular quarter-watt resistor may not be strong enough. Also, since you'll be dealing with fairly low resistance values, you may need to use several resistors in parallel to get the right value.

You REALLY don't want a value smaller than what you really need (or else, KA-POW !) but also, a value too large will vastly increase the time it takes for the cells to balance each other.

In fact, as the cells balance each other, you may want to gradually reduce the resistance to make the process faster. In order to know when to do that and by how much, measure the voltage across the resistor : using Ohm's law, this will tell you the current flowing through it. To reduce the resistance, simply connect more resistances in parallel with the first one.

When you cannot detect any more current through your resistance, it means the cells are balanced and safe to connect in parallel or series directly.

I REALLY MUST STRESS, IF THERE'S EVEN THE SMALLEST DIFFERENCE IN VOLTAGE BETWEEN BOTH CELLS, YOU WILL DAMAGE THEM. YOU REALLY WANT TO HAVE A GOOD QUALITY VOLTMETER, ANYTHING BOUGHT AT WALLMART WILL GET YOU HURT.

On another note :

- for PARALLEL setups it is not necessary for the cells being balanced to be of identical capacity BUT it is essential that they come from the same manufacturer and use the same chemistry. You can get significant differences from two different manufacturers, regarding the actual EMF of the cell, its internal resistance and how much current it can take (or deliver).

- for SERIES setups IS IT VITAL for the cells TO BE IDENTICAL, especially in terms of capacity.

- to balance more than two cells, you MUST balance them pair by pair, then balance pairs together, and so on : while you can balance a pair, then balance that pair with another cell, and add cell after cell, if you do that you increase the currents you'll have to deal with, and it's never a good idea. If your number of cells is not a power of two, you'll have to do that eventually, so just be careful and remember to use high power resistors.

- it's a good idea to keep ANYTHING flammable WELL AWAY. A shorted battery will put out 60+ amps, usually in wires that aren't rated that high. I've seen people set fire to wooden benches that way.

- it is a GREAT idea to keep anything metallic WELL AWAY. The last thing you want is for a screw driver to roll against your naked cells and short them up : the current will fuse everything in place, in which case you'll have to run for your life.

Now, I know I seem a bit paranoid about naked LiPo cells, but I really can't stress enough : just like rare-earth magnets, if you're not careful these things will kill you, or at the very least maim you, blind you, burn you. You have to treat them with respect until they are finally assembled as a working battery.

Now, balancing the cells is actually only the first step in safely making a custom LiPo battery. What you need next is to ensure the cells will remain balanced during charge, and that they won't overcharge or undercharge. In the early days, this used to be a complex engineering challenge, luckily we've come a long way since then.

What you need now is to equip your battery with SUPERVISOR CHIPS. How many chips you'll need depends on how many cells your battery has. A great place to start is the BATTERY MANAGEMENT section of integrated circuit manufacturers. I personally use Texas Instrument chips because they are nice enough to send engineers free samples of almost anything we need. I can't post a link because the TI site is way too dynamic (so it would probably be broken after a few days or weeks) so :

- Go to TI's web site

- Look for a section such as "power management"

- There will be a subsection about batteries

- What you are looking for are any chips relating to lithium battery protection

Now, some bad news : LiPo batteries are intended for portable devices (mainly). That means the chips you'll be dealing with will always be small and hard to solder by hand. This cannot be avoided. Since this post is already getting gigantic I won't digress into this, but there are many ways to work with these small chips without breaking out the super-expensive professional tools. If you need pointers, you know who to ask.

And now the good news : LiPo batteries are intended for portable devices (mainly). That means they have many excellent features, some of which might go beyond what you need. The most important advanced feature is that these chips all have some sort of high-level management interface (typically an I²C bus) that will allow you to manage very large batteries using a microcontroller or any computer with a suitable interface.

You can use that interface to do any number of nifty things, such as checking the status of each cell in the battery independently and balancing the cells in your battery. You can decide yourself at which point the battery should be disconnected from its load to avoid undercharge, or disconnected from the charger to avoid overcharge, and of course the chip will provide you with an accurate reading of how much juice is left in the battery. In addition, supervisor chips often feature outputs to directly drive LED's that inform you on the status of the battery.

The most important feature of the chip is its ability to disconnect the battery in order to protect it. This is usually done through an external MOSFET transistor that acts like a switch. MOSFET selection is, in itself, worthy of its own discussion as the wrong MOSFET can reduce the battery's performance.

It's worth noting that you don't HAVE to use one of these chips but you MUST somehow implement their protection features. Many amateurs find it more comfortable to do so using PIC microcontrollers, but this will always be a suboptimal choice, typically shortening battery life AND reducing its available capacity.

Nefastor

#32 Hedgehog

  • Guest
  • 462 posts
  • 1

Posted 11 March 2008 - 11:22 PM

I have one word regarding the Aptera: Oversteer. I think the handling in this thing is going to be scary.

Comparing the Aptera to a more normal PHEV (e.g. upcoming Prius or Volt)

Aptera: 12000mi/yr / 300mpg = 40 gpy
Normal PHEV: 12000mi/yr / 100mpg = 120 gpy

savings over normal PHEV: 80 gpy; $240-$400/yr. So for this modest fuel (equivalent) savings, you give up... handling, luggage capacity, climate control that actually works, who knows what else? It does look cool though.


What do you think the cost would be per year if you factored in the cost of recharging the batteries? I don't know that much about the these types of batteries nor how much NRG it takes to charge them.

TIA

Edited by Hedgehog, 11 March 2008 - 11:24 PM.


#33 niner

  • Guest
  • 16,276 posts
  • 2,000
  • Location:Philadelphia

Posted 12 March 2008 - 02:15 AM

Fuel cells have it all over batteries until we develop fast charging ultracaps because of the recharge turn around that is a tactical vulnerability of electric tech.

This is also one of the reasons that hybrid fuel electric plug-ins have it all over pure EV's in the real world. Who wants to wait hours to use your discharged vehicle when the need is NOW?

If you have a high amperage charger, (and that may be a big if in some circumstances) can't you get batteries now that can be charged to something like 80% capacity in a time frame measured in minutes rather than hours? Certainly ultracaps will solve this problem if or when they appear. (I keep an eye on EEstor, hoping for the best there. If they can deliver on their promises, it will be a whole new ballgame.) I note that a typical trip to the gas station could take me close to ten minutes, and given that a recharging station represents a vastly lower capital cost than a gas station, I would expect to find them all over the place eventually, particularly in parking lots where I would be leaving my car for extended lengths of time anyway. I think the requirement for rapid recharge is a bit overdone; all we really need is to be able to get the vehicle charged adequately during existing periods of normal downtime.

sponsored ad

  • Advert

#34 Lazarus Long

  • Life Member, Guardian
  • 8,116 posts
  • 242
  • Location:Northern, Western Hemisphere of Earth, Usually of late, New York

Posted 12 March 2008 - 03:00 AM

Niner I think that large capacity charging is not going to be an available infrastructure anytime soon and definitely not if you have to carry the charger around with you.

I made my comment mostly about the tactical disadvantages in terms of modern military applications but the way around the whole problem for smaller vehicles is simple, rechargeable detachable packs analogous to how we change battery packs in a cordless tool but scaled to about the size of barbecue LP tanks. This way a standardized power pack could be pulled and traded at any local charging stations and exchanged for fresh packs that would have SOC indicators on them. It would also allow a flat exchange rate per pack or some variation of SOC plus exchange and routine inspection/maintenance cycles because of on-board status circuits built into the price.

An error code would trigger service for a power pack and doing it under the right conditions would expedite the return to service, maximize safety and proper recycling of the dead cells that get removed, all while keeping such service costs to a minimum.

A small vehicle would use one or two packs and a larger vehicle would use more but fuel hybridization (whether with combustibles or fuel cells) means that recharging is a function of operation and this is tactically necessary for the logistics of military ops that can never depend on recharging *stations* in the field nor afford the *time* recharging requires in critical situations. Fuel (combustible or cell)-Electric hybrids also require smaller battery packs.

Plug-In fuel hybridization also mitigates the need for as many charging stations since the Plug-In option would allow many non inner city owners to also charge their own vehicles at home and sometimes at work. I doubt we are going to see too many public charging stations in the inner city except when attached to a parking meter that will also be automatically deducting payment for a charge along with parking directly from your ATM the way E-Z Pass works now. :|o

I do see urban centers getting into this business for the revenue.

I agree about the ultracaps and I have high hopes for them but 80% of charge in large battery packs could still take 45 minutes to an hour (instead of 3, 4 or more) and that is only after you are hooked up. A gas station that had 25 vehicles waiting to be charged at that rate would have a riot on its hands today. The actual fueling today takes under 5 (more like 2-3) minutes for the majority of vehicles and the added time is waiting for a pump and/or paying and those parts would not change.

#35 Lazarus Long

  • Life Member, Guardian
  • 8,116 posts
  • 242
  • Location:Northern, Western Hemisphere of Earth, Usually of late, New York

Posted 12 March 2008 - 03:14 AM

TIA thank you for reminding everyone that TANSTAAFL (there ain't no such thing as a free lunch) still rules. Yes there is a charge for pulling power from the grid and how you recharge will have a built in cost but it tends to be considerably less than the fuel equivalent and also will extend the life expectancy of the battery packs.

Also *if* you have solar panels and/or wind generators then storing that energy in the vehicle batteries when not sending it back through grid tie is a very reasonable option.

Currently all grid based electric vehicles are really roughly 60% coal, 15% hydro, 20% natural gas and 5% nuclear powered since that is where most of the electricity is coming from. At the present time all the solar, wind and other alternatives combined account for less than 1% of total electricity production, though it's changing.

#36 niner

  • Guest
  • 16,276 posts
  • 2,000
  • Location:Philadelphia

Posted 12 March 2008 - 03:42 AM

Niner I think that large capacity charging is not going to be an available infrastructure anytime soon and definitely not if you have to carry the charger around with you.

I made my comment mostly about the tactical disadvantages in terms of a modern military applications but the way around the whole problem for smaller vehicles is simple, rechargeable detachable packs analogous to how we change battery packs in a cordless tool but scaled to about the size of barbecue LP tanks. This way a standardized power pack could be pulled and traded at any local charging stations and exchanged for fresh packs that would have SOC indicators on them. It would also allow a flat exchange rate per pack or some variation of SOC plus exchange and routine inspection/maintenance cycles because of on-board status circuits built into the price.

An error code would trigger service for a power pack and doing it under the right conditions would expedite the return to service, maximize safety and proper recycling of the dead cells that get removed, all while keeping such service costs to a minimum.

A small vehicle would use one or two packs and a larger vehicle would use more but fuel hybridization (whether with combustibles or fuel cells) means that recharging is a function of operation and this is tactically necessary for the logistics of military ops that can never depend on recharging *stations* in the field nor afford the *time* recharging requires in critical situations. Fuel (combustible or cell)-Electric hybrids also require smaller battery packs.

Plug-In fuel hybridization also mitigates the need for as many charging stations since the Plug-In option would allow many non inner city owners to also charge their own vehicles at home and sometimes at work. I doubt we are going to see too many public charging stations in the inner city except when attached to a parking meter that will also be automatically deducting payment for a charge along with parking directly from your ATM the way E-Z Pass works now. :|o

I do see urban centers getting into this business for the revenue.

I agree about the ultracaps and I have high hopes for them but 80% of charge in large battery packs could still take 45 minutes to an hour (instead of 3, 4 or more) and that is only after you are hooked up. A gas station that had 25 vehicles waiting to be charged at that rate would have a riot on its hands today. The actual fueling takes under 5 (more like 2-3) minutes for the majority of vehicles and the added time is waiting for a pump and/or paying and those parts would not change.

Laz, I totally agree about the military applications. However, consider the capital cost of a charging station. Once you have a power controller and credit card processor, all you have to do is lay cables to the various parking spaces. A docking station-like architecture would allow cars to plug in just by driving up to a spot with a pop-up connector. Such a thing could be easily built, if manufacturers could agree to a standard. A gas station could have 25 slots for a lot less money than the tanks, pumps, and environmental clearances for hydrocarbons. Because of the relatively low cost and lack of significant environmental issues (relative to a gas station, certainly) such hardware could be installed anywhere there was a big enough power line.

I found an old article about an insanely fast charge battery that Toshiba had prototyped; I don't know what became of it. I'm sure that I've seen more recent reports of fast charge batteries that are pretty quick; like 15 minutes as I recall for something like 80%, though I don't have them at hand.

From http://www.greencarc...barsquos_f.html

Toshiba has developed a new fast-charging lithium-ion battery with an extended lifecycle that has significant potential for application in hybrid and full-electric vehicles.

According to the company, the prototype of the battery can recharge 80% of its energy capacity in only one minute, approximately 60 times faster than the typical lithium-ion batteries in wide use today, and will lose only 1% of its capacity after 1,000 cycles of discharging and recharging...


I also agree that the near term future of electrified transport will be hybrids and PHEVs, not pure EVs. I suspect that range extenders will be almost entirely hydrocarbon ICE, at least in civilian applications where money is an object.

#37 nefastor

  • Guest
  • 304 posts
  • 0
  • Location:France

Posted 12 March 2008 - 05:48 AM

Currently all grid based electric vehicles are really roughly 60% coal, 15% hydro, 20% natural gas and 5% nuclear powered since that is where most of the electricity is coming from. At the present time all the solar, wind and other alternatives combined account for less than 1% of total electricity production, though it's changing.


Well, that's really a societal problem, not a technical one. I know I bring this up a lot, but in my country electricity either comes from nuclear or hydro.

Ultimately, I think solar is what we should aim for (as a species), because it's about using energy that hits us in the face (literally) no matter what we do. It's not like tapping an oil deposit, damming a river or defacing the countryside with thousands of windmills (which require serious maintenance). If we improve the energy efficiency of our technology, we may not need to do more than install solar panel above our rooftops. That'd also eliminate the need for nation-wide power grids, which have proved to be vulnerable to extreme weather and human error.

Between batteries and solar panel, you get what I think is the alternative with the smallest environmental footprint AND the least need for maintenance. We'll probably still need nuclear, though, because sometimes you just need the gigawatts (say for the heavy industry)

I was just thinking : civilian vehicles, especially in cities, for commuters, could be equipped to receive power wirelessly through the road they drive on, or maybe recharge wirelessly when on a parking spot. It'd be as easy as embedding large coils into the pavement and fitting secondary coils on the cars. Think of it as heavy-duty RFID. While this might not make batteries superfluous on city cars, it could reduce the size of the cars' batteries, and therefore the (initial and recurring) cost of the cars. Economically, some cities like London already require drivers to pay a tax for driving in town (to help with congestion) : the tax money could go towards paying for the electricity.

Niner, your charging stations could be based on this kind of technique.

EDIT : I totally forgot, I actually have those "insanely fast charge batteries" you're talking about. You can buy them at Wallmart, I suppose (got mine at the French Wallmart, Carrefour) - Varta makes those :

They are called IC3 (stands for in-cell charge control, IIRC). Basically this is a standard NiMh cell (AA-size, 1.2V and 2.3 Ah) which you can really charge 100% in 15 minutes flat.

The secret ingredient is an internal thermic switch that will cut-off the cell's positive electrode when the cell reaches a temperature consistent with full charge. And they do get really hot : the charger I use actually has a miniature fan to cool them during charge.

How did I manage to forget about these ? I've had them for a year now.

Nefastor

Edited by nefastor, 12 March 2008 - 06:00 AM.





0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users