Eric's Project #002

[D-Man]

Here's what I use for soldering. The 100 watt iron for soldering batteries and tabs. The smaller one is an adjustable one but the big iron works better on batteries. Just use sandpaper like mentioned above to roughen the surface.

 

[xyster]

I'm missing your point of why you wish to break the connection between the subpacks.

Hi xter. You're charging with individual chargers. I'm talking about one big charger charging all the batteries. If you connect the positive from one battery to the positive of another battery at the other end of the pack, you'll get a big short.

Yes, but why would you need to "connect the positive from one battery to the positive of another battery at the other end of the pack" in order to charge with a single charger?

The positive from the charger goes to the main positive power lead (or a tributary off that lead), and the negative from the charger goes to the main negative power lead (or a tributary off that lead). By using tributaries off the main power leads for charging, the main power leads needn't be disconnected in order to charge -- and the controller can be turned on without causing a short. I do this with my scooter and currie ebike no prob.

Of course, your batteries won't be balanced using this method. But as far as connection circuitry, using a battery balancer is no different than using multiple single-cell chargers.

 

[fechter]

It's pretty easy to destroy batteries when soldering to them. You need to complete the job in the shortest possible time to minimize the heat spread. If you don't get flow in about 10 or 15 seconds, back off and let it cool before attempting again. A wet rag or sponge can cool the end quickly after soldering.

Sanding the metal first is a good idea. The proper flux makes a big difference too. With the right flux, you can make a good connection to stainless steel (or any steel). Nothing I've seen works on aluminum.

You can test the battery with a magnet. If it's steel, the magnet will stick.

If the solder is balling up and not spreading, it probably means the iron doesn't have enough heat to get the metal up to soldering temp.

Tin the spot on the battery before attempting to solder a wire or tab. Allow it to cool completely between passes.

Welded tabs are much better to solder to. If you could find someone with a tab welder, they could put tabs on for you.

 

[Beagle123]

The positive from the charger goes to the main positive power lead (or a tributary off that lead), and the negative from the charger goes to the main negative power lead

That would be the case if I were charging at 54 volts (which is still a possibility). However, I was planning on charging at 4.2v. If you don't beleive me, try touching a few of your positive charging leads together. (Don't do it! )

Thanks for the tips fetcher. I sent my batteries to the guy xter suggested (mike @ bigerc.com) to get spot welded. I simply don't trust myself around batteries anymore. I've sparked about 5 times so far. Luckily these batteries seem durable.

 

[xyster]

That would be the case if I were charging at 54 volts (which is still a possibility). However, I was planning on charging at 4.2v. If you don't beleive me, try touching a few of your positive charging leads together. (Don't do it! )

OK, I see now your plan. So you'll need a total, 15-cell series/parallel switching scheme. That should be a challenge for a guy who says he doesn't trust himself with his batteries.

 

[fechter]

That would be a big block of wire connections.

Some guys are charging those with a simple regulated power supply (no BMS). A BMS would be safer, but perhaps unnecessary for charging.

I've seen some surplus power supplies that are in the right voltage range.

What is the maximum charging voltage for the whole pack? 63v?

 

[Beagle123]

That would be a big block of wire connections.

Actually, I don't think it will be that bad because I'm going to split the battery pack into two halves.

Let me explain my plan again. This is the plan that I currently plan to use:

6p, 14s

I plan to make 14 sub-packs with 6 cells in parallel. Each will be 4v, 18ah. The 14 sub-packs will be split into two sides (of 7 packs). Each side will have a plug with 14 andersens connectors. When the plugs are put together, all the sub-packs will be connected in series making 56v. When I pull the plugs apart, all the sub-packs will be disconnected from each other, and connections to each sub-pack will be exposed for charging.

To charge, I will use a power supply like this:

http://bigerc.com/product_info.php?products_id=170&osCsid=8b05d427c8d2fd7805cfe55d4f6b8339

It will have a charging cord with 2 plugs of 14 andersens connectors (7 X 2) to connect to the batteries.

See diagram. The boxes are andersens connectors. Imagine that the bottom block plugs into the upper one. WHen plugged together, the battereis are in series. The charging plugs wouldn't be identical.

Some guys are charging those with a simple regulated power supply (no BMS). A BMS would be safer, but perhaps unnecessary for charging.

Actually, it seems to me that charging with a quality machine like the one in the link above at 4.2v until the current is zero seems pretty safe to me. Please let me know if I'm wrong. Also, I'd never have to worry about balancing the cells which is one of the more dangerous situations (if I understand correctly). It seems pretty idiotproof to charge everything at 4.2v like xter does.

Sorry xter.

Why would I want to charge 14 sub packs in series and have to worry about what the voltage of individual cells in the middle of the pack? One of the only ways to make a cell explode is to overcharge, so why do I want to charge at 60v? Escpecially when 4.2 can't cross their threshold? WHy would I want to mess around with battery balancers, when charging at 4.2v will be the best possible banancing?

If there's some advantage to wiring them in series, please let me know.

Actually there is one advantage I can think of: If I used that power supply to charge two 28v packs, I could charge them at 28v, 20A. So I could completely charge my scooter in less than 2 hours. Charging at 4.2v, it will be only 1.4 amps per pack, so it will take 10 hours to charge. If I could find a 4.2v, 60 amp charger I'd be happy.

 

[xyster]

Actually, it seems to me that charging with a quality machine like the one in the link above at 4.2v until the current is zero seems pretty safe to me. Please let me know if I'm wrong. Also, I'd never have to worry about balancing the cells which is one of the more dangerous situations (if I understand correctly). It seems pretty idiotproof to charge everything at 4.2v like xter does.

Sorry xter.

Seems sensible to me. After an initial period of idiot-testing and idiot-proofing by said idiot, which resulted in two dozen dead, smoking cells, I'm now confident my system is indeed me-proof. And in the process I learned that an idiot can idiot-proof a system. Indeed, an idiot may be the best choice for idiot-proofing a system -- so long as the idiot accepts he's an idiot. As you obviously do too.

Why would I want to charge 14 sub packs in series and have to worry about what the voltage of individual cells in the middle of the pack? One of the only ways to make a cell explode is to overcharge, so why do I want to charge at 60v? Escpecially when 4.2 can't cross their threshold? WHy would I want to mess around with battery balancers, when charging at 4.2v will be the best possible banancing?

If there's some advantage to wiring them in series, please let me know.

The only advantage to charging in series that I see is not having to disconnect and reconfigure all the parallel subpacks in parallel and series with every charge and discharge, and create the unnecessarily complex system to do so.

Since you already bought a mess of cheap 4.2v chargers, I'm at a loss for why you don't just wire the packs in series and charge each subpack from it's own 4.2v charger like I do. Is it because your single-cell chargers (also the same as I use) don't charge your emoli batteries properly? Did you fry a bunch with one of your mishaps? Do you think the cells can't be charged in a balanced manner whilst remaining connected in series (they can), or that doing so will result in a short (it won't)?

 

[Beagle123]

Seems sensible to me. After an initial period of idiot-testing and idiot-proofing by said idiot, which resulted in two dozen dead, smoking cells, I'm now confident my system is indeed me-proof.

:lol: Blowing-up your cells is perhaps the best way to insure that they won't blow-up again. This makes sense to me too. That's what worries me.

as the idiot accepts he's an idiot. As you obviously do too.

I did a great idiot test the other day. I accidentally had the wire on a resistor slip under the paper barrier and short one of my cells. It got so hot so quickly, that my reflexes took over and it flew out of my hands.

I think that's proof that these batteries are good. That's a lot of power for 4v

"Stupid is ass stupid does"

I'm at a loss for why you don't just wire the packs in series

The way I'm planning is almost the same thing. If you think about it, I'm going to have to use connectors to connect the batteries anyway. I'm just putting all the connectors together. I also really like having access to each battery sub-pack. I can easily make a plug for load testing and voltage testing. I could easily test every one in about 10 minutes. Since I'm relatively new to this, I think it would be great to be able to be able to see what's going on in there.

I'd also prefer to just have one set of wires from the sub-packs (eliminate charging leads). I didn't like having those skinny little wires running around my battery box. Just more potential for a short. I like the simplicity of battery sub-packs with 10 guage wires from each one leading to the exterior plugs. Less can go wrong.

I'd prefer the big charger because its a quality machine. Those cheap chargers already failed on me, and I spent $260 on them. I can buy this machine for $250, and not have the risk of failure. Theres an additional risk of a single charger failing and I could be unaware of it (it took me a while to realize the others had failed because the green light was on). Also, this machine can be reused for other projects and be used for any type of batteries I use in the future. Another great thing is that you can see the current flow into the batteries. If I suspect a pack isn't charging I can charge it individually to see how the current flows. Really its more than just a charger, its a valuable tool. Watch this video:

http://youtube.com/watch?v=q-1fb9_V2II

Cool huh?

Disadvantage = 20 lbs

 

[Beagle123]

This is How I Want to Configure My Batteries:

THis picture shows how I'd like to configure my battery packs to fit into my battery box. There will be two layers of the batteries pictured below.

The triangles of masking tape represent two more triangles of batteries similar to the ones to their left.

 

[Beagle123]

Still Contemplating Batteries

I have a couple of options for charging. I'm thinking about both of these machines:

<a href="http://cgi.ebay.com/MASTECH-VARIABLE-REGULATED-DC-POWER-SUPPLY-0-30V-0-50A_W0QQitemZ120175445414QQihZ002QQcategoryZ32720QQssPageNameZWDVWQQrdZ1QQcmdZViewItem">50A power supply (30V max) $418</a>

or

<a href="http://cgi.ebay.com/NEW-MASTECH-REGULATED-LINEAR-DC-POWER-SUPPLY-0-60V-0-3A_W0QQitemZ120175814526QQihZ002QQcategoryZ58286QQssPageNameZWDVWQQrdZ1QQcmdZViewItem">3A power supply (60v max) $118</a>

The advantage of the 50A version is that I can charge the cells individually. However, if I'm charging at 4v, 50 amps (1/2 amp per cell or 1/6th C), the total power of the system is 200 watts. Its not bad, but these 6 cells can be chargeed at three times fasster rates.

Connecting two packs in series wouild redcuce the charge time because it increwases the charging violtage and watts. So charging douible packs together would mean an 8v charge @ 50 amps would give the cells. Each cell would receive 0.45 amps @ 8v. THis is equivelent to 400w.



A much faster charge time.

So charging them individually would be the slowers charge rate at 200 watts.

This would be about a 5-6 hour charge time

Pairing battery packs together allowss me to double the charging ouitput using 8v, 50amps, producing a 400 amp charger

The smaller 60v machine could charge my entire pack to 60v with 3 amps of current. I could then charger the whole pack in about 7 hours. (its 180w) THis is a very tempting situation. However, what if the batteries get out of balance?

I'm planning on havgin a 6p,14s configuration. What happens if pack 5 is 1 volt higher than the others. WHen I charge at 60 v, all the other batteries will eat 3.5v = 45.5v total, so pack #5 wil be at 15volts. If pack 5 refuses to draw any current when its full, no problem will happen excepth that the other batteries will be undercharged.

As long as I can run a check to see if the batteries are balanced, I can correct the situation so its fine.

With the 60v charger, I can charge at a max of 180watts which is livable)

Should I charge at 60v and just hope things don't get out of balance?
Should I charge induividual cells at 4.2 vols so everything is always balanced perfectly?
Should I pair the packs together, so I can double the speed of my charging time?

Should I separate both halfs, and charge them both at 30v, 50 amps (1500 watts?) THis would only take an hour or so.

 

[xyster]

As we've explored already, pulling lithium cells from packs basically gives us the choice, for charging, of:

A) Using multiple single-cell chargers, one per cell in series.
B) Using a single, larger charger coupled to a battery balancer.
C) Using a BMS on the pack to balance the batteries.

Lithium cobalt and lithium manganese cells have a very small margin for charging voltage error. 4.20v is the accepted value. 4.25v is OK, but life will be dramatically shortened. 4.30v and above risks immediate cell destruction and resultant fire or explosion. Therefore, the voltage to each cell in series must be tightly regulated. Are there any basic methods for doing so other than the ones listed above?

The voltage to each lithium cell in parallel thankfully regulates itself.

 

[Beagle123]

As we've explored already, pulling lithium cells from packs basically gives us the choice, for charging, of:

A) Using multiple single-cell chargers, one per cell in series.
B) Using a single, larger charger coupled to a battery balancer.
C) Using a BMS on the pack to balance the batteries.

Hi xter:

I think there is another option:

D) Disconnecting all the batteries and charging with one big charger at 4.1v.

The only element missing from this design is a component that instantly switches the batteries from being in series to being in parallel. I beleive this is an invention that would solve everything. I'm envisioning a big switch that contacts all negative to positives when thrown, and disconnects everything when thrown in other direction. I wonder if that exists?

I've really been enjoying looking for a solution to this problem. I'm thinking of the future. What are our options going to be? What are our biggest hurdles. Here are a few of my thoughts:

Small cells are a big problem. The current small format cells cause huge headaches because we have to string so many cells together. Current balancers only handle 8 cells or so, and they must be linked for more cells. Our problems go up exponentially with the number of cells.

More cells = more electronics = more potential for failure = more danger

Imagine if we tried to power a car. At some point we're going to want something that's 6hp or so.


The dreamer in me hoped that the car companies would slip-up and use a safe LiFePO4 batteris as a car starter battery. That would be awesome. Imagine if we could go to AutoZone and buy 5 12v batteries. This would "slam-the-door" on most of our problems. Imagine a fully-self contained 12v, 20ah battery that protects itself from overcharge and balances its cells. I don't beleive any additional electronics would be required.

If we can get any company to consolidate cells into safe, balanced packs, our lives improve exponentially.

I actually think that the approach people have taken to battery balancers is backward--they try to make one machine balance many batteries. I think if we could invent a simple circuit to balance two batteries we could make a better solution. First, you would make a pack of 2 batteries with the balancers circuit. Then, yo take two of those packs and put them together with another circuit etc etc. At least that is the idea I'd explore.

But back to reality.

Fortunately, a company called lifebatt.com has made a much bigger cell. This is very promising. They made a 3.3v, 10ah cell (3X Times bigger than mine!) Also, these cells have threaded rods on the ends so anyone can engineer a connection method (no more spot-welding and soldering!).

In the immediate future, these bigger cells should help us tremendously.

As far as charging, I think you had a stroke of genius by bypassing all the compllicated electronics and charging at the cell level. I'd only offer one improvement: I think using a big power-supply might be more reliable and faster. I plan to buy a 50A power supply. I will be charging at 4.1v so it will be charging at about 200 watts. However, this machine is capable of producing 1500 watts (50A, 30V). I will have the option of hard-wiring my sub-packs into pairs, and charing at 8.2v. That's the same as charging at 400w.

As a long term solution, I'd like to see another invention-- a 100 amp, 4.1v power supply (charger). I don't beleive this is a complicated invention. It just has to be a simple power supply that cuts off when the current drops to about 0.5 amps.

I just thought of another possibility: Why are you able to charge with individual chargers without a short? It seems to me that your chargers must be isolating themselves from each other using diodes or something. Perhaps there's a way to use diodes with a single charger so the 4v can be transferred to each cell without shorting out. I'll need professional advise on this one. I'm simply too dangerous. But I will reasearch it.

Actually, I could take that idea one step further--perhaps I could make a component that takes 30v, 50 amp input, and transforms the output into 4.1v, 365A output. That would be a 1.4C charge rate for my bike. I could charge it completely in about 1/2 hour. Think about that.

I could also achieve the same results by charging two 28v sub packs @ 50A and just forget about balancing.

I've been studying electronics a bit lately.

Lithium cobalt and lithium manganese cells have a very small margin for charging voltage error. 4.20v is the accepted value. 4.25v is OK, but life will be dramatically shortened. 4.30v and above risks immediate cell destruction and resultant fire or explosion.

That's true of li-cobalt, but not for li-mn. Look at this chart:

http://endless-sphere.com/forums/viewtopic.php?t=2252

Actually I don't see safety as being too much of a problem going forward. It seems like the safer batteries will be phased-in, and everything on the market will be battle-tested, or be charged individually like we both favor. Imagine if you had a car powered by LiFePO4 that you were charging at 4.2 volts using a quality power supply. Would you be able to sleep at night? I would.

Also, I beleive I read in battery university that charging to 4.2 volts is not optimum, and that charging to 4.1v will lengthen battery life.

from battery university:

3.92V/cell is the best upper voltage threshold for cobalt-based lithium-ion. Charging batteries to this voltage level has been shown to double cycle life. Lithium-ion systems for defense applications make use of the lower voltage threshold. The negative is reduced capacity.

Charging only to 4.10V reduced the capacity by 10% but provides a longer service life. Newer cell are capable of delivering a good cycle count with a charge to 4.20 volts per cell.

From this article:
http://www.batteryuniversity.com/partone-12.htm



I think I read the figure that overcharging reduces battery cycles from 800 to 300 somewhere. It seems like charging to 4.1v is a far better choice.

If you could buy a battery pack that's 10% heavier, but lasts twice as long, wouldn't you want it? I think its a no-brainer.


Summary:
My current opinion is that the best solution is to charge cells individually with a huge 4.1v (100 amp) charger. Hopefully, someone will make a 12v "Black box" battery soon, then we can charge with a 12v 100 amp charger. I'll look forward to when people ask me, "how long does it take to charge?" (been there?).

I estimate that my clothes dryer uses about 7200 watts. If it was a battery charger, it could produce 600 amps at 12v. If you had a bike with a battery pack that was 90v, 30ah (drool), my dryer/charger could charge it in 15 minutes. Its time to start converting laundromats into charging stations.

 

[xyster]

Hi xter:

Hi b123.

I think there is another option:

D) Disconnecting all the batteries and charging with one big charger at 4.1v.

Agreed. I missed that one.

The only element missing from this design is a component that instantly switches the batteries from being in series to being in parallel. I beleive this is an invention that would solve everything. I'm envisioning a big switch that contacts all negative to positives when thrown, and disconnects everything when thrown in other direction. I wonder if that exists?

Just a big bunch o' switches that seems unnecessary to me. You seem to be operating under a mistaken impression that chargers in series are not isolated without extra precautions. Two prong chargers like ours' are are always isolated. Hence, no shorts. Hence, no need for a complicated series/parallel switching scheme. And it's not like you didn't already purchase a butt load of single cell chargers anyways.

Small cells are a big problem. The current small format cells cause huge headaches because we have to string so many cells together. Current balancers only handle 8 cells or so, and they must be linked for more cells. Our problems go up exponentially with the number of cells.

No big problem for me, and with 300 on board, I'm the reigning small-cell ebiker 'round here. It was just a lot more upfront work is all.

More cells = more electronics = more potential for failure = more danger

Agreed. But in the case of small cells, also a lot more redundancy.

Imagine if we tried to power a car. At some point we're going to want something that's 6hp or so.

Tesla motors employs 6,831 of the same laptop cells I use.
I agree though that having the option to use larger cells is definitely better.

The dreamer in me hoped that the car companies would slip-up and use a safe LiFePO4 batteris as a car starter battery. That would be awesome. Imagine if we could go to AutoZone and buy 5 12v batteries.

It'll happen in not too long. 12v lithium batteries are now entering the market as a replacement for motorcycle lead-acid starting batteries.

If we can get any company to consolidate cells into safe, balanced packs, our lives improve exponentially.

It'll take a lot more than batteries to improve my life exponentially. However, if I could replace my need to eat with a 105-year replaceable battery pack, that could be exponentially freeing -- you think?

But back to reality.

Reality is overrated anyway -- at least this present reality is.

As far as charging, I think you had a stroke of genius by bypassing all the compllicated electronics and charging at the cell level.

There was no genius stroke involved. Others had shown how they had done the same with SLA and SLA chargers, NiMH and NiMH chargers.

I'd only offer one improvement: I think using a big power-supply might be more reliable and faster.

Agreed. But I'd note that lithium charging time is not inversely proportional to charging current. Charging at 2X current takes more than half as long as 1X current.

I just thought of another possibility: Why are you able to charge with individual chargers without a short? It seems to me that your chargers must be isolating themselves from each other using diodes or something.

Chrimany man (how is cry-man-e spelled ? oh well...) as I've tried quite unsuccessfully to explain, they are already isolated from each other by the very nature of the circuit. You should stop and think about it. There's no added danger of short circuit involved.

I've been studying electronics a bit lately.

Good. Study the part on short circuits and common electrical points again.

That's true of li-cobalt, but not for li-mn. Look at this chart:

http://endless-sphere.com/forums/viewtopic.php?t=2252

So it doesn't blow up, that's very good. Do you happen to know the voltage at which LiMn cells are at immediate and substantial risk of permanent death or irreversible damage?

Would you be able to sleep at night? I would.

Again, it'll take a lot more than that for me to sleep soundly.

Also, I beleive I read in battery university that charging to 4.2 volts is not optimum, and that charging to 4.1v will lengthen battery life.

from battery university:

3.92V/cell is the best upper voltage threshold for cobalt-based lithium-ion. Charging batteries to this voltage level has been shown to double cycle life. Lithium-ion systems for defense applications make use of the lower voltage threshold. The negative is reduced capacity.

Charging only to 4.10V reduced the capacity by 10% but provides a longer service life. Newer cell are capable of delivering a good cycle count with a charge to 4.20 volts per cell.

I mostly agree. After reading the Tesla motor's blog, other materials, and integrating my previous experience accidentally killing some of my cells, I'm of the opinion that LiCo, and probably LiMn, have a "happy range" between 3.80 and 4.0 volts, and that operation outside this range progressively and slowly damages the battery faster than operation within this range.

If you could buy a battery pack that's 10% heavier, but lasts twice as long, wouldn't you want it? I think its a no-brainer.

Unless for some reason (like the weight limit on a bike rack) that 10% made a disproportionate difference on the total systems performance or viability, then yes, that trade-off is also a no-brainer to me. Except I just used my brain to decide that. So maybe not. Arrrggghhh! Do you have to always do that to me!?

Summary:
My current opinion is that the best solution is to charge cells individually with a huge 4.1v (100 amp) charger. Hopefully, someone will make a 12v "Black box" battery soon, then we can charge with a 12v 100 amp charger. I'll look forward to when people ask me, "how long does it take to charge?" (been there?).

My response continues to be that your series/parallel switching scheme is unnecessary added complexity because our chargers are already isolated, and as my experience has proven, happily charge each subpack individually whilst they remain firmly wired in series.

I estimate that my clothes dryer uses about 7200 watts. If it was a battery charger, it could produce 600 amps at 12v. If you had a bike with a battery pack that was 90v, 30ah (drool), my dryer/charger could charge it in 15 minutes. Its time to start converting laundromats into charging stations.

Don't forget that charging faster shortens longevity just like discharging at a higher rate does. Why? Either way the cell heats up more, greatly speeding fatal chemical oxidation degradation. A 10 degree rise in temperature generally doubles the speed of a chemical reaction. I remembered that from General Chemistry -- impressive, eh? But I forgot if that was 10F or 10C. Probably 10C since metric is the standard in the university science education realm.

As for the laundromat, perhaps they should also work on harnessing the power of static cling.

 

[Beagle123]

Hi xter:

We seem to be in agreement about most everything except the one really important issue about isolating the cells for charging. I'm consulting bobmcree about this.

I wrote him this pm:

Hi Bob:

I've been having a disagreement with xter, and I want you to tell him he's wrong.

He's telling me that if I connect my battery pack in series, that I can simply connect a charging lead to each terminal of every battery, and charge them at 4.2 volts using a power supply.

I think he's trying to get me to explode.

Also, on his bike, he's able to charge his batteries while connected in series because he's using about 20 isolated chargers. Is it possible to charge batteries that are connected in series using a single power source by perhaps separating them with diodes?

Please comment on this thread:
http://endless-sphere.com/forums/viewtopic.php?t=1833&start=360

Thanks,
Eric

We'll see what he says.

 

[xyster]

We'll see what he says.

Cheese Whiz, Man! Sure, appeal to authority instead of using your head. OK, well what if Bob's wrong too? Less likely I agree since he's much smarter and more experienced with this stuff than we are, but still well within the realm of possibility, right?

To that end, I've attached incontrovertible photographic proof that I'm right and you're not. It's even got the word "proof" in the label, so it must be proof. If this were a court of law, we'd label this "evidence" because we wouldn't have known it was proof until a jury decided it was proof, and a judge accepted the jury's ruling. In this case, I'm quite sure I'm right, and so this is proof this is proof.

Follow along if you would, please.

Pic one: some of my 20, 15-cell parallel subpacks. Note the positive and negative labels, proving the parallel interconnectivity of each subpack (how could I have known in advance we'd be having this argument, right?). All twenty are constructed the same way as these.

Pic two: six subpacks wired in series and stacked atop each other, with charging leads coming out from the orange wire nuts through the box and out the top. There's another six subpacks arranged the same way on the opposite side of this box. There are no diodes, nonodes, hoodads, mossyfoots, or other such tricky componentry anywhere in this box, except for the power relay and switch -- the power switch circuitry -- which is wired in series with these subpacks, and sits between the two sides. I trust you accept that this main power on/off circuitry can not impact charging?

Pic three: The charging leads, terminated in all their powerpole connector block glory. No hoodads or mossyfeet here either.

Pic four: Charging the pack. Note: there is no smoke or flames, and the lights on the chargers are all red. As you ought accept because you have the same chargers, red means the chargers are actively working doing their job -- in this case charging each subpack individually whilst the subpacks remain wired in series. There are also 8 more subpacks in a box on the back actively charging simultaneously with the 12 in the box you can see.
It does not matter whether the main power switch is on or off. If it is on while charging, the bike is on, the voltage shows on the voltmeter, and if I twisted the grip the bike would shoot through the external garage wall. But until the point where the charging leads were ripped out, the chargers would continue charging the batteries happily.

Note that all 20 chargers are firmly plugged into a six-outlet power strip with some number of similar power-strip like thingies providing the 20 outlets for the 20 chargers. There are no trannyresistors, howdydoo's, or pieholes in this circuit either. I did not alter the chargers. I'm fairly certain you'll agree that I'm not smart enough to do that successfully (it would take Bob to do that). And Bob I trust will tell us truthfully that he had no hand in rigging this system just so I would win this silly argument which is silly only because it's so obvious to me that you're wrong.

Watching geeks argue is a popular spectator sport. We should team up and make a buck or two.

 

[Beagle123]

I'm not sure if we're having a disagreement or a misunderstanding.

I didn't mean to suggest that your love cruiser isn't roadworthy. I'm sure all the noodle wire, tuct tape and chewing gum is in perfect order.

What I'm saying is that you can't charge a bike like yours with a single power supply withouit isolating the chargers.

In the top picture, notice that the top battery's positive termainal is connected to bottonm of the battery below it. Short circuit.

 

[xyster]

I'm not sure if we're having a disagreement or a misunderstanding.

Yes, you are sure. See above where you wrote to Bob,

"Hi Bob:

I've been having a disagreement with xter, and I want you to tell him he's wrong."

I didn't mean to suggest that your love cruiser isn't roadworthy. I'm sure all the noodle wire, tuct tape and chewing gum is in perfect order.

All right then. I'm very sensitive you know about my wire, tape, and chewing gum, so I can't have anybody dissin' it.

What I'm saying is that you can't charge a bike like yours with a single power supply withouit isolating the chargers.

Correct. But that's not what you were stating before. You have changed your argument to conform to the photographic proof above. See where you wrote, <blockquote>"Why are you able to charge with individual chargers without a short? It seems to me that your chargers must be isolating themselves from each other using diodes or something."</blockquote>
And I was replying that no diode isolation blackbox is necessary -- just wire any chargers in series that don't have a third ground prong.

In the top picture, notice that the top battery's positive termainal is connected to bottonm of the battery below it. Short circuit.

Yes, duh, agreed, don't do that. All I've been arguing all along is that the series/parallel switching scheme diode/hoodad isolator machine is unnecessary complexity for charging. But this is Burger King, you can have it your way. Just don't expect others not to reply in kind when you try to take their bike in front of the judge (Bob).

 

[dirty_d]

what if instead of having a charger for every 4.1V subpack you have one for like every 3 or 4, it wouldn't get that unbalanced like that would it? maybe check it every month or something and charge the 4.1V subpacks individually if they get too far out of balance?

 

[Beagle123]

Yes, duh, agreed, don't do that. All I've been arguing all along is that the series/parallel switching scheme diode/hoodad isolator machine is unnecessary complexity for charging. But this is Burger King, you can have it your way. Just don't expect others not to reply in kind when you try to take their bike in front of the judge (Bob).

OK Ok Ok Already. Don't get your duct tape in a twist. I agree we're on the same page now. I'm trying to say that there are diodes in your chargers separate them from each other naturally. Keep in mind that all these chargers are plugged into the same power strip. I'm just trying to understand how to use a power supply to charge at 4.1v. We agree that the above diagram "won't work."

So my question to you is: How can I use a power supply to charge my sub-packs at 4.1v without separating them?

Also, about charging, its my understanding that emoli and A123 cells can be charged at faster rates. At battery university, he claimed that li-ion batteries should be charged at 0.8C or less. However, in the "notes" section of his chart, he claims "fast charging" for emolis:

http://www.batteryuniversity.com/partone-5A.htm

A guy on a RC Group posted this:

http://www.rccombat.com/forum/post.asp?method=ReplyQuote&REPLY_ID=108958&TOPIC_ID=12386&FORUM_ID=39

Cell type- Emoli from Milwaukee batteries
Mah-2800
Weight-99.2g or 3.5 oz
Voltage-4.2
Size-
Discharge rate-10C-12C
Charge rate-3C
Case type-Steel
Cost-$10
3s-297.6g,10.49oz
4s-379.6g, 13.4oz
3s2p-569.4g, 20oz
4s2p-759.2g, 26.7oz
+’s-Durable, low cost, Uses standard Lithium charger, Ease of building packs
-’s-Heavy, low C capablity

He claims a 3c charge rate. I don't know where he got the numbers, but I don't doubt them. I think it might be possible to charge these newer cells at a 1.5C charge rate without losing much because they don't heat up like the older ones do. These batteries could really be revolutionary.

Think of these three ideas together:

1) Fast 1.5C charging rates
2) Partial charge doesn't hut them (It helps)
3) You can use them immediately

So, I'm thinking that any EV with batteries with a 1.5C charging rate could be charged in about one hour. it would have a graph like this except that instead of 3 1/2 hours, it would be 1 hr:

<img src="http://www.batteryuniversity.com/images/partone-12.gif">

Now imagine stopping charging after about 35-40 minutes (this is about the 2hr mark on the graph). The area under the curve of the current is the amount of charge given to the battery. Visually I"d estimate its about 75%. So, in my ideal dreamworld, you could be riding out in the world, and have your batteries be empty and recharge in 35-40 min. If a 2C charge rate is realistic, its even shorter. And usually you don't drain batteries to empty, so it would be even faster still.

I think it may be realistic to "stop at a gas station" now.

That's why I'm so interested in this power supply.


Dirty: Yes, I agree that pairing the packs could be a good idea. As soon as I get my batteries going, I'm sure I'll start tinkering with it. I can try dismantling the charging plugs, and pairing the packs by connecting their andersens connectors. Then I'd make a charging plug that has 1/2 the connections, that charges at 8.2v and charges in half the time.

 

[xyster]

OK Ok Ok Already. Don't get your duct tape in a twist. I agree we're on the same page now.

Who me?

I'm trying to say that there are diodes in your chargers separate them from each other naturally.

It's really an academic point at best, but I don't see that diodes are required to separate the separate chargers when in series as I use them. We can agree to disagree on this point though, and yet somehow probably manage to remain friends and get on with our lives just fine.

Keep in mind that all these chargers are plugged into the same power strip. I'm just trying to understand how to use a power supply to charge at 4.1v. We agree that the above diagram "won't work."

Yes, totally agree. Your "won't work" diagram definitely won't work. As you've noted, the short is obvious and can be seen by tracing the wires.

So my question to you is: How can I use a power supply to charge my sub-packs at 4.1v without separating them?

Definitely problematic if not impossible with a single power supply. I don't know how it could be accomplished. This is probably an excellent question for the Bob's and the Fechter's here. The simplest, most assured solution -- if you can find and afford 15 such power supplies -- would be to use a single, 4.1v power supply of 1.5C amps per each subpack, and leave the subpacks in series.

Also, about charging, its my understanding that emoli and A123 cells can be charged at faster rates. ...

That's why I'm so interested in this power supply.

I don't offhand know the max charging rate of emoli's, but the 3C sounds reasonable for a 12-15C battery. 1.5C sounds twice as reasonable, and should extend battery service life as the cell will heat up less than at 3C -- though I also don't know by how much less. We have temperature versus discharge rate graphs available, but I don't recall seeing any temperature versus charge rate graphs. Running a chemical reaction the opposite direction I believe can be more (or less) exothermic (heat producing).

 

[fechter]

To charge your pack using the power supply, it would be best to leave the cells in series and charge the entire string at full pack voltage.

If you go this route, you may or may not need some kind of BMS to prevent any cell from going overvoltage. As long as none of the cells are bad, and they are balanced at the beginning, they should stay balanced without a BMS.

To charge all the cells at 4.1v without separating them, the only way would be to charge one cell at a time, then switch the charger wires to the next cell. This could take a long time, and you'd be in trouble if you forgot to charge one (not idiot proof). Or you could use the Xyster approach with a separate charger for each cell.

Making a massive switch or plug to change the cells from series to parallel is another way, but probably not practical.

One other possibility would be to "multiplex" the charger, having an automatic switch that puts the charger on each cell individually, but switches from cell to cell over a fairly short time period. This way all the cells get charged the same amount even if you run out of time to do a full charge. The multiplexer switch could be FETs or a mechanical stepper relay. This could be a bit complicated and the switch would need to take quite a bit of current. A BMS would be easier.

 

[Beagle123]

To charge your pack using the power supply, it would be best to leave the cells in series and charge the entire string at full pack voltage.

Why? I'm under the impression that the opposite is true--I think that charging on the cell level is totally superior to anthing I've seen so far becasue:

1) Safety. Zero chance of overcharge at 4.1v.
2) Auto-balancing. All positives and negatives connected when charging. Every single cell is at the exact same voltage every time you charge.
3) Easily diagnose problems. After a ride, bad packs will be at a lower voltage. Since all cells have an exactly equal starting point, its easy to see problems. Also, since the individual plugs are exposed for the charger, you can just touch a volt meter to them to get a reading instead of having to dig through your batteries.
4) packs are easier to replace because they're not even wired together.
5) You can use superior charging equipment. Like this:

100 amp dream-charger
http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&rd=1&item=320173610976&ssPageName=STRK:MEWA:IT&ih=011

Its more reliable, much faster, and gives feedback so you can see any problems.

6) Scalable. If you want to go to 90v, just add batteries. If you want it to charge faster, add another charger.

Am I missing something?

If you go this route, you may or may not need some kind of BMS to prevent any cell from going overvoltage. As long as none of the cells are bad, and they are balanced at the beginning, they should stay balanced without a BMS.

For how long will they stay in balance? A year? How will I know when they're out of balance unless I unpack all my batteries? Also, if they're out of balance in a year, I will have to balance them by hand, then keep testing, testing testing.

Making a massive switch or plug to change the cells from series to parallel is another way, but probably not practical.

I think it will be practical. I'm going to try the 7 X 2 andersens connectors. Hopefully, I'll just be unplugging the two plugs, then attaching a charging plug to each one. If not, I'll have to unplug 2 plugs and attach 4 charging plugs. I think that will be acceptable.

Strangely, the problem is finding a powerfiul charger. That 100 amp one looks like the best, but even with 100 amps @ 4.1v, it would only be charging my bike at 7.14 amps per pack. That's 0.4C. Ideally, I could use a 300 amp charger for 1.2C.

I'm planning to buy a 50 amp charger for 0.2C unless something better comes along.

I was thinking about pairing the batteries and charging at 8.2v. Hower, I think that might be even worse than charging at 54v because if a battery shorts it will overcharge at 8.2v, whereas if a pack shorts at 54v, each pack will overcharge at 4.4v. It seems like the best solution is to charge at 4.1 or 54v as fetcher suggests.

 

[Lenk42602]

my word - this thread gets the verbosity award for 2007!

 

[Beagle123]

I'm just glad xter and fetcher put up with me. Actually this is a huge issue that I plan to work on for a long time.