PINNED - Lithium battery conversion/info

[Burgerman]

Yep you could. But I hate modular. Prefer to build something designed as a unit to work properly together rather than a kit of parts. The live forever lithium, depends on cell matching and very careful management that never allows a cell to go over voltage at the end of charge, and never stored full, or run too low. But basically yes. Of course the ideal setup is as much Ah on board any given design as possible. Because this means longer life, as well as better performance or range.

Why?
Because you get 1500 cycles at 100 percent.
And 2000 at 80 percent.
And 10s of thousands at say 60 percent.

But the bigger the pack the longer you go in days between charges. So I may charge only once a week in winter... That way these cycles may last a very long time.

Lithiums last longest with average discharge rates (say 1c or less), not being "full" too often, not being discharged too deeply, keeping cool, careful balancing and no overvoltage on charge/balance.

They work best when warm. So will give less range in winter. Say down by 15 percent, and shelf life is expected to be 10 to 15 years.

[Burgerman]

I've been playing around with Woodygb's sketches and these cells allow for so many layouts, provided you keep the 8 (2 x 4) cell-groups to get 24v.
Using 6, 7, 8 or 9-cell groups gives you 72, 84, 96 or 108 (lithium) ah packs.

This allows for use in existing chairs. But sticking with 24v kind of misses out on some of the lithium advantage.

Increasing voltage is possible because of the reduced BATTERY resistance and no voltage drop when the going gets tough unlike lead where it drops like a rock. Meaning further efficiency gains for free, such as extra speed, without loss of torque or range. Or smaller lighter motors instead, with slower speed gearing so "actual" speed stays the same and we get free increased torque... Etc.

You cant do this with lead as increasing voltage means decreasing Ah in the same space. Which means it cannot handle the amps, and it struggles now with gel and faster chairs...

[Step]

I agree with the unit design idea.
But you didn't build everything from scratch either though, you selected the best parts available for your needs, improved them if needed and built custom ones when what's available didn't do.

I think modular design with decent parts could work if done right.
Most means of personal transportation are modular to an extent, aren't they? bikes can change steer, saddle, gears, wheels, tires... so can cars, ...

For the extremes... custom design will always be the way to go.
I think you classify for extreme. no offense.

[ex-Gooserider]

To me the difference is between whether the custom changes alter the basic package of the item or not... Changing parts like seats or tires on a pedal-bike don't really alter the basic design.

Playing with casters and some of the other things you mention would be sort of like building a car with modular drop-in engine choices (and yes it is possible to change engines, but it usually is pretty major surgery...

Keep in mind that it is NOT a good idea to add new batteries to an existing lithium pack (at least as I understand it) so the pack really can't change to grow with the user...

IMHO if one is going to do anything modular, I would say that the current rehab chair pattern of a "power-base" that a "seating system" fastens onto is probably the best approach... I would deal with having different size chair needs for growing kids by having a better "exchange" system so that parents can "trade up" as the kids grow - unfortunately we keep getting enough handicapped little kids coming into the system to use the outgrown chairs as well...

ex-Gooserider

[Martin O Refurbisher]

Sounds exciting!

Best,

Martin

[Step]

Ex-Gooserider,

I wouldn't compare changing the battery box shape and bolt on casters and seating with changing to a different engine in a car.
That is a daunting task, no doubt.
As long as you stick to certain rules like 'Casters are best placed in a 90° angle to the floor' etc... It seems pretty straightforward.
BM already bolts the motors/rear wheels to the battery box. A smart man/woman might come up with a decent caster bolt-on design as well, maybe even with some sort of suspension.
A standard could be set for attachment hole patterns for instance, making switching to larger motors or casters a possibility.

In my imaginative chair, you just bolt 4 corners (2 casters, 2 motors) to a rectangular box of wider or smaller shape, depending on your needs.
The top of the box is a rail kinda thing that allows C of G adjustment by sliding the seat forward or backwards as needed. Central footrest is attached to the seat frame.
In a way, that's what my Handicare Alex chair already is like right now. with lift, tilt & recline as well.

This makes the wheel components the 'base' part, the battery box the 'structural' part determining chair width and battery capacity and the seat the 'grow' part for kids or any other user's changing needs.

From what I've learned sofar on lithium batteries, adding more cells to an existing pack sounds like a bad idea indeed.
But putting the same one in a larger or different shaped box would work perfectly...

And upgrading to a completely new, larger set + battery box would still be cheaper than buying a complete new chair, wouldn't it?

[Seajays]

How about these?

http://evolveelectrics.com/CALB.html
http://evolveelectrics.com/GBS.html
http://currentevtech.com/Lithium-Batter ... l-p25.html

[Burgerman]

Ok...
Went through this before...

CALBS cheap, but turned out to be hard to get, and in the sizes I wanted (60Ah discontinued), and half te C rate of the headways...
Thundersky have a bad reputation on EV sites for reasons unknown
The third one no idea!

[nandol]

ok fellas,in the next 2 weeks Nandol=tech looser=me? lol,i´ll start the lithium tests with one 24V battery http://www.autosil.pt/abrir/gama_de_pro ... _iao_l_tio
the young engineers want me to test it in " normal user " conditions,they want that i use the chair charger,and they will install a small device near the joystick pod to do several measurements...
tests will be made first in this chair i´ve bought in Germany http://www.ebay.de/itm/Elektrorollstuhl ... 43a9a98ef5 (now the seat is a recaro elec.)
but i´d really like to test it at Martin s type of bm2,to feel the 13km + rnet,range and torque...so Martin,take care of your " squid legs " and give speed at the workshop

obrigado

[Blade Runner]

Just musing out loud, and thinking budget. I have group 22's which are 50ah. From what BM has said
I really only have about 25ah at the 1c rating, If I used 16 12ah headways that would give me 24ah at 1c.
32 would give me 48ah at 1c which would still be twice what I have in lead. I know it's not taking full advantage of using lifepo but it would be more affordable.

Bob

[Burgerman]

With lead, there are a bunch of things conspiring against you. A 50Ah batt has a "usable" approx 25 to 30Ah. Because:

a) you discharge it below 80 percent a few times and it goes in the rubbish... So that means a 50Ah battery is 40Ah Max...
b) You have that Peukert (surface charge/discharge) guy hitting on you heavily if you take out power as fast as we do. Remember big dollops of power are removed at say 100 amps (a 50 amp controller). Thats 2C or 30 min rate... So you lose lots of that 40Ah we had left.
c) the battery voltage falls fast under load as the acid strength weakens during use... So even if 1/3rd charged the batter cannot maintain volts to shove you along. So when you run out of go, theres still enough to light a bulb for 2 days... Its just not accessible Ah.

A lithium battery of say 1/2 to 3/4 the Ah (30) will give the same range. bECAUSE ALL OF ITS CAPACITY, 100 PERCENT, IS USABLE.
But you didnt win anything by changing so its pointless. Unless a marketing point or two, as a manufacturer.
Plus you will be working them very hard (high C loadings shorten life). Discharging them fully (100 percent is usable but less cycles) giving say 1000 to 1500 cycles as long as you dont go below approx 2.5v per cell. Thats half or less than you will get with a bigger battery.

You save initial cost but lose the main benefits. One of which is economy. And how long they will last is debatable, as you are going to be working them very hard. Same current, half the battery and deeper average discharge...

When the manufacturers start fitting lithiums it will be like this. With no or very little real gain, but good marketing kudos for those that dont really get the whole picture. Lithium is lithium? But its not.

Ideally you want to:
1. increase the capacity as much as possible as they allow this in the same space.
2. decrease weight - a 50 percent capacity increase, means a 50 percent weight reduction usually.
3. increase longevity to maximise investment - by having lower loads per Ah (bigger battery) and lower average DOD and less charge cycles. (weekly in winter?)
4. Increase potential range by fitting more Ah in the same space
5. improve reliability by stressing the batteries less.

But this will cost more initially since you are actually buying 2x the battery capacity.

[Lord Chatterley]

Just out of curiosity - how would you configure the charge leads and the balance wires to charge an 8S pack with the Hyperion?

LC

[Burgerman]

You can do 4 cells to each (7 cell) balance connector.
Or 7 to the first one (a) and 1 to the other one (b).
All can be routed through one D connector on the chair. Keep all balance wires short, or less short and a bit fatter that the short ones...

[malamar]

i think that the availability of an easy swap lith batt is to be welcome by all: its expensive, right front, but convenient in the long term if you can do the initial forking out.
But i have priorities,in my chair, before the cake of lithium. The dismounting in pieces act makes it doable but extremely inconvenient an weaken the chassis to no real practical advantage (thought to exchange for samba light rigid one) but the real McCoy w'd be mayor surgery to see how to install a center longitudinal batts case as in BMs, to narrow a too wide back path for installing turf tyres All alloy big tubed frame, of course... Second and more of a terra incognita project wil be front caster suspension arms a la Permobil or Storm way, the best there are it seems for kerb "guerrillas"...

Buen dia

[Lord Chatterley]

You can do 4 cells to each (7 cell) balance connector.
Or 7 to the first one (a) and 1 to the other one (b).
All can be routed through one D connector on the chair. Keep all balance wires short, or less short and a bit fatter that the short ones...

Like this-




LC

[Burgerman]

Yep, but I use a single connector on my 13s pack on the chair, for ease of use..

[Lord Chatterley]

Yep, but I use a single connector on my 13s pack on the chair, for ease of use..

Yes, through a D-connector.

I assume the balance boards are not just for different cell sizes but because some packs have unique-ish proprietary balance taps so you require different boards?

Do the balance wires both charge and discharge the individual cells or only discharge - I imagine overcharging is the main hazard with lipo especially.

LC

[Burgerman]

The charger charges the lot, in series, via the thick wires at up to 550 watts, or 20 amps, whichever comes first.

The balance taps, dont require any board, go straight to the chargers balance connections. As it charges, ANY cell that goes above 3.600v has a 1 to 300ma "brake" or pull down applied via the tap as required.

And if this isnt adequate to hold the cell to 3.600 during charge, the charger lowers the max charge amps, until it absolutely is enough. In an extreme situation, with out of balance cells, this can result in a max charge rate, at the end of charge of just 300ma, As more would sent a cell, or cells, higher than 3.600v.

[Lord Chatterley]

The charger charges the lot, in series, via the thick wires at up to 550 watts, or 20 amps, whichever comes first.

The balance taps, dont require any board, go straight to the chargers balance connections. As it charges, ANY cell that goes above 3.600v has a 1 to 300ma "brake" or pull down applied via the tap as required.

And if this isnt adequate to hold the cell to 3.600 during charge, the charger lowers the max charge amps, until it absolutely is enough. In an extreme situation, with out of balance cells, this can result in a max charge rate, at the end of charge of just 300ma, As more would sent a cell, or cells, higher than 3.600v.

Thanks. So the balance wires do not bring a low cell up a little - they just take a high cell down a little as I suspected - and I realise the board is just for convenience when swapping lots of packs, I just wondered if there were different types of boards for different brands of packs of the same size for model aircraft etc..

What gauge is the standard cable ie "the thick wires at up to 550 watts, or 20 amps" - I think it would be handy to splice some parallel/series leads for the Andersons and that gauge seems about right - 2 thick marine cables are too big to fit in a single Anderson connector "spade-plug[?]" thing.

LC

[Burgerman]

The balance circuits pull down a high cell only. Or pull down all the cells other than a low one as needed. The extra power goie in via the heavier charge wires.

As for the size of these then the relatively small ones are fine in SHORT lengths. As when charging mopdel heli packs. But I use a short 4 inch long 4mm to anderson connector cable. And then everything is 8 guage from here on. Anderson connectors for parallel, dont need to be as fat since half the current goes down each.

See these cables:

[Burgerman]

http://www.wheelchairdriver.com/faster- ... arging.htm

more...

[Lord Chatterley]

The balance circuits pull down a high cell only. Or pull down all the cells other than a low one as needed. The extra power goie in via the heavier charge wires.

As for the size of these then the relatively small ones are fine in SHORT lengths. As when charging mopdel heli packs. But I use a short 4 inch long 4mm to anderson connector cable. And then everything is 8 guage from here on. Anderson connectors for parallel, dont need to be as fat since half the current goes down each.

See these cables:

Yep, that's exactly what I'm looking for- thanks.

LC

[ex-Gooserider]

Just note that BM's 8g cables are a UK size, actually 10mm^2, which corresponds roughly to a size 7 AWG cable, which doesn't really exist as a size one can get in the US.

An AWG 8 cable is significantly smaller, though it will probably do for most applications. AWG 6 is bigger, but works for those seeking to go in the "overkill" direction. It is also the biggest size that (at least in the US) will fit in an Anderson SB-50 / PP-75 contact.

Not sure what wiring options this gives for the EU folks - though presumably you would be looking for the 10mm^2 size. However it does matter for the US given the difference between whatever system BM's 8g is, and the US standard AWG sizing...

I have also tried a couple of different styles of cable in AWG 6, an "automotive / marine" cable and a "welding" cable. Both were the same size, though the wire stranding was different. The marine cable had a harder "poly" plastic insulation, while the welding cable was more rubbery, but both were quite robust. IMHO the welding cable is slightly better for our applications, as it was more flexible and easier to fit into corners and other tight spaces that we see in our chairs.

ex-Gooserider

[ariveroll]

Has anybody tested this on a Quickie P222?

[ariveroll]

You can do 4 cells to each (7 cell) balance connector.
Or 7 to the first one (a) and 1 to the other one (b).
All can be routed through one D connector on the chair. Keep all balance wires short, or less short and a bit fatter that the short ones...

Like this-




LC

WOW! This diagram really clears things up. In RC terms this as an 8S or 2x 4S battery, correct?
Now, how may inline cells, and which, on each group to get a good range?

What charger?

Thanks

[ex-Gooserider]

The diagram is for an 8S pack - which is what you need for use with a chair, as that works out to comparable voltages to the lead acid batteries the controller was built for. It could be any number of cells in parallel, as parallel cells are usually drawn as just one bigger cell in order to keep the diagram simple (and electrically they are the same) Essentially you want a pack with 8S by as many Parallel as you can stuff in the battery box.

Remember cells in series keep the same Amp-hour capacity, but add their voltages, while cells in parallel keep the same voltage, but add up their Amp-hour capacity. You also need to keep the pack symmetrical - i.e. all the strings must have the same number of cells in them.

Thus you need 8 cells per string in series to get the nominal 24V, and sufficient parallel strings to give you a reasonable Ah capacity, and so that the controller / motors can't draw more peak current than their C-rating.

Ideally you could build a pack like BM's only doing 8S x 9P for 72 cells, but that requires an end-to-end box, which AFAIK is unique to his chairs - if you have the more common side-by-side box, the numbers don't work for this, as the 12Ah Headway cells are 40mm wide with their building blocks, but a Group 24 lead battery is only 300mm long, so the numbers don't quite work out.

In my Redman conversion, I'm looking at going with an 8S x 8P pack, or 64 cells, in an 11 x 6 matrix with 2 cells missing - however this is also a bit of an unusual box as I have the two group 24 batteries standing on end, with their tops facing each other. (I may also be able to go bigger if I have the clearance, something I haven't really measured yet...)

However the basic idea is the same - figure out how many multiples of 8 cells you can get into the available space - the wiring can be worked out afterwards.

For the charger, the first choice here seems to be the Hyperion - which we know should work well, as BM helped develop some of the charging algo's for it, and has been doing a lot of work with the manufacturer on making it work right...

ex-Gooserider

[Burgerman]

http://www.wheelchairdriver.com/lithium ... rchair.htm

Why monoblock batteries dont work for us (yet?)

http://www.wheelchairdriver.com/convert ... ithium.htm

What you need to do, build 2 group 24 replacements...

http://www.wheelchairdriver.com/BM-MK3- ... rchair.htm

How, and why I did it, and why you are wasting about half of the added benefits by sticking to 24v, and so on.

[sin85]

I wonder if this post could be placed at the top (like robotecq, etc)

[ariveroll]

You can do 4 cells to each (7 cell) balance connector.
Or 7 to the first one (a) and 1 to the other one (b).
All can be routed through one D connector on the chair. Keep all balance wires short, or less short and a bit fatter that the short ones...

Like this-




LC

WOW! This diagram really clears things up. In RC terms this as an 8S or 2x 4S battery, correct?
Now, how may inline cells, and which, on each group to get a good range?

What charger?

Thanks

[ariveroll]

Yep.

The thing with lithiums is that they are not really 3.2v or 3.3v Thats a nominal figure. They are generally 3.4 or above for at least half the discharge cycle.

When you charge FOUR cells you do so at 3.600v x 4 = 14.400. Exactly the same as lead batteries. The difference is that they will STAY at 14.4v for a very long time and only drop a fraction, where the lead batt drops instantly to say 13.2 and eventually 12.8v ish.

So a fully charged 4 cell lithium phosphate stays at the 14.40V you chargeD it to.
Some are 3.6v some 3.65v, 3.7v... But the last "bit" doesent do anything. 1/1000 of an amp hour will send the voltage from 3.55 to 4v plus! This is why balancing and care is needed. So almost every charger is 3.60v CC/CV.

As you use the chair, the volts will fall from 3.6 (14.4v) to 3.35 to 3.40 in the first few meters. It will then stay exactly the same and only drop a tenth of a volt - even under load on a hill - for say the next 15 miles. And lose another 1/10th only over the next 15 miles... Then it will drop off a cliff. And fall very fast at about 95 percent depleted. At this point its done. The next half a mile will see the volts drop not by 10ths of a volt but by a volt, and after that its not good for the battery.... But its easy to see/feel this happen as it will fall from 3.2 volts to 30 percent less. And will slow accordingly. So just stop and do not try to drive.

The other thing is that when you come to a hill (do a wheelie, climb a ramp etc) the volts of a lead battery drop like a rock due to the surface charge (peukert) affect. Lithiums do not change. They stay exactly the same.

Why not use 100ah batteries such as these?
http://www.evpst.com/productshow.asp?id=136

Thx