Voltage multiplier needed

[dnix71]

If you string cells in series to boost voltage the weakest cell limits the string and the total usable capacity of the string. When charging a long string the same limitations apply. Incorrect charging will damage the weaker cells and ruin the pack.

I have some used nicad packs from Bruce. I have no simple way to make a well balanced 24v pack since I can't I can't know the true capacity of each cell, and in this case there will be 100+ cells when I'm done.

Is there a circuit that would take low voltage, say 1.2 or 2.4 and step it up to 26v and store it in a super cap to be drawn from by a device.

That would allow me to build a mass low volt pack. The weakest cells would fail first, but as long as they are not shorting across they would not limit the rest or take down the whole array.

Some inefficiency in conversion is fine, since this will be compensated by increased use of all the cells and longer pack life.

[madlabs]

Dinx71,

It sounds, if I am understanding, that you wnat to make a large low voltage battery pack and then run a higher voltage load. If this is true, it seems as if what you want is a DC to DC boost converter. You don't mention how much current you need, but Digi-key for example has plenty of converters. If you can provide more info on the load, (and I've gotten what you are trying to do) I'm sure we can help identify a unit that will meet your needs.

Jonathan

[OperaHouse]

Charge up the cells, individually load with a 10 ohm resistor, and see how long it takes to discharge to say 1 volt.

A nice job for a computer or you could just use a LM339 operating as a comparator that powers a cheap 1.5quartz clock.  Count the minutes by resetting to 12:00 each time.  The cell even powers the clock.  A diode can be the voltage reference.

[dnix71]

Madlabs, that's what I want to do. 16v supercaps are cheap and easy to get because they are used in car stereos. If I had two 1.2v to 14v upconverters, I could charge 2 supercaps each to about 14v and then series them for output. The output would go parallel with the existing 24v sla pack I already have. The supercaps are intended as a low impedance storage buffer.

My ebike officially has a 450 watt motor but if I full throttle it the battery lights on the handlebar control drop way off so I suspect the real full load is nowhere near that.

Even at 450 watts, thats 20 amps or less, so if the boosters' output is in series I would need two 20 amp/14v controllers for full throttle assuming no help from sla pack.

Does that sound doable? There is a circuit called a Joule Thief, but it's not intended to produce that kind of power.

[joestue]

let me get this straight, you want to parallel 100 cells and convert 200 amps at 2.4 volts into 14 or 26 volts?
not going to happen...

with 6 cells in parallel x 18 in series you really don't care if one cell dies, they are pretty reliable ya know.. but in any case, 18 comparators will let you know if one group dies.

for a while hybrid cars were running 200 single cells in series, and the reliability was *good enough* 5 or more in parallel is plenty good, provided you aren't mixing and matching manufacturers.

[willib]

Madlabs, that's what I want to do. 16v supercaps are cheap and easy to get because they are used in car stereos. If I had two 1.2v to 14v upconverters, I could charge 2 supercaps each to about 14v and then series them for output. The output would go parallel with the existing 24v sla pack I already have. The supercaps are intended as a low impedance storage buffer.

My ebike officially has a 450 watt motor but if I full throttle it the battery lights on the handlebar control drop way off so I suspect the real full load is nowhere near that.

Even at 450 watts, thats 20 amps or less, so if the boosters' output is in series I would need two 20 amp/14v controllers for full throttle assuming no help from sla pack.

Does that sound doable? There is a circuit called a Joule Thief, but it's not intended to produce that kind of power.

The converter you are looking for is called a boost converter.
It uses pwm a FET and a coil  to slingshot the voltage to a higher value.
" oztules " had sucess using an old PC powersupply controller (PWM)chip
http://focus.ti.com/lit/ds/symlink/tl494.pdf
so did " xyzscience " IIRC
i had limited sucess though, cuz i just couldnt get the coil right.

[madlabs]

Dinx71,

That is a lot of current. I haven't seen one that will go from so low an input voltage up to that output voltage at that current. I'm off to work, but I'll poke around later and look, but I have doubts about finding something premade of that description. And homebrewing a high current boost like that won't be easy.

Jonathan

[Fused]

Hi
Ive been playing around with nicd's too.
I built mine so I can isolate a single 11 battery group cell if one happens
to go bad. Built a few 66 battery "packs" to run at 12v range.
Can charge complete pack to 15vdc let it rest full charged, drops to 13.9v.

It costs the extra wires but its handy at 7.8Ah.

My soldering is sloppy at best but here's a picture of what I put
together.





have fun

Fused

[dnix71]

Fused: Maybe taking 1.2v to 6.5v at 20 amps and stacking 4 controllers with supercaps would be more reasonable.

[commanda]

Dnix,

Motors are rated at continuous shaft output.
My 180 watt Cyclone draws 680 watts peak.
My 1500 watt EVT scooter draws 6000 watts peak.
So typically, your peak power input is 3 to 4 times the rated output power.
So your 450 watt motor probably draws 1500 watts peak.

I know my figures are correct, because I've got Watts-Up meters on both vehicles.

So, your peak battery current is probably in the order of 60 Amps.

Amanda

[dnix71]

Amanda- I don't have a WattsUp meter, but I probably should get one. Does it have non volatile memory or does it forget like the KillAWatt meter when it loses power?

The price seems to still be around $55 delivered no matter where you get one.

My motor can't draw that much power even if it wanted to. I don't know if it's the controller, wiring or both that limit things. There is a fuse in the battery pack and I have never blown one. The fuse is 40 amp fast blow and only intended for a dead short condition. I've full throttled from a standing start (not recommended) and nothing failed.
The lights on the handlebar controller drop off scale, though. Even going fairly easy with a full charge the indicator lights drop under moderate throttle and immediately recover on coasting. When they don't come back up to green I know the pack is getting low. The setup is properly "battery assist" rather than battery driven. There is a switch on the controller to allow you to choose, but the use manual doesn't recommend going battery driven.

Dead shorting a 12v 12ah sla usually gives no more than 25 amps. Those are the same sla's used in APC backups under 500 watts. Dead shorting a big pack of new nicads on the other hand might give a bigger current rush.

I really don't want that much raw power into the motor anyway. I would rather have the range since it's about 12 miles to work. Even with 2 packs, if I don't want to pedal to work that's more than 1/2 a charge. I work in a warehouse on my feet all day pick and pack so I don't want or need to pedal much and my feet don't like it at the end of the day, either. I can recharge the pack at work without a problem. I have a set of no-flat inner tubes on order for the bike, due next Wed. It would be nice to have a bullet-proof setup soon since the rainy season is finally over.

[joestue]

Dead shorting a 12v 12ah sla usually gives no more than 25 amps. Those are the same sla's used in APC backups under 500 watts. Dead shorting a big pack of new nicads on the other hand might give a bigger current rush.

!!!

Shorting a lead acid battery should deliver at least 20 times the amp hour rating, for example, a new 7 amp hour battery will deliver at _least_ 200 amps.. and yes, i've measured this.
it sounds like the current delivered is dependant on the resistance of your amp meter..
let us know how quickly a standard 3/32" dia screw driver melts in when you short out a 24 amp hour, 12 volt battery, if it doesn't, the battery is either not fully charged, or its old.

you can download data sheets from respectable manufactures that list this, in most cases its between 30 and 40 times the amp hour rating, anything less means your battery is on its way out. Using walmart as an example , a new 50$ car battery can deliver 500-600 amps at 9 volts... its much less than 30 amp hour battery.
This isn't a case of what type of battery is it, but the simple fact that all lead acid batteries are more or less the same.

Nicd's and nimh cells have much lower internal resistances, and there are D cell batteries out there that can deliver 100 amps at 1 volt, for a nominal 1.2 volt cell.

[commanda]

A watts-up meter is worth every penny, even if the memory is volatile. You can power it off a separate battery if you wanted to.
Turnigy make a similar meter, cheaper, check out hobbyking.
http://hobbyking.com/hobbyking/store/uh_viewItem.asp?idProduct=10080

Going by the voltage sag you describe, I'd say the main reason you haven't blown the fuse is simply because the batteries aren't up to the task.
Of course, you wouldn't have that problem with a decent Lithium pack.

Here's my notes for my cycle with the 180 watt Cyclone last time I took it out.

Sunday     22 August 2010   6.0 Km Chopper. M2 cycleway to Kingsgrove rd & return
   2.176 AHr
   50.2 WHr
   666 Wpk
   30.89 Apk
   21.39 Vmin

This was on 24 volts, 9AH, SLA.
I've now fitted 24 volts, 20AH LiFePO4, and am hoping, weather permitting, to get my first decent ride tomorrow.

Amanda

[dnix71]

I just ordered 2 of the Turnigy meters. Whats the fix for the display?  Everyone says it's unreadable unless you crack the case and add a resistor.

[klsmurf]

Mine reads fine.

[commanda]

I've seen a number of people on another forum using them. No mention of display problems.

Amanda