Cutoff Voltage

[myelectricbike]

You can also use on-off and on-off-on switches in a matrix of resistors to provide the cutoffs you'll need.

[mustangman]

  If you design it and build it, I would buy it!

[movintoelectric]

Thank you all(myelectricbike, pdonahue, oneeye, and mustangman) for your input.  When there is time I will add a resistor to lower the cut-off voltage and then monitor battery voltage with a voltmeter as I ride. The whole world is waiting on good batteries at a reasonable price and efficient solar cells at a reasonable price to charge them.  Who ever can get a good handle on either of those will be the next Bill Gates.

[mn_engr]

I don't know if anyone is reading this string yet, but I can share some info on my e-bike project.

I have a 36V front hub motor kit.  I live 12 miles from work. 

I bought 120 NiMH AA cells for my battery pack.  (2600mHR cells by T-energy ~$1.20 each.)
you can find these cells at www.all-battery.com.

They are arranged as 4 sets of 30 cells in series, housed in eight ~2' lengths of 1/2 Sched 40 PVC from menards.
 
I use B360B schottky diodes to tie the 4 sets together to make one 36V 10.4 AHr pack.

In the beginning I wasn't able to reverse engineer the charger and limit the charge current.  As a result, I used resistors in series to limit the current.  (messy)  I also switched to 28 cells in a string (112 total batteries) because the pack voltage was too high for the charger.  I now have the charger modified to 1A current limit (250mA in each string) and got rid of the resistors.

With the 28 cell configuration, due to cuttoff voltage, I was not able to deliver more than 5A to the controller 
I also discovered R71 and change it so cuttof should be 29V, but haven't tested it.

I am now in phase 2.  I'm making a 'smart pack' of 120 cells.  I monitor each pair of cells (60 voltages) using 8 MSP430F2012 uP's.  (each uP has 8 a/d inputs)  If any cell goes to 0V, I can signal the controller to stop draining power by connecting the brake input to the 430's.  (Ultimately I want to be able to switch out the string and run on the other 3 until they go as well.

If anyone is considering alternatives to Lead Acid, I'd consider NiMH cylindrical cells. 
If you're willing to work with AA's (lowest cost)  I paid ~$140 for 36V @ ~10AHrs.  (This included 60 AAA's @ 1AHr as a bonus) and free shipping!
all-battery also offers D cells (10AHr) You could get 30 for ~$225, which isn't  but probably the best deal you can get.  This would simplify the design (only 2 MSP430's required!) but I don't know if there is a common PVC size that would fit the D cell.

I will keep posting on how well they hold up, but 500 cycles should be realistic, especially if my 430 protection design works.

[myelectricbike]

I use NiMH AA cells for things like mice and cameras and portable power for home made electronics projects, flashlights and all the rest so further investment in the technology is a possibility for me.

My only complaint so far is that of the 10 or so 4 battery per packs I recently purchased 4 have only held up for 20 to 30 cycles at the most and neither Wal-Mart or Target will take open battery packs or light bulbs that pop on the first day back.

[mn_engr]

Here is a datasheet of a NiMH cell very similar to the ones I use:

http://www.all-battery.com/datasheet/2300mah.pdf

It gives a nice curve on capacity vs cycles if you properly charge/discharge the cell.

Perhaps the reason NiMH didn't work out is that your devices are expecting alkaline voltages (1.5V) and NiMH is 1.3V at light currents and 1.2V when discharging at C.

Recharging can also play a big role.  This website has been very helpful in understanding batteries:

http://www.powerstream.com/
Click on Technical Resources, then "How to charge NiMH batteries"

If you want to see what I am trying to do with my MSP430 voltage monitoring,

http://www.powerstream.com/
Click on Technical Resources, then "Special Design Rules for Large Battery packs"

The device pictured at the bottom of this page is very similar to what I am planning to make for my e-bike.  With active monitoring like this, the probability of getting 500 cycles should be rather high.

[myelectricbike]

Since the stated cycle life of NiCad is double NiMH why do you prefer NiMH, if cycle life is of concern?

[mn_engr]

I would have gone with NiCd if I could have found a source for cylindrical cells at a good price. 

With my NiMH selection from all-battery: $1.20/cell where each cell is 1.2V@2.6AHrs, thats $ 0.384 /watt*hour. 
If I could find a NiCd battery that would beat that, I'd switch. (They are lighter to boot!   BTW, my pack weighs 10.5lbs with PVC)

Of course this is with an unproven vendor (T-energy) so I've risked $140, and
all I have is a datasheet for a 'similar battery' that tells me I should still have 80% of capacity after 500 cycles. 
Time will tell if I made the right choice.

One cool thing about my 'multiple strings of monitored cells' approach is that if later on I decide I want to add new cells to the pack, or even cells of a different chemistry, I can just switch out dead strings made up of old cells as they die early and let the newer cells keep trucking, or biking in this case.


Have you found any good sources for cylindrical NiCd's?

[myelectricbike]

Its been several months since I did my last search. To be honest my doctor has order me off the ebike until I loose 20 lbs. I never considered that gaining weight might be an ebike hazard. I've tried to stay with lead acid and pedal a lot using my regen controller but to be honest again I am a pitifully lazy creature. 

Actually, I can not recall where I found the best price with the greatest possibility of quality assurance but the total price was well above the price of either the Golden or the PowerStream packs and Golden may no longer sell the Ni Cads. http://www.goldenmotor.com/batterypack.htm

If you do a search on this site there have been a fair amount of posts suggesting the cheapest source of Nickle based battery technology may be from used Toyota Prius packs recovered from disabled vehicles.

Otherwise its the LiFePo4 technology everyone here seems to be eying. If you check the batteryspace.com site and search for LiFePo4 you will find a lot of information but no cells with more than 2 amp hour capacity whereas cells with 10 to 30 amp hour capacity are needed unless you design a circuit to cut in new strings when the current string is discharged.

Also the forums.batteryspace.com has a moderator that will answer most any tech or sales related question except something like asking how fragile lead acid batteries are.

If I do a price search in the next few days I will post it here just in case I find something with better quality assurance if not lower price.

[ahend]

I bought 30 T-Energy F Size 14000mAh NiMH cells from All-Battery. They seem to have some of the best prices. My bill was $335, not dirt cheap, but it’s a good cell and packs a hefty punch. I broke my pack up into three 12 volt sections which get charged simultaneously with separate peak chargers. The power plug is then jumpered to connect them in series. The pack is fixed into a trapezoid shape 10.5” long by 7” wide by 2.25” thick.

mn_engr
Did you solder your batteries end to end or do you rely on a compression contact inside the tubes?
I like your micro control scenario. I bet it looks like an atomic device when you are done. I’m working on a datalogger, based on a basic stamp, to monitor the three 12 volt packs. It’s not as sophisticated but should help me identify problems before too much damage occurs. I could easily rig a similar 1 volt/cell/pack cut off circuit, it’s a great idea.

Andy

[myelectricbike]

30 T-Energy F Size 14000mAh NiMH cells from All-Battery are now $403.50 so that must have been a seasonal thing. 25 are listed as $335, however. If the only and best requirement for battery management is a circuit to monitor each battery and cutout at 1 volt then I would use the no tab batteries with a 2 layer piece of circuit board between each battery to facilitate both monitoring and switching. That way if you have a few extra cells than needed and one goes down you can replace it instantly without shutting the whole pack down.

[OneEye]

Nickel prices have been doing wierd things over the past few years, driving the cost of Ni based batteries higher.  Recently the metal cost started falling though.  Hard to say whether that will translate to Ni battery prices falling or not.

[myelectricbike]

If its anything like coffee the higher the highest price the higher the recovery price will be. We were at $5.89 at th end of February, $9.19 all Spring and half of the Summer and now at $7.33.

[mustangman]

 That is why we need the LiFePO4 technology to level the playing field on pricing.

[myelectricbike]

Humm... don't suppose the battery industry might be raising the price of Nickle based technology so they can maximize the volume starting price of LiFePo5...?