Eric's Project #002

[xyster]

But its looking like these batteries are only good for about 1C or so.

You can find more similar discharge charts and dozens of individual descriptions of 1s emoli packs pulling 45 amps for the RC folks (30 amps each cell with minimal sag):
http://www.google.com/search?hl=en&q=emoli+site%3Arcgroups.com&btnG=Search

Not to mention their inclusion in powertool packs where they certainly have to pull more than 1C (probably around 10C) for the life of their 5 year warranty. I agree with Fechter that something is probably amiss with your testing procedure. Could there still be epoxy or other residue on the cell caps? Are you measuring the voltage right on those caps?

Xter, did you measure the amps with a multi-meter or did you compute them with math? I know you're using four 10 ohm resistors together, which should produce 2.5 ohms, but maybe you're getting more resistance than you think?

For individual cell testing, I computed the amperage. But for the whole pack, the amperage is communicated by the ammeter on my handlebars. Sag varies a bit with temperature, but averages about 6 volts at 35 amps for the whole pack, which jives pretty well with my individual cell measurements.

 

[Beagle123]

Update: Charging Manually Led To Better Results

I made a make-shift charger, and charged three new batteries manually. I was getting about a 0.2 voltage drop at 3.4 amps. I was very happy with the improvement. Then I realized that my connections to the batter may not be the best. I was pressing aligator clips to the terminals of the batteries for my test. To make a better connection I clipped the aligator clips to an aluminum bar, and pressed the aluminum bar against the batteries. My results all improved by about 0.05 volts.

Here are my final results:

Test done at 3.4 amps (1.13C)


Battery #1 = 4.17v (starting), 4.05v under load, Diff = 0.12v!
Battery #2 = 4.19v, 4.02v, 0.17v
Batt #3 = 4.20v, 4.06v, 0.16v

Much better. I'm going to repeat this test on my batteries X,Y and Z above.

You guys were right. Again. I'm just relieived that I'm not sitting on $1400 of crap batteries. Whew.

 

[D-Man]

Was there a reason you didn't use the matching milwaukee chargers to go with the milwaukee batteries? Seems like with such an investment, you would want the matching chargers to do the charging. I know jondoh and maytag used them on their milwaukee powered e-bikes.

 

[dirty_d]

good ol' lead-acid for me.

 

[fechter]

Actually, I think all your batteries are complete toast.

You can send all of them to me and I will properly recycle them.
(into my scooter )

 

[Beagle123]

Was there a reason you didn't use the matching milwaukee chargers

Yes, I didn't have room to fit all the plastic packs on the bike, and I was planning to use about 14 of them so unplugging and charging them all would be a burden. Actually, it sounds good compared to what I'm going through now.

I'd prefer to get off lead-acid becuase part of my motivation is to save the environment.

 

[Beagle123]

The NIghtmare Continues...

It turns out that about 7 of my chargers are fried. I don't know how it happened. When I plug them in with no battery attached, the green light doesn't go on. When I plug them into a battery pack, the green light goes on, but when you plug it in, the green light stays on and it doesn't charge.

Another overheats badly, but works. I have about 3 that function properly.

Sometimes when I charge a few batteries it goes for hours and hours and never gets above 4 volts. Then I put the same batteries into another and they charge ok. Everything is haywire.

Does anyone know a better way to charge these things?

 

[Beagle123]

Another Question

I'm considering how I'm going to manage charging my batteries. My latest idea is to get DC chargers that run on 12v DC an install them into the neck of my bike. Then I could have an external power supply like this:

http://cgi.ebay.com/Power-Adapter-12-Volt-5-Amp-12V-5A-DC-Supply-LCD_W0QQitemZ320166549977QQihZ011QQcategoryZ58286QQcmdZViewItem

I think this is a good idea for these reasons:

1) The heat created from 15 chargers in the confined space of the neck will overheat. I beleive that the most heat must be created from converting the AC to DC. Do you guys agree?

2) That power-supply isn't a problem to use. Its just a bit bigger than an extension cord that I'd have to connect anyway.

3) The DC chargers should be smaller without the AC transformer, so I could fit 15 of them in the neck.

Here are the drawbacks to this design:

1) It doesn't exist. As far as I can tell the chargers that I have could be the only 4.2v, 1+ amp, single cell units out there. And considering that I have 7 dead ones on my patio table, I'm not sure I want to spend another $250 for a new set. There are ones that will automatically charge 1-3 cells etc, but then they're bigger. I'd like to find the little, simple charging component.

I'm going to keep looking.

 

[Beagle123]

One Good Thing Happened Today

I got my cut-off switch today, and I think its a great thing. Its a big switch with a removable key. The terminals look like they can handle lots of current. If you guys need a way to disconnect your controller, and make a key for your electrical system, you should look at this switch. It solves these problems for me:

1) My controller has no "off" switch.
2) My bike doesn't have a key to secure it.
3) The positive leads from the battery need to be connected to the motor and the controller, so I can wire the positive lead to one end of this switch, and attach the motor and controller to the other terminal, making a solid junction.

And it only costs $6.00

You can get them here:

http://www.harborfreight.com/cpi/ctaf/displayitem.taf?Itemnumber=92688


<img src="http://www.harborfreight.com/cpi/photos/92600-92699/92688.gif">

 

[xyster]

The NIghtmare Continues...

It turns out that about 7 of my chargers are fried. I don't know how it happened. When I plug them in with no battery attached, the green light doesn't go on. When I plug them into a battery pack, the green light goes on, but when you plug it in, the green light stays on and it doesn't charge.

I'd send them back. 20 out of 20 of these chargers have worked fine for me for a year now.

 

[fechter]

I think this is a good idea for these reasons:

1) The heat created from 15 chargers in the confined space of the neck will overheat. I beleive that the most heat must be created from converting the AC to DC. Do you guys agree?

2) That power-supply isn't a problem to use. Its just a bit bigger than an extension cord that I'd have to connect anyway.

3) The DC chargers should be smaller without the AC transformer, so I could fit 15 of them in the neck.

Certainly part of the heat will be from dropping the voltage, but depending on the charger design, there could still be quite a bit of heat generated by a low voltage charger.

Seems like there should be some way to use the guts out of a Milwaukee charger.

 

[Beagle123]

Hi Fetcher:

The milwaukee chargers won't do me much good at this point. I'm using 6p, 14s configuration.

 

[newbie electric rider]

Nice job Beagle, that is a one of a kind bike for sure. Sorry to hear about your battery problems, but i am sure you will solve them very soon with excellent results.

 

[Beagle123]

Thanks Newbie. I'm currently considering my options for charging the batteries. My problem is that I can't find individual cell chargers that can charge my packs at 1-2 amps. The chargers I tried before would have been perfect if they worked. I spent $250 on those little bastards. I'm weary of spending another $250 to replace them, especially when I can buy a great power supply like this:

http://bigerc.com/product_info.php?products_id=170&osCsid=5da05f9cc745fd79cb967165559de752

I have another idea of how to configure my system. I thought the ideal solution would be to take all the charging leads from my 4v sub packs and connect them together and charge them with a big 10 amp, 4.2v charger. However, I realized that it is impossible because connecting the charging leads would short-circuit the system (duh).

But then I thought of a solution:
I could run big 10 guage wires from each sub-pack and connect them in parallel to charge the batteries, and connect them in series to run the bike. The only problem is that this involves unplugging 14 connections every time I want to charge (or ride) the bike.

I think this can be accomplished by making a grid of plugs (possibly with andersen connectors). It would be 14 X 2. Then you make a "dummy plug" that simply connects each battery to the one next to it (series). The end wires would go to the controller. This plug is for riding the bike.

Then, I could make another plug that fits into the same place for the charger. This plug would just connect all the positives together (and negatives).

The benefits of this system is that I don't have to mess around with a bunch of plugs. But the biggest benefit might be safety--it would be impossible to short circuit becausse I'd have to disconnect the "riding plug" to plug in the charging plug.

This diagram shows the configuration. It shows a 3 battery pack (12v) system. Just imagine 14 batteries:

 

[fechter]

Your plug idea will work, but with 14 batteries, that's going to be one gigantic block of Andersons.

Another possibility would be to get a surplus power supply that can be adjusted to the proper finishing voltage for the whole pack. It would run in a CC mode until the batteries get up to voltage, then switch to a CV mode. The only problem with this approach would be maintaining balance on the cells.

From what I've gathered from some users is that balancing during charging may be unnecessary with these cells. Some kind of BMS would typically be used with this setup, but if the cells are well matched to start with, it may not be needed.

Any idea why your individual charges failed?

 

[JEB]

Hi Beagle

Maybe I am missing something but- why are you chargeing the li batteries in parrell instead of series parrell? it would be a lot easyier to check the statis of each bank- or with the terminals (wired out) of each cell out to a check point to check each cells voltage.

jeb

 

[JEB]

What I use to check the capacity of a single cell- in this case nmh- from Clockkit, the electric basic single cell clock turns off at about .8vdc, one could use the same system to check a li-ion cell using a zener diode in series with the clock-battery set to the cut off voltage you want to drop out at. (stop clock) To reset just turn the hands back to 12.

High wattage auto lamps give the flattest current draw for minmal cost (lots of them in parallel) power resistors work fine but don't give as true watt hour rateing. But you could bring the li cell below the recomended discharge voltage if you are not around to see the clock stop.

 

[dirty_d]

do you have any lead acid batteries laying around, you could use those until you get your pack made, winter is close.

 

[Beagle123]

Another possibility would be to get a surplus power supply that can be adjusted to the proper finishing voltage for the whole pack.

I was considering that. I founf this one on ebay:

http://cgi.ebay.com/NEW-MASTECH-REGULATED-LINEAR-DC-POWER-SUPPLY-0-60V-0-3A_W0QQitemZ120172112166QQihZ002QQcategoryZ58286QQssPageNameZWDVWQQrdZ1QQcmdZViewItem

It looks great, and would be the easiest solution. However, I'm worried about safety when charging at 60v. I understand that the only real safety concern is overcharging. If I have 14 cells in series, and aa bunch of them fail, there's the potential for there to be higher voltages across single cells. I don't feel so confident that I can detect and solve these problems, so I feel like it would be a bit more risky. Hopefully, if there were problems with some cells, it would be obvious, and I could intercept the problem (start voltage and load testing).

Another downside is that I wouldn't ever know how balanced the batteries are. Right now I don't understand how to balance 14 batteries. I've seen balancers for up to 6 cells with the option of linking them for more cells, but that sounds like a more complicated (complicated=risky) plan. Also, its another potenial for failure of all sorts.

The method that I feel most comfortable with is xter's method of using a 4.2v charger so there isn't even the potential of overcharge. This seems idiotproof. When you think of the possibilities for failure, there are very few. When charging, the single cells would be connected directly to the charger with no electronics inbetween. It would automatically balance itself. Also, I can observe that the charge is going well from the display on the power supply. I could observe problems with current flowing into the batteries.

More later (I think you'll like my next idea)

Any idea why your individual charges failed?

No, I don't. Possibilities are:

1) My house may not have the best current. I've had some other electronics fail in this hous. My computer fried, and the guy said it may be due to power spikes. I need to check for that. Any ideas on that?

2) The epoxy I used added resistance. However, I don't beleive that should have caused any problem. I beleive that would have made the chargers just turn off sooner. I think several of them burned out as soon as I plugged them in. I think I remember many going green immediately.

I really don't know.

JEB, thanks for your replies. I don't know what clockkit is. I googled it and can't find it. I think the above answers your other questions.

 

[Beagle123]

New Variation on Battery Idea

When I was thinking about making my last battery configuration that uses the "dummy plug," I realized that it would be very difficult to make a block of andersen connectors with each plug connecting to its neighbor. THe problem is that 10 guage wire won't bend in such a sharp curve.

So I had a better idea: Make a block of andersens connectors for each half of the batteries so when the two plugs are connected, the batteries are in series. When you unplug them, you can plug a charging plug into each one.

Like this:

 

[JEB]

http://www.klockkit.com -
clock parts-
clock movements-
quartz time movements-
premium quartz movments-
under $5

I was reading about li batteries, to charge them correctly it appears to require more than a constant current, and a peak voltage turn off (there are some chargeing graphs showing correct method, on the internet)

what I was thinking was haveing cells, 7 in series, 7 in series, power relay [=50+vdc] (all electronics 40 amp 12 volt power relay @.200ma)
or (digikey 40 amp schottky diode 40cpq100)
parallel the 10 banks together, the diodes isolate the banks from each other, or use the relays, keep the relays off to charge.

I have never tried the diode method, but it was suggested on the internet- but it sounds like it is a good idea- worth a try- schottky's have only .5v drop, at 10 amps draw that is only a 5 watt loss.

purchace 7 cell chargers, charge the 7 cells (http://www.batteryspace.com) with one charger on each group of the 7. It is easy to check that you have a full charge on each cell when they are not in parrell.

the tab end cells, look like the are easier to connect together, wrapping them around a wire (tinned copper) and soldering- there is less heating than soldering directly onto the cell end.

 

[JEB]

some power strips have spike protectors- for computers, and other uses.

 

[Beagle123]

Hi JEB:

Using diodes to isolate the batteries is interesting, however, I don't feel confident enough in my electronics abilities to be certain I'm not going to make a mess out of it.

If I undersstand you correctly, you're suggesting having two 28v packs separated by a diode so it wouldn't short-out when you charged it. It seems to me that charging at 56 volts using a power supply is close to the same thing.

I don't see the advantafge to making the sub-packs in series. If cells go bad, they could slow down the whole pack.

 

[JEB]

I talked to a electronic eng. friend yesterday about li battteries, he said there are still a lot of problems with the charging-discharging of the cells- to the point they are conserned that they can explode- "no tesla car for sale yet" he suggested to stick with nmh batteries they don't have the problems of the li cells, even to the point that he would put up with the extra weight- (watts per lb)

back to your li pack- what I am suggesting is to put a total of 14 cells in series with a diode, which in turn that pack is parrelled with the others of the same configeration to form a bank. .

charging with (2) 7 cell chargers center tapping (/= charging points) the middle of the pack. *

-/xxxxxxxx/xxxxxxx/-)- + out this is one pack

[-)- = the diode] the loss of the one cell, or more (in a pack), it/they can't be reverse charged. The problem could be that if the packs are not balanced to each other, all of the current would be drawn from the highest pack until it went down to the level of the others. there is no problem using the relays- (direct connection) but then again cells can be reverse charged.

With what ever set up you go with One has to keep an eye on the voltage of the total battery bank as to not discharge it too deeply to start reverse charging cells

"To get carried away", because with the diode in series, one pack can not talk to the others in the bank- and not even when you are drawing current out (one way valve, as it were)- you can run a volt meter across each cell to see if there is a problem, to locate the defective one.

Checking before the diode on each pack you can see if they all have the same voltage after charging- or in the discharged state-

Placing a power relay in where the diode would go, in it's open state there is no problem charging the pack as above. * (no large power connectors to unplug) -talk about being long winded!-

Does anyone else have any comments?

 

[JEB]

the more I read about li-ion batteries--- http://www.buchmann.cal/chapt4-page8.asp I will stick with lead acid gell cells- use them up -throw them away- live with the weight- and bulk-

buchmann states li-ion have a 2 to 3 year life wheather you use them or not- charging/dischargeing conditions are too stringnent for me- less than 1% charge discharge voltage parameters- too high distroyed- too low 10% loss of capacity, deep discharge = failure.

dewalt in their 36vdc batt pack has a control ic, switches, fuse, thermistor, +lots of resistors and capacators to make the whole thing work! all mainly used to balance the charging current to each cell, to give the longest life and most capacity per charge.

also http://www.electronics-lab.com/li_ion_reconstruct/ one note-(don't solder directly onto the cell.)

dewalt spent a lot of $$$ to perfect their control system in their battery pack that the ordinary person could not afford to duplicate. (one system in each pack - or group of cells in series, NOT IN PARALLEL!)