I am having a problem and I need some help...

S

SlyCayer

100 W
Joined
Sep 6, 2010
Messages
173
Location
Limoges, Ontario
I am working on a E-Kart, it's a racing Go-Kart converted to a Electric Kart.

My setup is the following: Alltrax Controller(7245) and a D & D Motor(ES-15-6).

Batteries: 20P24S A123 26650 taken from Dewalt's A123 outputting 79.2V nominal and should be able to take a total draw amperage of 1400amps.

My problems:

My batteries are connected in parallel with 1/8" X 3/4" steel flat bar and springs carefully screwed to the flat bars and aligned to the batteries. Springs are rated at 100V 40A and I should only be pulling a max of 20amps per cell anyways.

Pictures of the strings and batteries: http://img42.imageshack.us/g/picture144c.jpg/

Problem 1: When I am pulling about 150amps, which should be nothing for 20 cells(7.5A per cells) my voltage sags to 58V(2.41V per string).

What am I doing wrong here? I need your help badly.

P.S: Whoever helps me find my problem will receive a donation.

Thank you and Greetings, Sly
 
You must forget about springs and steel at all and change it to a solid good old copper bars or car battery cables or 4 GA speaker cables or whatever copper of proper size you can find in your town for carrying enough current or more. For immediate contact between bars and each cell you must weld them with at least 8GA wires or something like this. You'll need a contact welder, industrial or home made... The bottom line is forget the steel springs at all cost if you making the battery, not the expensive lithium air heater. That's all I know...
 
http://endless-sphere.com/forums/viewtopic.php?f=14&t=17912&p=260888&hilit=a123+easy+pack#p260888

Some good ideas in Leamcorp`s thread. Copper is a much better conductor than steel.
 
Hi guys,

I tried soldering 8GA wire to the batteries before but I had to get the cells so hot to get a correct bond to the button or the flat side that it was ridiculous.

Please help me find a solution.

The Leamcorp's method I don't understand it, how does he get connections between each cell and the copper bars?

Is the reason for my extensive sagging have to do with Steel and string resistance?
 
I see that you want 1400 amps. Thats a lot of amps. I dont think that Leamcorps style will work for 1400 amps. Maybe you will see less sag if you use copper bars in place of the steel bars you are using now. Good luck
 
Steel is around ten times worse as a conductor than copper, so steel bars will drop around ten times the voltage for a given current than copper bars.

An added problem with steel is that any oxide on the surface (even a very thin, invisible layer) will decrease conductivity even more. Copper oxide is a reasonably good conductor, so copper bars don't really suffer badly from corrosion reducing conductivity.

If you're serious about this, then the best way to get that pack built would be to get it welded up. You can build a resistance tab welder fairly easily, see the thread in the Ebikes Technical section.

Jeremy
 
Icewrench said:
I see that you want 1400 amps. Thats a lot of amps. I dont think that Leamcorps style will work for 1400 amps. Maybe you will see less sag if you use copper bars in place of the steel bars you are using now. Good luck
Click to expand...

P.S: I am only drawing 460 amps max, which is about 20A per cell. 1400A is the maximum draw of my battery pack.
 
If you put a small load on the pack you can probe around with a voltmeter to see where the losses are, for example between the physical end of a battery and the bus bar where it connects. For a 1 amp load, 1 millivolt = 1 milliohm resistance. My copper straps screwed to Headway cells measure 2-5 milliohms, but a loose connection can raise that to 100 milliohms. The voltage drop across the length of a bus bar should also be interesting.

Edit-Oops, make that 0.2-0.5, milliohms, checking the logbook I had made those measurements at 10 amps.
 
Well, 80 1 ohm 1 watt resistors in series, 40 2 ohm 2 watt, etc. A ~100 watt AC light bulb or heater like a hot plate (they will draw less with lower voltage DC, don't use a transformer load like a soldering gun, that might overheat the transformer). Several 1500 watt AC heaters in series (the kind that goes on/off with a mechanical switch or thermostat, DC will confuse ACpower electronics).

Simpler might be to make the measurements with your normal charging current, if your charger puts out clean enough DC.
 
SlyCayer said:
Really? So a sag of 10V with steel bars would be only 1V with copper bars? And the Steel bars had started to rust before I screwed the springs down...
Click to expand...

Yes, really............... well, almost. The exact figures are somewhat better than my guess:

Copper resistivity is 1.68 x 10-8 ohms-metre (0.0000000168 ohms-metre)

Iron resistivity is 1.0 x 10-7 ohms-metre (0.0000001 ohms-metre)

The real ratio in resistivity between these two metals is about 6, so clean steel will be around 6 times worse as a conductor (i.e. will have about 6 times the voltage drop of a bit of copper of the same size). Add in a bit of surface corrosion and I reckon you could easily have a voltage drop that's around 10 times or more of copper bars.

As has been suggested, probing various points with a meter whilst drawing some current, or even charging, would give you confirmation of where the voltage drop is occurring.

Jeremy
 
One thing about a123s is that they do naturally sag under heavy amp draw. When we used them for R/C helicopters you had to over gear for the headspeed that you wanted after the initial spool up. a123 cells are kinda like "torque" where as lipo's are like "horsepower". a123s drop voltage and then hold there with a super flat discharge rate where as lipo's hold a higher voltage for longer.

If you need 70 something volts under load then make the pack about 20% higher voltage than what you want. You will thank me for it in the end. The nominal voltage you should use under load is 2.7 that is what worked best for my high current drain on R/C helicopters pulling 50 to 100 amps for a 12s1p pack. Your poor 3C results are probably results from your battery build.

I also totally agree on the spring/steel idea being really bad. The best way is the spot welder and next would be battery bars and a large 100 watt plus soldering iron with a large tip. Mine is a 175 watt Weller with a 5/8 diameter tip. You need the large tip to quickly add heat to solder fast and then quickly remove the heat. I can solder cells in about 1 second. https://webvia.techni-tool.com/VIA/viaImagePageIndex.jsp?row=0&pgName=viaListProducts.jsp&searchText=272IR272&modifier=SEARCH&reqTitle=TITLE_VIASEARCHRESULT&newWindow=Y

Tom
 
Hi, checking voltage drops are various points is hard because the setup is 20P12S. I have 20 Cells connected in Parallel and then I have those 12 strings connected in Series to create 39.6V Nominal. 43.2V After charge.

I have another question, after charging my cells, they are at 3.6V, but a few hours later they go down to 3.5V and then the next morning they are back at nominal which is 3.3V. Is this normal? After a charge should voltage be 3.6V until there is an actual load being put on them?
 
Checking for losses should be just as easy with that set up (provided you can get meter probes to the cells) as any other. What you want to do is set up a known fixed load of maybe 10 amps or so (the exact value isn't important as long as it's steady during the measurements), using light bulbs or a load resistor, then, with the meter ground connected to the load negative (directly, right at the load) record the voltage on the negative bus at the end of the cell furthest from the load connection, followed by the voltage on that cell's negative terminal, then the cell's positive terminal, then the bus next to the positive terminal and finally the voltage right at the load.

The resulting set of measurements will tell us exactly where your system is letting you down and help us give an answer that has a reasonable chance of helping you fix the problem.

Jeremy
 
SlyCayer said:
You are losing me here. I can connect a AC Hot Plate with A123S?
Click to expand...


yes, use a resistance heater type load. do not use a space heater with a fan, just the oil type or ones with a wire wound element.

for a 36V lifepo4 pack, a 700W electric heater element draws 1.74A.

put several heaters together in parallel. plug them all into the end of the extension cord and connect the other end to your pack, through an ammeter for measuring the current, then the voltage delta will show the resistance of your conductors.

2 of the 1500W oil bath type heaters in parallel for the load and the current would be close to 10A.
 
Maybe someone could start a thread on dummy loads. I would but am not sure where to put it.
Here is my 10 ohm resistor, a 1.5 kilowatt space heater from the dump. I added a small analog 0-5 amp meter for a rough indication of the current.
 

Attachments

  • ten_ohm_load.jpg
    107.4 KB · Views: 1,496
dnmun said:
2 of the 1500W oil bath type heaters in parallel for the load and the current would be close to 10A.
Click to expand...

I have 1 wok that as a 120AC - 1550 WATT MAX and a 1 Lean Mean Fat Grilling Machine(George Foreman) 120AC - 1440 WATT MAX would these do the job for my 39.6V nominal battery pack?

So you are saying to connect these 2 "heaters" in parallel(+ and + together and - and - together)? Do I need a 10 ohm resistor?

I would like to Thank you for helping me, I am fairly new to this EV stuff and I am learning a lot. I thought my Electric course in Science in high school and my 2 years in Computer Technology would of learned me this but it doesn't' seem so...
 
If you need to solder 8 or 4 GA wire to the terminal rings or the anderson contacts, use the heat gun. Point the gun up and put between your knees provided you're sitting. Drop the anderson contact in the slot in the tip of the gun. Turn the gun on and when you see the tint of the contact rapidly changes fill it up to the half with solder which almost immidiately melts and liquefies, then just stick tightly in the stripped end of the wire in it and turn the gun off. Now you can remove cooled down wire soldered very well this way. Don't try it on batteries though. Tab or spot welder is the only thing that works the way it just should, beside perhaps the ultrasound one. Also protect your eyes with some eyewear and be extremely cautious since the heat gun is almost 2KW thing blowing up almost 500 degrees Celsius :mrgreen:
 
ebkr said:
If you need to solder 8 or 4 GA wire to the terminal rings or the anderson contacts, use the heat gun. Point the gun up and put between your knees provided you're sitting. Drop the anderson contact in the slot in the tip of the gun. Turn the gun on and when you see the tint of the contact rapidly changes fill it up to the half with solder which almost immidiately melts and liquefies, then just stick tightly in the stripped end of the wire in it and turn the gun off. Now you can remove cooled down wire soldered very well this way. Don't try it on batteries though. Tab or spot welder is the only thing that works the way it just should, beside perhaps the ultrasound one. Also protect your eyes with some eyewear and be extremely cautious since the heat gun is almost 2KW thing blowing up almost 500 degrees Celsius :mrgreen:
Click to expand...

Thanks a lot, any help on my previous post? On Dummy loads?
 
if you plug the two heaters into the end of the extension cord, they will be in parallel. this makes their already low resistance even lower so that it will draw the maximum current out of the pack.

the pack will drain fast though so you need to measure the resulting voltage drops quickly. like all at once essentially so be prepared.

oh, and use the male end of the extension cord to connect to your battery in series with the ammeter. use the 10A range on your voltmeter. make sure to put the probe in the right place, you have to move the red one over to where the hole says 10A. don't use the probe tips for your contactor or it will burn the tips off.

make sense?
 
I have a clamp meter...

Can I simply hook up the heaters to my batteries and check the voltage at different location like Jeremy was suggesting?

I can clamp my meter to one wire and check the amps and then proceed to checking voltage from different location on in the battery pack.

Also, I am wondering if Aluminum can be a substitute for copper? Let me know.
 
You must log in or register to reply here.

Similar threads

G
donor hub motor needed to help a stroke mechanical engineer
Replies
1
Views
445
calab
C
D
Build a bike to commute not sure what I need.
Replies
13
Views
493
E-HP
E-HP
P
Lost noob, where can I read up on this stuff?
Replies
1
Views
195
E-HP
E-HP
ERN
I think I may have fried my mind.... help!
Replies
17
Views
1,172
DomW
DomW
P
Need help with a build
2
Replies
27
Views
1,434
powerpeddles
P
Back
Top