Valid Pack Building Opinions Sought.

[Penny K]

OK, so I'm at a stalemate with this.
Any opinions as to keeping the connections to Minimum 4 connections on each series will be appreciated.

 

[spinningmagnets]

spot-welding? Soldering?

What model of spot-welder, or soldering iron?

There is a current discussion over the various benefits/drawbacks of every material and method. If you are committed to spot-welding nickel-strips onto the cell-tips for the parallel connection, fine...

If so, I would like to suggest that you can use copper for the series connections, since it is the series connections that must carry the full pack current. If you agre that this is a viable option, the copper can be attached onto the spaces between the cells on the nickel strips.

just a thought...

 

[Penny K]

spot-welding? Soldering?

What model of spot-welder, or soldering iron?

SUNKCO 709A with 8mm .010" pure nickel strips.
Improvement on the series connection pattern? I would like a min of 4, I have 3x 3 on the back side that I cant figure out.

The other consideration I had was pure nickel sheet and do a full trace of each S/P connection.
201 Ni .010" is cheap for a 6x12" sheet, I need 6x14", so 2 sheets for me.
http://www.knifemaking.com/product-p/pn010.htm?CartID=1
1 problem that I cant overcome with this method, my Prof wants the pack submitted so the cells can be individually balanced and replaced.

 

[Penny K]

Ummm, 2x 14S3P might do the trick here.
Running 6 batteries through 3 connections, copper bus lines or not, almost make the most sense, at 2:18am my time.

 

[kdog]

Once they're welded into p groups you can't individually balance or replace cells only the group.
Is this the layout you've been given or can you change it?
If it's fixed then you can improve a couple of things.
Ditch one of the parallel strips on each group and use that nickel to put in the missing series connects. There are a few missing You only need one P strip per end. It'll save cost in nickel.
For the problem groups you can either rearrange ( preferred) or run copper connects in wire form between the un-connected cells. I'd insulate them well and secure them over their P group to avoid at all cost an abrasion and short.

 

[kdog]

What discharge rate are you looking at?

 

[Penny K]

Once they're welded into p groups you can't individually balance or replace cells only the group.
Is this the layout you've been given or can you change it?
If it's fixed then you can improve a couple of things.
Ditch one of the parallel strips on each group and use that nickel to put in the missing series connects. There are a few missing You only need one P strip per end. It'll save cost in nickel.
For the problem groups you can either rearrange ( preferred) or run copper connects in wire form between the unconnected cells. I'd insulate them well and secure them over their P group to avoid at all cost an abrasion and short.

I decided that 2x 14s3P will prevent the bottlenecks I was going to have with the original layout. Trying to push 6 cells through 3 connections...not good.
I'm just thinking it through now, any suggestions as to a really effective layout would be appreciated.
I want the original '+' and '-' to reane in the respective Left and Right if at all possible....But whatever works.

Im using a 14S 30Amp BMS, so there is the limitation as to what the pack will actually put out, but I'm after the 52 volts for the dive.
The original design was for 14S10P, but that wasn't possible, so we might have to build another 14*4 and locate it in a different spot.

 

[kdog]

Unfortunately .1 ni has poor current carrying ability. At 6mm wide you'd be looking a 3 maybe 3.5 amps per strip. So you'll need 8-10 series connects to run 30amps for anything more than a few seconds. If you can't weld any thicker, then you have a few options. Go the broad sheet approach (.5A/mm), order 0.1x 10mm and still have 5-6 S connects, or double/triple up the 6mm ( only the S connects)
Have you tried reconfiguring before going the split pack option? Two series strings paralleled will need two Bms's.

 

[Penny K]

Unfortunately .1 ni has poor current carrying ability.
Two series strings paralleled will need two Bms's.

(.010"/.255mm) x 8mm Ni strips, I recalculated the scale.
You might be familiar with this as its posted here.

As for the circuit design, there is no difference in the output voltage to the motor or the BMS connection points.

As for the amperage, the BMS I have to use is rated @ 30A.
Dividing the Amp output to 2 banks will split the total single draw in half, so in theory there should only be 15A draw on each bank.
Each cell is 10A constant/20A burst to 30 seconds.
Keeping 3 strips per group will allow, at max, 42 ACCEPTABLE Amps.
At 15A it should be well under the 28.2A OPTIMAL rating for the Ni strips, or 53% of calculated capacity.
Unless I'm wrong, please provide some math to disquantify mine.

I have 2Lbs/1Kg of Ni. No shortage here or reason to skimp out.
This is where I am with this:

 

[jonescg]

Looks solid to me!

 

[Penny K]

Looks solid to me!

TY.....
The cells are 3.6V, so the packs center charge is 50.4V.
I'm only charging to 80%, and every 10th charge doing 100% (59.4V).
At 100% I will run the pack down to 40.5V (2.88V) then charge to 80% for the next 9 charges again.
Is this the best way to condition the cells? Or is charging to 90% better for 9 and 100% for 1?

 

[jonescg]

I'd say charge to 100% right before a ride, but if you aren't going to go riding for a while, it's best to keep store it at about 3.9 V (60-80%).
As long as they are in balance (same cell voltage) they are considered 'run in'.

 

[Penny K]

I'd say charge to 100% right before a ride, but if you aren't going to go riding for a while, it's best to keep store it at about 3.9 V (60-80%).
As long as they are in balance (same cell voltage) they are considered 'run in'.

I will be riding every day during summer here.
I'm using new TESLA cells from a 90D. TESLA preconditions/run in the cells before they are assembled into a module. What that is exactly and how much no one knows....The cells have an estimated 2000 cycles lifespan. Charging 6x a week, from March to October (35 weeks) is 210 chargers.....So out of the lifespan of the cells, that's 9.52 years.
Really, should I be concerned?

 

[jonescg]

No, I wouldn't be too concerned Most of us upgrade to new/better/different cells before they start to show signs of loss. Or the bike gets stolen before then... However all cells have a calendar life which is shortened by prolonged high states of charge and high temperatures, so my best advice is to ride the shit out of it and beat the clock.

Are you in Vancouver or further east? The humidity in Vancouver can creep in and corrode the connections, so try and waterproof the pack as best you can.

 

[Penny K]

No, I wouldn't be too concerned the bike gets stolen before then...
Are you in Vancouver or further east?

Noooooo! No stolen! Big chain and lock!
East in Edmonton.

 

[kdog]

My metric mistake! I thought you were using 0.1mm nickel... Which is pretty crap.
.25mm is much more solid- looks wider than 6mm too, 8mm in your chart?