PSI -- LiFePo4 -- Testing.. 1 ... 2... Testing..

[rkosiorek]

the price on the G10 from that hobby shop is a lot lower than the price i got from Digikey. i wish i had known that a couple of weeks ago.

the Digikey part number is PC8-UNCLAD-ND for a 1/16 X 12 X18 piece. price 23.99USD.

another nice thing with the PSI buy is that the copper strips and connection stuff was included. i had to source my own for the LIfeBatt cells.

visually the BMS from PSI looks identical to the "VMS" sold by LifeBatt. Even the board arrangement is the same. 1 board for every 4 cells daisy chained together with a single communication board with the 2X DB9 serial connectors. even the board number matches what LifeBatt has on thier documentation.

the square chip with lots of legs on each of the 4 cell boards looks like it is a controller of some kind. it also looks like it still has a part number on it. any chance you could haul out a magnifying glass and read the part numbers off of it?

i'll be interested in how this all works out.

rick

 

[marty]

Ypedal,
If you stack batteries as shown in this picture, you will fit more in a smaller space.

 

[Dee Jay]

Original post Thursday... this latest post Sunday evening... humph... I'm kinda disappointed they aren't built already!?! Can't wait to see the final product and here about the performance!!

Yeah Ypedal, let's go. Get Cracking on them cells! Chop Chop! We ain't getting any younger here. We don't need no petty excuses, we need some battery pr0n here, let's go! *Looking at watch*

:wink: Just kidding, man. Don't hit me on the face, don't do me ugly (than I already am) ! We're just excited for ya.Thanks for sharing details and great pics!

Yeah.. i know.. but i have not seen the surface of my kitchen table in months, my house has turned into a horrible mess to the point i have forbiden myself from bolting these cells together until i clear a path and organize myself a bit.. ( should have that done tomorrow ! )

Totally with you on re-organzing and regrouping . It's the way I prefer to work. On *a single project*, I regroup about 3 times or more depending of the task at hand.

Don't forget to have fun tinkering!

J

 

[rkosiorek]

Next step, to build these babies in packs.

I may change my mind, yet again, but right now i'm looking at building these cells in 24v 10ah each. 8 Cells.

Gary built his packs using " G10 " fiberglass board :
Random google result :

http://www.jamestowndistributors.com/userportal/show_product.do?pid=8758&engine=yahoo!6457&keyword=g10_fiberglass

Or lexan/acrylic/polycarbonate

http://stores.ebay.com/affordableplastic_W0QQfrsrcZ1QQfsubZ4802319QQtZkm

They even got colored !! i like the red!

I see the ebay guy has tubes !!!

http://stores.ebay.com/affordableplastic_Acrylic-Tubing_W0QQcolZ4QQdirZ1QQfsubZ4802321QQftidZ2QQtZkm

why not try a Canadian source for the plastic. I have bought from a company called Johnston Plastics. they have offices across Canada but unfortunately none in the Maritimes.

Toronto 1-800-268-2220, Montreal 1-800-363-8742, Winnipeg 1-800-665-7455, Edmonton 1- 800-661-5658

rather than Acrylic (which is brittle and cracks easy) i would recommend either PVC or Lexan. you can get both in clear. PVC has the added benefit that it is easy to heat and bend.

you can set your oven to about 275F or 130C and heat the sheet for a 5 or 10 minutes to heat it through. it will turn flexible and rubbery. you can then mold it. once it cools it will be set into shape. you can even stretch it a bit. it machines very easily and can even take a thread.

rick

 

[GGoodrum]

Those are definitely the same LiFeBatt VMS boards. They have balancing shunt circuits, but are limited to 100mA per channel. Each board has serial ports that can be daisy-chained, and I believe there is a PC program that can be used to get info from each cell. In addition, each board has "under-voltage" and "over-voltage" TTL signals that are all ganged together. These signals are logically high, when active, but with a couple of parts, the low/under-voltage signal could be reversed so that it can be used to pull down the controller ebrake signal, like we do with our BMS board.

It is nice that they supply it with the copper bars. You could contact my friend Andre, at http://www.askmanap.com, if you want G10 plates done. He's still got pattern he did for me. He gets the material at a much better price than I've seen advertised so you could end up paying less, and have everything custom fitted.

-- Gary

 

[Ypedal]

Doh.. my shipment of 12x24 pieces shipped yesterday from performance hobbies, i got one of every sheet they had so i can test different thicknesses for flex etc.

However, Once i figure out the exact size of everything i want, i may contact him for the production run if he's equipped to cut the stuff.. i imagine that " fiberglass " sheet must be a PITA to cut ie: itchy.. :x

 

[GGoodrum]

Doh.. my shipment of 12x24 pieces shipped yesterday from performance hobbies, i got one of every sheet they had so i can test different thicknesses for flex etc.

However, Once i figure out the exact size of everything i want, i may contact him for the production run if he's equipped to cut the stuff.. i imagine that " fiberglass " sheet must be a PITA to cut ie: itchy.. :x

He's got four CNC machines, all setup to cut G10 and carbon fiber sheets. It's all automatic.

I tried various thicknesses as well, and ended up using 3/32", as it provided better rigidity.

One thing you need to be careful of, and that is to not overtighten the nuts, especially on the positive (red...) end. The studs can snap off. You need to use a pair of nuts to hold the copper strip, and not just use the nut that is provided.

 

[Ypedal]

One nut ( already on the cell ends ) stays on, aparently important to prevent the cell from leaking, i have the spec for the torque in an email ( 8 something.. will confirm in a bit ) then the copper tab, G10, Lock washer, Nut.

Thanks for the heads up on the threaded bolts ease of breaking gulp.. that would suck.. Do you know if this threaded rob is part of the cell of if it can be removed and replaced ? I have to ask PSI this ( longer or shorter studs might be handy at times )

I have a table saw, plunge router, drill press, sloor model jig saw, most have vaccum ports and a dual breather 3M safety mask lol..

 

[rkosiorek]

Wood Working Tools? bet you buy most of your stuff at Home Depot or Lowe's.

best thing to cut G10 is a shear and the only thing to use for the holes is a punch. these also work great for sheat metal parts like the Nickle or Copper strips used for interconnection.

View attachment n_200.jpg

available for under $200 at places that sell metal working tools.

here is a picture of my shop QC inspector.

View attachment n_201.jpg

note the "critical" lok he has about him.

rick

 

[AndyH]

here is a picture of my shop QC inspector.
<snip>
note the "critical" lok he has about him.

rick

Rick - that's a rough looking inspector. Not only has he a critical look, that 3-legged balance leaves a back leg to kick someone in the head when they screw up.

 

[miro13car]

Ypedal,
Is there any way to figure out production date, any date from those bar codes on PSI cells?
Any other way to figure this out.
I am close to deciding to open my LifeBatt battery, but I am still testing before opening case.
MC

 

[Ypedal]

IT BEGINS !!!!

Micro: I asked the question twice so far, and Raymond either does not know, or is not spilling the beans yet.. i'll keep bugging..

So i find myself on Sunday, end of my vacation, but i've cleared out my schedule, i had a list of things 20ft long that HAD to get done before i could allow myself to dive head first into this project.. and i'm happy to report.. i've knocked it down to a 4ft list and it's good enough !

Triage - Went thru all the boxes and found 8 cells at 3.328v and 362 to 363 grams, and another 8 at 3.329v with the same exact weight.

I have no clue if this makes any difference or not, but it's my starting point. Logic says that they have all been cycled before shipping and have been idle the same amount of time, having cells grouped in equal self discharge by voltage would help the BMS in not having to ballance the packs as much over time.

 

[Ypedal]

All bolted up temporarily, 10 awg discharge wires with 45 amp powerpoles.

 

[Ypedal]

Bright in here.. 8)

 

[Ypedal]

 

[miro13car]

Ypedal,
I am really interested in your testing of PSI cells really, because they are in my LifeBatt pack.
For now from my WatssUp reading I can conclude that cell voltage do realy drops below 3V in my HPS pack when I draw 47-48 Amps.
At 48A WattsUp records 35V UNLESS WatssUp doesn't capture momentary voltages right.
It is hard to use regular voltmeter , because 48A would last only like 2-3 seconds, because TForce controller works this way.
I would probably won't be able to capture on multimeter.
But as I remember there are multmeters which can capture miliseconds - lasting values
not mine.
MC

 

[miro13car]

your test can be much meaninful to us electric bikers than Sandia Labs tests for sure if you can get like 50A drawn from them and still maintain 3V/cell.
MC

 

[AndyH]

...For now from my WatssUp reading I can conclude that cell voltage do realy drops below 3V in my HPS pack when I draw 47-48 Amps.
At 48A WattsUp records 35V UNLESS WatssUp doesn't capture momentary voltages right...

MC - how do these track with your experience? This is single cell performance at 45A and 50A. Andy

 

[Ypedal]

Hmm.. 35v would mean 2.1 to 2.2v per cell and that just can't be right... not at 50 amps anyways.. either the amp draw is way higher or the voltage drop is not being read correctly somehow.

I have 20, 35, 40, 50 amp with 4110 fet controllers and a big monster on the way with 80 amp aprox ( still being worked on ) , and if that don't do it i have a 200 amp kelly brushed controller to really push things ( That 20C rating on the lifebatt sticker bothers me.. but i won't get into that just yet.. )

I don't know about you guys, but i just HAD to see what's under the shrink, since these cells showed up at my door this has been eating at my patience and tonight i decided i had to scratch the itch..

 

[Ypedal]

Oh, hey Andy !! nice graph .. i really have to invest in one of those pronto..

Skirt comes off !

 

[Ypedal]

Under the end stickers ( red / black ) you can see bulges where the Vents are i assume, i don't want to break the seal just yet so for now what's under there is gonna be a mystery..

 

[Jozzer]

Is that a lifebatt you are testing Andy? THe pic says its a Yesa....

 

[Ypedal]

Lifebatt, BMI, Yesa, Ecosomething, and more, = PSI cells ! Phoenix Silicon Inernational

http://mvp090-1.104web.com.tw/cetacean/front/bin/home.phtml

 

[miro13car]

Ypedal,
I have 36V HPS pack that's 12 cells. 35V divided by 12 is 2.9V, just below3V/cell, but below.
Or WattsUp is not accurate.
Jozzer says he can stay above 3V/cell at 5c DRAW.
MC

 

[Ypedal]

36v.. not 48.. wooops.. sorry about that.

3.0v or 2.9v is not a great difference when the motor meets the road, .. depending on how the HPS pack is built, if the discharge goes thru the BMS/VMS or not can make a difference compared to Jozzer who may be reading the cell voltage directly....

Testing takes time, and rest assured all my waking hours are spent working on it ! :wink: