Arkmundi's saga making & maintaining an A123 AMP20 battery

[999zip999]

Iguess I will post here. My 24s as 2-12s A123 20ah packs. I was looking for weak cells and on pack 1 the number one cell neg when high 3.8v and all the rest 3.5v something and found a loose sense wire on it. I put bad on the charger and it looks find so far. Don't know how to find a weak cell. But think after charge I will see. I did take ride with the cell at 3.8v and it level out as my 1420 don't stop it and didn't have a way to take it down. I started charging with the 24s connected and disconnected it after 4 min. How back is it to charge half the pack when connected as 24s and charge half or 12s ?

 

[999zip999]

Yep That cell ran to 3.61v will the rest of the 12s or 11s at 3.48 to 3.51v. So it a runner I have a spare and will match charge and install. Pumping 3 voltfreaks right now so 6a. maybe 4.5amp ? one cell. Will take pics and beg my roomate to download or hit up that beachcrusier guy. It was low tide at sunset and beer in a bag plus. Love this part of town.
P.S. I just though it was rock solid and the phase was loose and kicked.Plus this was why i had a spare.
Arkmundi's hope I didn't jump your thread just letting you know on the balance of full tab cells. Looking to buy some more but how many spares ?
Oh Oh will check with my 4 pin I.R. meter the cheap one.

 

[dnmun]

did you balance the cells together before you assembled the pack? the first thing to do when making a pack is to short all the cells together so they have the identical voltage to begin charging from. then they will be in balance from the beginning.

 

[999zip999]

Yes put 4 voltpheaks on 6 cells 4 times for the 25 cells and let them sit for a day still all the same 3.5v. 2 months ago. This cell is on the bottom of rear rack. I need to pad better for one. I don't like to rest the cells flat as I have with the other 12s.There in the triangle lenght wise.

 

[999zip999]

So sitting and watching the raise in voltage it takes a while to hit 3,4v but after 3.42v the cell fills fast real fast.

 

[arkmundi]

P.S. I just though it was rock solid and the phase was loose and kicked.Plus this was why i had a spare. Arkmundi's hope I didn't jump your thread just letting you know on the balance of full tab cells. Looking to buy some more but how many spares ?

No problemo, that's what we're here for. So, to anwer "how many spares?" I'm advising as many as you can aford. Obviously, the A123 AMP20 salvage market will disappear at some point in time. Johnson Control's now own the manufacturing plant. All the production problems will be worked out, so the plants are spitting out perfect cells. They'll also put in quality assurance controls, so their automotive clients will NEVER ever again run into what Fisker did, forcing a recall. Any less-than spec batteries will be yanked before they leave the factory. There will NEVER be a "factory window" for folks like us. The lower quality of cells we got from Victpower between the first & second carton, less than 4 months, tells me that they're reaching down into the pile and pulling up cells that are third rate at best. Also their drop in price to reflect that. TIME IS RUNNING OUT FOLKS!!! If you want to build a pack, or get spares to protect your pack, now is the time!!

 

[999zip999]

Didn't want to hear that but felt that that is what is going on. I replace the one cell that was running high. Will see on my test ride. I guess I could test that one cell later somehow ?

 

[arkmundi]

Just a quick update, after having replaced 3 of the original cells in my 12S AMP20 pack. The surgery went well, with Agniusm's kit allowing easy reconstruction of the pack. I now believe that the "cause" of cell failure was mostly due to having left my pack connected, through a Watt's Up meter, and on, while at the bike shop. That meant 10 days of continuous discharge. Apparently the Watt's Up meter has enough of a draw to do that. As a result, 3 cells died. They were fully discharged and would not hold a charge thereafter.

Since I only had 2 extra cells from the original buy, I did a Victpower carton split with Barkmau 7/7 from a carton of 14, the minimal shipment. So I now have 6 extra cells. I was uncertain how the new cells would hold up in the pack. But they're doing just fine.

I'd be willing to let some of those cells go if anyone is interested. They're short, cheese-grater tabs but full voltage and holding a charge just fine. Not-used, but obviously salvage. Leave me a message if you're interested.

 

[ohzee]

Nice glad you found the root cause at least.

Any problems threading the punched tab into the kit or bending it over ?

 

[arkmundi]

Nice glad you found the root cause at least.

Any problems threading the punched tab into the kit or bending it over ?

No problems whatsoever. Just have to do a bit of work with thin-nosed pliers to straighten out the tabs so they slide out OK. I also used copious quantities of tape so as to not let the back bar fall out of place. A little foresight and deftness of hand made it all go well.

Agniusm: that may be an upgrade to future kits, if you're going to manufacture more. Instead of making a back bar like you do, make it part of the PCB, so its integral. Both construction & reconstruction would go easier that way.

 

[JoramsWeapon]

Sent PM with interest in leftover Victpower cells.

 

[agniusm]

Agniusm: that may be an upgrade to future kits, if you're going to manufacture more. Instead of making a back bar like you do, make it part of the PCB, so its integral. Both construction & reconstruction would go easier that way.

This was thought over and over again many times. The thing is i cant solder aluminum blocks. There is an option to press in special nuts into PCB but the slots for cell tabs leaves little meat for the press nut. I was wandering if steel is easy to solder? Perhaps getting square nuts instead of blocks would make it work?! Copper is way too soft for this.

 

[arkmundi]

This was thought over and over again many times. The thing is i cant solder aluminum blocks. There is an option to press in special nuts into PCB but the slots for cell tabs leaves little meat for the press nut. I was wandering if steel is easy to solder? Perhaps getting square nuts instead of blocks would make it work?! Copper is way too soft for this.

Yea, you're the engineer here. Glad you've been thinking about it. Aluminium is OK. The bars are OK. Either a way to adhere the back bar to the PCB or place square nuts into the PCB during manufacture. Don't know how you contract the components out, nor want to know. Maybe you PCB supplier has a good idea. Just sayin that if there was improvement I could make, in retrospect, It'd be this. Getting tape down to those back bars without disassembling the whole pack (the cells I replaced were in the middle) was time consuming. In retrospect, I should have taped all the bars onto the board during assembly, and left the tape on. Problem is, even good tape will loose its sticky over time and changing climate, hot & cold. wet & dry. There's an opportunity for you and your fellow engineers to make an improvement. How big a deal is it? Not very. Still love & recommend the kit. Best.

 

[agniusm]

Its not a big deal to make it, quiet different story to make it cheap. All additional work and labor adds to the cost which at the moment is on a high side. Will think about it and if A123 prismatics will be available still, make changes.
Cheers

 

[arkmundi]

The cost of pack

What is the complete cost of your A123 battery pack build?

Good question. Here it is - costs include price, applicable tax and shipping to me, USD:

• A 14 count carton of A123 AMP20 cells from Victpower .... $383.30
• Another 7 cells, spit carton ..........................................$198.15
• Agniusm's 12S kit ..................................................... $120.68
• Smart 6A Charger for 12 cells LiFePO4 Battery Pack ........ $92.26
• Watt's Up RC Meter .................................................. $58.84
• 3 celllogs at $17.95 apiece ............................................ $53.85
• Master Lock 7111D Locking Cash Box ........................... $23.70
• Miscellaneous wiring & hardware ................................ $25.00

Total: $955.78. Prices of course are likely different now - you'd need to check with my vendors. That covers everything, including casing & charging. UPDATED 19Nov13

 

[onlineaddy]

Thanks so much for the detailed accounting.

 

[arkmundi]

Thanks so much for the detailed accounting.

You're welcome. Many here at ES will advocate for a BMS, which would be another expense and additional time for the build. I'm still neutral, saying that for now I'm OK without a BMS, but willing to add one in if I see its necessity. I believe my pack stays reasonably balanced doing just a bulk charge with smartcharger. I'll check cell voltage every week or so. If a cell is wacko out of charge (less than 3.2V), I'll pop on a single cell charger - I've got a voltsphreak 2amp for that. But you should add in the cost of a BMS if you want to be on the safer side of the question.

 

[JoramsWeapon]

What is the complete cost of your A123 battery pack build?

Good question. Here it is - costs include price, applicable tax and shipping to me, USD:

• A 14 count carton of A123 AMP20 cells from Victpower .... $383.30
• Agniusm's 12S kit ..................................................... $120.68
• Smart 6A Charger for 12 cells LiFePO4 Battery Pack ........ $92.26
• Watt's Up RC Meter .................................................. $58.84
• Master Lock 7111D Locking Cash Box ........................... $23.70
• Miscellaneous wiring & hardware ................................ $25.00

Total: $703.78. Prices of course are likely different now - you'd need to check with my vendors. That covers everything, including casing & charging.

Hobbyking's Turnigy Wattmeter does the same thing as the Watt'sUp Meter but at less than half the price, is frequently out of stock so you should put on your wishlist to be notified when its in.

 

[cassschr1]

I feel the cells might be ok with out a bms low amperage charges 2-4 amps but not with even 10amps
when these cells hit 3.5 it goes fast ......especially with more amps. Maybe you canput a timer on your chargers to stop at a guess of how many ahr to put in
This experence comes from to many hours spent doing this building packs, charger is powerlab-8
If you don,t have a rc balancer/charger to initially get cells correct you should at least run celllog 8s
I type this because I've lost 8 cells out of 108 Probably from bulk charging mostly with meanwells
For those that do have a PL8 try the accurate charge for a123 2300 cells set at 10 ch and discharge
I picked up about 3.5ah by cycling packs from 3.25 to full about seven times. Every time it would pick up about .400ma

 

[arkmundi]

found 1200 lumen bike light 12-80V $20.00 posted by by ecat » Mon Jan 14, 2013
... now installed:

Placed on the back rack so light strobes my pedalling - I want to be visible to road beasts more than I need to see road conditions. Soldered in after the Watts UP, before the controller, so its ON whenever I've plugged the battery in. Nice bright blue luminosity pulling all of 0.25 amps - almost nothing.

 

[arkmundi]

Quick update on that CREE LED back-mounted light in blue: it gets more comments than anything. One of the benefits of not being enclosed by 4,000 pounds of steel & glass, on an ebike, is that you're eye-to-eye with pedestrians everywhere, able to interact, and in a friendly walkable city, you do. So what gets noticed is not the fact I'm on an ebike or have a state-of-the-art A123 AMP20 nano-LiFePO4 battery, its that bright blue light. By the way, I used silver duct-tape to add in strategic reflective surfaces, all with the idea of safety in mind, being more visible to those monstrous killing machines I have to share the road with. And geezzzee, am I seen!

 

[ambroseliao]

Sounds good Ark...

D you have a night photo or video of your bike's lights?

 

[arkmundi]

Sounds good Ark...Do you have a night photo or video of your bike's lights?

Yea, here it is:
So there it is doing double duty. I'm at ClarkU promoting an ebike workshop to be held with the local earn-a-bike. I've got the whole thing there for students to see, including the battery. FYI, I got 19 sign-ups for the workshop. Plan is to get them all on this forum site, do the research, and then some group-buys. After which the real in-the-workshop begins with battery making 101, assembly, completion, testing, and maintenance. I'll go over the whole tamale.

 

[agniusm]

This is very good indeed. To get young people involved in technology and fabrication is most of value experience for life. I was thinking myself to start a after class activities for youth from school twice a week but lack of time keeps me from doing it.
Thumbs up, Arkmundi

 

[999zip999]

Love. But be carefull a battery cell is very powerful. So be careful.