Creating my own battery pack

[cwah]

Hello there,

I've always been using HK lipo and A123 20AH until now, but they are not the most lightweight. Even the zippy compact at 167wh/kg.

I'm now thinking to replace my pack with the Panasonic NCR18650PD:
http://www.aliexpress.com/item/180pcs-lot-10A-High-Discharge-Rate-Battery-for-Panasonic-NCR18650PD-18650-2900mAh/759290109.html

They are at 230wh/kg and have more than 500 cycles. With a 20s5P pack, I can get 1000WH pack for 5kg.

I'm thinking to get this spot welding to do the job:
http://www.aliexpress.com/item/wholesale-POWER-787A-MCU-Spot-Welder-Battery-Welder-AC-180V-240V-Applicable-Notebook-and-Phone-Battery/898890556.html

What do you think?

 

[imorton]

Hmmmm, looks interesting.

I just noticed that the Spot Welder is 220v, so I don't know if that is an issue for you?

IAN.

 

[arkmundi]

Why would you move from A123 AMP20's for something else?

 

[bowlofsalad]

Energy density is a really big deal in the EV world, we all would like it to be radically higher. Relatively speaking, our current batteries are very dismal in energy density in my opinion. While it would be great to have better energy density, there are always pros and cons. How is the safety of these cells compared to lifepo4?

 

[spinningmagnets]

Making your own pack should prove to be a fun and educational process. But, as part of doing your homework before spending a lot of money I suggest you read both of these threads:

"Samsung INR18650-20R .. yes these cells can punch"
http://endless-sphere.com/forums/viewtopic.php?f=14&t=48074

"14S/6P Triangle Pack build using Samsung INR18650-20R cells"
http://endless-sphere.com/forums/viewtopic.php?f=14&t=50538

Bear in mind, the vendor that is promoted here (ebay: supowerbattery111) will do the spot-welding of the cell groups for you, and they will use any cell you are willing to purchase (should you still prefer the Panasonic 29PD).

 

[cwah]

Hmmmm, looks interesting.

I just noticed that the Spot Welder is 220v, so I don't know if that is an issue for you?

IAN.

i live in europe so 220v AC is good for me

 

[cwah]

Why would you move from A123 AMP20's for something else?

Because I'll divide by 2 the weight of my battery pack

 

[cwah]

Making your own pack should prove to be a fun and educational process. But, as part of doing your homework before spending a lot of money I suggest you read both of these threads:

"Samsung INR18650-20R .. yes these cells can punch"
http://endless-sphere.com/forums/viewtopic.php?f=14&t=48074

"14S/6P Triangle Pack build using Samsung INR18650-20R cells"
http://endless-sphere.com/forums/viewtopic.php?f=14&t=50538

Bear in mind, the vendor that is promoted here (ebay: supowerbattery111) will do the spot-welding of the cell groups for you, and they will use any cell you are willing to purchase (should you still prefer the Panasonic 29PD).

Thanks I'll look at it carefully. I'm thinking maybe I shall invest in one of these spot welding machine so I can make and repair my packs easily

 

[arkmundi]

Because I'll divide by 2 the weight of my battery pack

What kind of weight are we currently talking about, and how critical is that weight?

 

[cwah]

Because I'll divide by 2 the weight of my battery pack

What kind of weight are we currently talking about, and how critical is that weight?

A123 20AH have 130wh/kg. It also needs special compression box to be protected. My current 17s weight about 11kg with box and bms.

Now the panasonic cells have 230wh/kg. Because it's cylindrical and quite safe I can use a lighter box. I expect about 6kg finished for a 20s5p

 

[cal3thousand]

Because I'll divide by 2 the weight of my battery pack

What kind of weight are we currently talking about, and how critical is that weight?

Ark...

We understand your advocacy for the safety of the A123 cells and LiFePO4 in general (And I totally agree with them), but there are reasons for using these other chemistries.

I think it has to do with the way people prioritize battery characteristics. ALL cell chemistries come with their trade-offs, with each person weighing each trade-off differently. The scary part is not fully understanding the "danger" trade-off and underpriortizing it. But for those fully aware of the differences, the safety trade-off sometimes is not worth the penalty of the weight and size and energy density (for their particular application).

Having said that, I see these NMC 18650 cells as a smart trade-off in power and energy density and relative safety (to LiCo). If OEMs feel that these cells are safe enough as a consumer product, I think that says quite a lot. And since no OEMs (that I know off) are shipping the high performance/ high danger LiCo (LiPo) pouch packs, these things deserve more "hobbyist" attention. I see too many people wanting to jump to LiCo because they want the power/energy density, but don't FULLY understand their safety trade-off and the work required to keep them safe. Many see LiCo as the only other option, since these NMC cells have not had enough exposure and they don't know about them.

So as hobbyists begin to report back successful builds (such as Orlando: migueralliart), this NMC chemistry should provide a relatively safer alternative to LiCo, for other (would-be) hobbyists looking for more than what A123 and the like can deliver.

Personally, I want to start looking at these cells more closely and have a desire know more about them. I have just completed my transition to the "LiPo Camp" by selling off the last of my LiFePO4-specific equipment. But I will try a build with these NMC cells soon and may do a complete move-over again, but this time it will be easier since the nominal voltages between NMC and LiCo are closer (3.6v and 3.7v; vs ~3.2v) and my chargers will be able to be tuned for that. OR it may be that I decide to keep the LiPo and reserve it for my "danger" bikes

 

[arkmundi]

A123 20AH have 130wh/kg. It also needs special compression box to be protected. My current 17s weight about 11kg with box and bms.

Now the panasonic cells have 230wh/kg. Because it's cylindrical and quite safe I can use a lighter box. I expect about 6kg finished for a 20s5p

So you'll loose 5kg (11 lbs). I understand in your calculation that its significant weight loss for you. If I were to also want weight loss, I'd first aim to take it out of my fat ass! Seeing as hows I built an ebike to move that ass around to begin with, it kinda defeat the whole purpose!

 

[Toorbough ULL-Zeveigh]

if you're not in a hurry, i'd suggest build your own using good caps like the 1f rockford fosgate. if you don't want to build one, then maybe try one of the china spot welders in the 300-500 range. the cheap $100 ones in the black box are pretty sketchy.

more details

 

[cwah]

So you'll loose 5kg (11 lbs). I understand in your calculation that its significant weight loss for you. If I were to also want weight loss, I'd first aim to take it out of my fat ass! Seeing as hows I built an ebike to move that ass around to begin with, it kinda defeat the whole purpose!

If weight wasn't important, I'd still get the Panasonic NCR18650PD and double the battery capacity over A123 20AH. I'm actually planning to get 2 packs so when I need more capacity I can bring these 2 together in my bike bag

 

[cwah]

if you're not in a hurry, i'd suggest build your own using good caps like the 1f rockford fosgate. if you don't want to build one, then maybe try one of the china spot welders in the 300-500 range. the cheap $100 ones in the black box are pretty sketchy.

more details

Can you help on a correct spot welder? I'd like not to pay too much for that

 

[arkmundi]

Can you help on a correct spot welder? I'd like not to pay too much for that

I'll follow your saga, as someone with an A123 AMP20 pack, moving in a different direction. We should all remain flexible and make the best pack we can in any moment in time, as its a VERY dynamic, competitive market. Have you considered conductive epoxy? Don't know why more folks don't consider that as an option to either spot welding or soldering.

 

[deVries]

So you'll loose 5kg (11 lbs). I understand in your calculation that its significant weight loss for you. If I were to also want weight loss, I'd first aim to take it out of my fat ass! Seeing as hows I built an ebike to move that ass around to begin with, it kinda defeat the whole purpose!

If weight wasn't important, I'd still get the Panasonic NCR18650PD and double the battery capacity over A123 20AH. I'm actually planning to get 2 packs so when I need more capacity I can bring these 2 together in my bike bag

You could salvage your A123 cells that are good to make a hybrid pack. Use the Panasonic to boost the A123. The booster pack would kick-in once the voltage dropped to its lower level. You protect the lower voltage pack w/schottky.

But, I guess you don't want that added A123 weight? Are you going to trash the A123 or ???

 

[cwah]

I'll follow your saga, as someone with an A123 AMP20 pack, moving in a different direction. We should all remain flexible and make the best pack we can in any moment in time, as its a VERY dynamic, competitive market. Have you considered conductive epoxy? Don't know why more folks don't consider that as an option to either spot welding or soldering.

I've never heard of conductive epoxy. Can you link me to few exemple?

 

[cwah]

You could salvage your A123 cells that are good to make a hybrid pack. Use the Panasonic to boost the A123. The booster pack would kick-in once the voltage dropped to its lower level. You protect the lower voltage pack w/schottky.

But, I guess you don't want that added A123 weight? Are you going to trash the A123 or ???

No these Panasonic are rated for 3C. With a 74V13AH pack I only use 1000W constant. So about 1C. No need to use A123 to boost anything.

I'll keep the A123 until these Panasonic are proven reliable.

 

[arkmundi]

I've never heard of conductive epoxy. Can you link me to few exemple?

MG Chemicals 8331 Conductive Epoxy, at Amazon

 

[cwah]

Ok I see.

I found the answer:

Solder has a resistivity of about 0.00000016 ohms per inch, so if conductive glue is around 30 ohms per inch then it's about 187,500,000 times poorer as a conductor than solder.
Jeremy

http://www.endless-sphere.com/forums/viewtopic.php?f=2&t=9395

So no conductive epoxy for building battery pack. I need a reliable and cheap spot welding machine

 

[arkmundi]

Ok I see.

I found the answer:

Solder has a resistivity of about 0.00000016 ohms per inch, so if conductive glue is around 30 ohms per inch then it's about 187,500,000 times poorer as a conductor than solder.
Jeremy

http://www.endless-sphere.com/forums/viewtopic.php?f=2&t=9395

So no conductive epoxy for building battery pack. I need a reliable and cheap spot welding machine

You can do whatever you want of course, but I think you've jumped to a conclusion there.
At: http://www.altex.com/MG-Chemicals-Silver-Conductive-Epoxy-14g-035-oz-8331-14G-P143200.aspx
Air cured (at room temp. for 5 hours) 4.0 - 6.0 ohms
Which is perfectly acceptable for electronics. It'd be a micro-layer between cell tabs and connectors, spread over the surface. There's a lot of usage in the electronics industry. But, whatever....

 

[cal3thousand]

Ok I see.

I found the answer:

Solder has a resistivity of about 0.00000016 ohms per inch, so if conductive glue is around 30 ohms per inch then it's about 187,500,000 times poorer as a conductor than solder.
Jeremy

http://www.endless-sphere.com/forums/viewtopic.php?f=2&t=9395

So no conductive epoxy for building battery pack. I need a reliable and cheap spot welding machine

You can do whatever you want of course, but I think you've jumped to a conclusion there.
At: http://www.altex.com/MG-Chemicals-Silver-Conductive-Epoxy-14g-035-oz-8331-14G-P143200.aspx
Air cured (at room temp. for 5 hours) 4.0 - 6.0 ohms
Which is perfectly acceptable for electronics. It'd be a micro-layer between cell tabs and connectors, spread over the surface. There's a lot of usage in the electronics industry. But, whatever....

Is that 4-6 ohms per joint? At 20S, that would be a range of 80 - 120 Ohms, no?

 

[deVries]

Ok I see.

I found the answer:

Solder has a resistivity of about 0.00000016 ohms per inch, so if conductive glue is around 30 ohms per inch then it's about 187,500,000 times poorer as a conductor than solder.
Jeremy

http://www.endless-sphere.com/forums/viewtopic.php?f=2&t=9395

So no conductive epoxy for building battery pack. I need a reliable and cheap spot welding machine

You can do whatever you want of course, but I think you've jumped to a conclusion there.
At: http://www.altex.com/MG-Chemicals-Silver-Conductive-Epoxy-14g-035-oz-8331-14G-P143200.aspx
Air cured (at room temp. for 5 hours) 4.0 - 6.0 ohms
Which is perfectly acceptable for electronics. It'd be a micro-layer between cell tabs and connectors, spread over the surface. There's a lot of usage in the electronics industry. But, whatever....

Is that 4-6 ohms per joint? At 20S, that would be a range of 80 - 120 Ohms, no?

Never build a battery using that epoxy crap! It would be a hot heater or a meltdown!

Use solder or spot welding (or pressure spring techniques for lower amps).

 

[arkmundi]

Nonsense! The only reliable evidence of usage is actual usage. Its used a lot in the electronics field these days. I'm sure there are some shit brands out there, but not the one I linked to, which has a silver metal embed. The accurate way to determine resistance, if that's a concern, is to actually measure it with a multimeter. Since it would be a micro-thin layer, there would I expect be very little resistance measured. There would be a film of silver particles and epoxy particles, so many, many channels, electrical and adhesive. When I get around to building a 18650 cell pack, I'm going to experiment around with a design that yields both the electrical and strength properties I desire, so will report MY actual experience. Best.