Soldering iron for 18650 packs

Why using bigger dimensions on the parallel connecting wires than on series connections?

Load on series connections are bigger right, so the other way around would make more sense, wouldn’t it?

1 Like

Hey @longhairedboy what bland of oil solvent and flux do you use? Do you clean all your solder jobs with oil solvent after finish soldering or you use it before applying heat?

Instead of nickel strips I just 0.2mm copper sheet..

Its much easier to solder than nickel strips and for the same area can carry more current iirc.

You can go thicker than 0.2mm but that thickness is easy to cut with regular scissors to any size you need. If I need higer current I just use double or triple layers.

Copper has like 5 times better conductivity than nickel, I use copper braid, also with braid is easier to make it a bit more flexible the :battery:. Some flux and 100+watts solder and ready to go.

1 Like

i use liquid flux in a little needle tipped squirt bottle. Cleanup i use rubbing alcohol and a clean white rag.

https://www.amazon.com/gp/product/B005T8KZ0I/ref=oh_aui_search_detailpage?ie=UTF8&psc=1

and i pour it into this for application during jobs: https://www.amazon.com/gp/product/B00UG08QDC/ref=oh_aui_search_detailpage?ie=UTF8&psc=1

i use rubbing alcohol to clean up the tips on the bottles as well.

In this case for a 12s4p battery:

Series connections - 4 x 12 awg wires, one on each series line (6 awg equiv) Parallel Connections - 1 x 8awg wire

so technically the series has a better connection than parallel like you said.

Load per series is 20A x 4 which 12awg x 4 will handle, and load on parallel is 80a which 8awg, while a little undersized for continuous load, should handle. VESC 4 will likely overheat before the wires do, though the VESC 6 may change things.

Some pics to clarify:

http://esk8content.nyc3.digitaloceanspaces.com/uploads/db2454/optimized/3X/1/5/154a7070c89196cf2892ace326b542a92ef117a6_1_331x500.png

12awg on the series lines

http://esk8content.nyc3.digitaloceanspaces.com/uploads/db2454/optimized/3X/5/4/540ce1b672408e945628bb4de37c699c513af820_1_667x500.jpg

8awg on the parallel lines

3 Likes

The problem with using wire and solder directly to the battery ends is that for a low guage wire (12/14 awg) you need A LOT of heat to melt a large amount of solder to engulf the whole wire and make it grip to the cell. The problem with that is, the batteries are great conductors of heat and most of the heat from the soldering iron gets conducted into the metal of the cell, thus damaging the inners of the cell.

If you want to make a flexible battery but still use solder, the best option is to weld small tabs of nickel to the ends of the cells, with enough nickel so that it can be slightly bent away from the cell, then rough up that nickel with a sharp end or sand paper and solder to that. The nickel basically absorbs most of the heat and much less dissipates to the cell.

You can make a spot welder from a car battery, a solenoid relay and some heavy guage cable for next to nothing, like mine…

2 Likes

My cell fused parallel lines are running 22awg from cell to buss bar and used the .016" x .5" buss bar for series

1 Like

Soldering can be tricky - but I’ve found a good way to keep the battery temps low.

Use a ~100w soldering iron in an effort to be quick. Pre-tin the cells with solder+flux, only takes a few seconds to get the solder on and to stick with the big soldering iron. Cells get warm, but not too hot to hold. Then, with the braided wire, just set it on top of the pre-tinned solder, put the soldering iron on top of the braided wire, and let it melt in using a little bit of rosin(flux) core solder to start the process. Takes a few seconds and the cells don’t even get warm.

The 8awg wire was a bit trickier to keep the cells cool, but pre-tinning both the batteries and the wire separately, then soldering together worked well.

I decided to look through my VESC tool - I’ve got 632 logged miles on the battery so far, so about 35-40 discharge/charge cycles with no BMS and just a meanwell adjustable 50v 4a bulk charger (I know, I know) but all the Samsung 25R cells still check out as follows:

Bulk Charger set to 49.6v (4.133333333v)

4.14 4.14 4.13 4.13 4.14 4.13 4.15 4.14 4.13 4.14 4.14 4.13

The pack has never been balanced.

A quick pivot table of the CSV files I found some more specs to see how the battery has been used:

Number of ride logs: 98 Total Distance: 632 miles Max motor amps: 95.03 (single 6374 motor) Max Battery amps: 54.61 (13.6a per cell) Max Watts: 2283 Max (GPS) Speed: 30.1 mph / 48.17 km/h * Min volts (not counting disconnects): 37.4 (3.11v / cell)** Min Temp: 6.8c / 44.24v (cold morning test ride) Max temp: 81.4c***

*this is all it will do at 95% duty cycle. Should go faster with taller gears.

**this was one or two rides I managed to kill the battery completely after 22 miles and was trying to get something out of the battery to get home. VESC protection prevented me from going any lower.

***vesc de-powered on a LONG uphill run on a hot day

Overall - I believe this style of battery building holds up extremely well, provides plenty of large pathways for high power capacity, and carefully soldering the cells doesn’t have any impact on performance or longevity.

3 Likes

https://www.mcmaster.com/#grounding-braid/=16gi4yq – these guys are legit. Ordered this morning and received today. Maybe it’s my proximity. Thanks for the link.

Do you heatshrink or electrical tape the copper braided flat wire or just leave open?

I’d be worried about the longevity of electrical tape, and the heat resistance.

The plan was to buy some large heat shrink and encase the entire battery in that. For some reason I just never got around to doing it. Since it’s enclosed in a plastic case, it may not be necessary to cover the battery. The connections won’t move if soldered correctly, and there is nothing metal to touch.

There are some vendors on eBay that sell large diameter shrink tubing. A paint gun (or hair drier in a pinch) would do the trick.

This probably isn’t the right size, but it was what I saved in my bookmarks a few months ago.

Hey, i’ve just read the entire thread and it’s super informative, thanks.

regarding the thickness of the wire in the parallel connections - you said that you’ve used an 8 guage wire. Isn’t that a problem if one of the cell will get bad and shortout ? In that kind of situation the other cells in the group will pump the faulty cell and get ruined themselves.

In my understanding it is better to use a thick wire for the series connection and to isolate each cell with a thinner wire for the parallel connections.

would like to hear you opinion, thanks.