welding 18650 cells with DN-5 spot welder
- Thread starter madin88
- Start date
cut cat said:
Copper would be the best at performance, but there are some reason because nickel is chosen
- Nickel doesn't corrode over the time as copper does
- Nickel is easy and faster to weld than copper
- Nickel can be produced in laminations of different thickness easier than copper
- There are not better electrical conductor than copper (silver but..), so the poles of your welder will get stick to the tabs or disintegrate.
- You need to much amps to create the enough resistivity through a copper lamination to get welded to the cell, you won't be able to weld copper at the classical 220V/120V 15A breaker at home
(all tanking into account the typical needed for welding 18650 cells between 2.5A - 8A conducting capacity for each strip)
okashira
10 kW
Nickel is easier to weld, more corrosion resistant and stronger the copper.
For copper welding, I will probably want to do something like pot the copper strip in a coat of epoxy to ensure it can't be torn off.
For copper welding, I will probably want to do something like pot the copper strip in a coat of epoxy to ensure it can't be torn off.
I've just now put the DN-10 through some testing with 0.3mm nickel strip.
No sparks now, and the welds were consistant... though none were taking. Even at too much power the weld would not take.... by too much I mean the wood underneath the weld would vaporise and char... which would be devastating to a battery.... in your face.
Too much heat, the pulses are too long and not consistent enough, and the welds do not take with the 0.3mm. At 1.0mm it was blowing the pudding out of it.
Not impressed so far.
No sparks now, and the welds were consistant... though none were taking. Even at too much power the weld would not take.... by too much I mean the wood underneath the weld would vaporise and char... which would be devastating to a battery.... in your face.
Too much heat, the pulses are too long and not consistent enough, and the welds do not take with the 0.3mm. At 1.0mm it was blowing the pudding out of it.
Not impressed so far.
litespeed
100 kW
I think that is the problem I'm having which is too much power with my 800 watt second capacitor discharge welder. I'm trying different tips on mine, maybe that is what yours needs?
Tom
Tom
litespeed
100 kW
I'm surprised every time someone asks this question. Of course! if I pushed any harder I put a hole in the top of the cell.
Hmm..
Hmm..
okashira
10 kW
what voltage you running?litespeed said:
what kind and how long electrode wires?
What kind of connections?
litespeed
100 kW
okashira said:
0 to 23.2 volts
My leads are 1 gauge
400 amp twist lock leads and soldered 1 gauge to 3/8" solid copper rod.
Tom
team_EMUS said:
ultimately a pack would be assembled / disassembled without any permanent connections to the cells. People on the forums have build packs which use springs and pressure plates or bolted pressure plates to achieve weld free packs - but still in prototype phases. There was a kit (will have to find it again) which enabled the cells to be loaded into a cartridge system - which looked to be extremely good.
S.
Allex
100 MW
I usually remove the welded tabs with dremel, simply polish with a cut wheel until you see the can of the cell. Just aim for the weld spots and cut there.
Dont like the tearing methods as you can damage the cell and worst case make a hole in it while doing it.
If the nickel strip are thin, like 0,1mm than it is usually okej to tear them apart, but carefully.
BTW, this is 0,3nickel welded with dn10
Dont like the tearing methods as you can damage the cell and worst case make a hole in it while doing it.
If the nickel strip are thin, like 0,1mm than it is usually okej to tear them apart, but carefully.
BTW, this is 0,3nickel welded with dn10
litespeed
100 kW
riba2233 said:
I'm adjustable up to 30 or 60 ms....I forget right now. I think the copper or the tip style/shape is the downfall.
Tom
okashira
10 kW
litespeed said:
Way too high ...
and 1 awg is overkill
I use 4 awg, even 6awg would work
go to 4 awg, use .125" electrodes.
start with 7ms pulse time. or less.
I use ~4ms for .15nickel
Start with 7Volts
If you get blowing with that, then there's a problem with the welder/design/microcontroller
Allex said:
Hi Alex... these look like decent welds - at what volume did you have the welder set to?? Any higher than 1.5 I found that there was extreme amounts of heat produced... Have you tested welding 2 strips together sitting on a block of wood (so you can see the heat result)?? I have not been able to get a satisfactory result.
kdog
10 kW
hey willow
im having good results now that i have done 3 things.
first-i reduced my electrode tip SA- it was very sensitive to this. not exactly sure what my tip SA is now but im using old copper wire used to wire houses for my electrodes... at a guess roughly 1.5 mm^2. good thing about this is the gauge is uniform so it doesnt get fatter as they wear.
second- i tried using lots of pressure and it didnt work, by chance i reduced the pressure to quite light and got a good weld straight up. i think the lower pressure increases resistance slightly between the work peices and creates more heat for less current. now that i have built my proper electrode arms this is much more consistent. of course to little pressure tends to blow out a hole as an arc forms. i guess this is dependant on the power are using.
lastly (and im still waiting for a few bits) ive got a a timer instead of the variability of my foot/thumb/whatever.
i realise with your machine, none of these may be applicable cause youve prob already got these things sorted, and my power levels are probly lower- but i nearly gave up until my lucky results with the first two point steered me in the right direction (for me)
ive been using the wood heat guide too
very techincal! but now with a good weld i leave no mark and it is only warm to touch within a second or so.
my best results are welding aprox .15-.2mm tab to a thicker backing~.3mm although i can do down to .1tab to .1backing (specially once my SSR turns up)
ran into a guy in SOHO yesterday sporting a sweeet looking bike- i stopped and talked and had a squiz. it was very nice.. turned out its one of eric's machines. not too many people in hobart doing that sort of work (i feel like a hack now).
anyway that ride needs to get done-tried last week but ran out of time- ill keep in touch re ride
K
im having good results now that i have done 3 things.
first-i reduced my electrode tip SA- it was very sensitive to this. not exactly sure what my tip SA is now but im using old copper wire used to wire houses for my electrodes... at a guess roughly 1.5 mm^2. good thing about this is the gauge is uniform so it doesnt get fatter as they wear.
second- i tried using lots of pressure and it didnt work, by chance i reduced the pressure to quite light and got a good weld straight up. i think the lower pressure increases resistance slightly between the work peices and creates more heat for less current. now that i have built my proper electrode arms this is much more consistent. of course to little pressure tends to blow out a hole as an arc forms. i guess this is dependant on the power are using.
lastly (and im still waiting for a few bits) ive got a a timer instead of the variability of my foot/thumb/whatever.
i realise with your machine, none of these may be applicable cause youve prob already got these things sorted, and my power levels are probly lower- but i nearly gave up until my lucky results with the first two point steered me in the right direction (for me)
ive been using the wood heat guide too
very techincal! but now with a good weld i leave no mark and it is only warm to touch within a second or so.my best results are welding aprox .15-.2mm tab to a thicker backing~.3mm although i can do down to .1tab to .1backing (specially once my SSR turns up)
ran into a guy in SOHO yesterday sporting a sweeet looking bike- i stopped and talked and had a squiz. it was very nice.. turned out its one of eric's machines. not too many people in hobart doing that sort of work (i feel like a hack now).
anyway that ride needs to get done-tried last week but ran out of time- ill keep in touch re ride
K
litespeed
100 kW
okashira said:
I am going to 1/8" electrodes from Sunstone and should be here tomorrow. The designer of my welder recommends the 1 gauge wires to lower the resistance. I'll try the electrodes first and start experimenting only if that doesn't work.
Thanks for the advise though and I'll report back.
Tom
as mentioned earlier i was going to try nickel-copper alloy sheets. i did a bit of research and almost bought this or similar stuff like in the link below.
http://www.ebay.de/itm/1-kg-Schweis...b&pid=100005&rk=4&rkt=6&sd=160974484305&rt=nc
this has 61% copper, 18% nickel and 20 % zinc, but a very poor conductivity of only 4% of IACS (copper is 100%) which made me skeptical.
than i found this chart:
http://eddy-current.com/conductivity-of-metals-sorted-by-resistivity/
electrolytic nickel: 25%
nickel: 18%
Copper 90%, Nickel 10%: only 9,74%
Copper 70%, Nickel 30%: only 4,5%
nickel coated steel (hilumin) should be above 10%
its not plausible for me why copper-nickel is so poor, but it looks like all the hype and promotoion seems to be only parody.
http://www.ebay.de/itm/1-kg-Schweis...b&pid=100005&rk=4&rkt=6&sd=160974484305&rt=nc
this has 61% copper, 18% nickel and 20 % zinc, but a very poor conductivity of only 4% of IACS (copper is 100%) which made me skeptical.
than i found this chart:
http://eddy-current.com/conductivity-of-metals-sorted-by-resistivity/
electrolytic nickel: 25%
nickel: 18%
Copper 90%, Nickel 10%: only 9,74%
Copper 70%, Nickel 30%: only 4,5%
nickel coated steel (hilumin) should be above 10%
its not plausible for me why copper-nickel is so poor, but it looks like all the hype and promotoion seems to be only parody.
beefed up the hilumin with nickel. hilumin was about 3,3mm² and nickel now 1,8mm² copper equivalent so entire connection has conductivity of about 5mm² copper. adaptto shows an IR in the range of 30-50 mOhm. thats great!
finished packs for peli 1200 cases:
finished packs for peli 1200 cases:
litespeed
100 kW
I'll tell you, no one builds a better looking battery pack than you. What a great eye for detail. Hope you don't mind but that was why I copied your design on my own packs. Only difference is I used .2 mm nickel everywhere and two strips (one on top of the other) for my single rows.
Thanks for the great idea!
Tom
Thanks for the great idea!
Tom
Attachments
thanks! building a pack like this and use 25mm nickel for both parallel and serial connection makes it more easy as welding many little strips and it has smallest possible size.
tom, do you weld both strips at once or step by step?
i noticed there is risk of short if you weld one sheet onto another and the dot or one of the dots is far outside the middle at positive side. now i use self sticking insulation washer to have more protetion there.
tom, do you weld both strips at once or step by step?
i noticed there is risk of short if you weld one sheet onto another and the dot or one of the dots is far outside the middle at positive side. now i use self sticking insulation washer to have more protetion there.
litespeed
100 kW
I welded them one by one and you are right. This welder I built has so much power that if I am on the edge of the electrode it will blow a hole in the nickel. After that happened twice I was very careful of the probe placement.
I was afraid to add all that heat to the top of the cells so I left my strips long, soldered the wire to it and tried to spot weld the nickel down but it won't work with solder on it. So I ended up soldering the strips together.
Seems to work well with resistance in the 13 to 15 range per pack as measured by my Thunder Power TP820CD charger.
I agree that next pack will get many more washers than I used on this one. I wrapped my packs in thin foam and core board to protect them so I know these are ok.
Tom
I was afraid to add all that heat to the top of the cells so I left my strips long, soldered the wire to it and tried to spot weld the nickel down but it won't work with solder on it. So I ended up soldering the strips together.
Seems to work well with resistance in the 13 to 15 range per pack as measured by my Thunder Power TP820CD charger.
I agree that next pack will get many more washers than I used on this one. I wrapped my packs in thin foam and core board to protect them so I know these are ok.
Tom
Joost, do you mean the fuse or circuit braker in your house that cuts off?
do you have 110V or 230V version and whats the amp rating of your fuse?
In my case it works well with 230V and normal 16A braker (up to time setting 5-6, no more tested) - even if it peaks at 5kW which would be 22A :wink:
DN-10 or higher definitely need a fuse or braker upgrade.
do you have 110V or 230V version and whats the amp rating of your fuse?
In my case it works well with 230V and normal 16A braker (up to time setting 5-6, no more tested) - even if it peaks at 5kW which would be 22A :wink:
DN-10 or higher definitely need a fuse or braker upgrade.
