kWeld - "Next level" DIY battery spot welder

[garolittle]

I use the 120 Amp rated Anderson connectors and feed both Negs and Pos into one connector for each pair. Matching connector is on KWeld.

Nice. Simple solution. Really like it.

 

[BVH]

Here's a mock up of how I balance charge the 2P pack. Not shown, but obvious are the balance board input cables and balance cable which are connected to my FMA Dual PowerLab 8 .



It may look a little "Mickey" with the alligator clips but they are rated for 10 Amps and I charge at about 4-5 Amps. I did make and use an Anderson 120 Amp-to-Anderson standard connector adapter, just didn't get it out for the pic. It's a little more refined! Any multi-pak balance board will work fine. Since the packs are permanently Parallelled, they are always at the identical Voltage and they are balance charged to within about 3-5 Millivolts. I keep them at 3.75 Volts Per Cell storage charge when not using them. I charge the pack just before use and discharge (if necessary) right after use.

 

[garolittle]

Clever. I really like it. Thanks for posting that information. Gary.

 

[SubnetMask]

Here's a mock up of how I balance charge the 2P pack. Not shown, but obvious are the balance board input cables and balance cable which are connected to my FMA Dual PowerLab 8 .



It may look a little "Mickey" with the alligator clips but they are rated for 10 Amps and I charge at about 4-5 Amps. I did make and use an Anderson 120 Amp-to-Anderson standard connector adapter, just didn't get it out for the pic. It's a little more refined! Any multi-pak balance board will work fine. Since the packs are permanently Parallelled, they are always at the identical Voltage and they are balance charged to within about 3-5 Millivolts. I keep them at 3.75 Volts Per Cell storage charge when not using them. I charge the pack just before use and discharge (if necessary) right after use.

Looks like a pretty good setup (not judging the alligator clips as they look beefy and charging, you're not really typically pushing a crazy amount of current - for the initial charge of my new packs, I used MUCH less substantial clips and wires from the charger, but I was only charging at 1A), but just to throw it out there for anyone new that doesn't know, you need to be VERY careful about parallel charging like that - You need to be very aware of your pack and cell health. If you parallel charge multiple packs and one of them has a bad cell, the bad cell in that pack can wipe out the same cell in ALL connected packs.

 

[SubnetMask]

So I FINALLY got the XT90 connectors that USPS was supposed to have delivered LAST Saturday and was able to complete my setup and actually use my kWeld. I think to say that the kWeld is better and more consistent than the Suckko welder I had originally tried is the understatement of the century. VERY happy with the kWeld. I tried some welds on some old worn out cells and every weld with .2mm pure nickel at 30J was solid. With the Suckko welder, with the exact same material and everything (pulses and current) set to the max, they'd fall off the battery if you looked at them hard. It could NOT do .2mm nickel. I then built my first 'real' pack, a little 2s6p deal for a 6v 'power wheels' type ride on toy. I had charged the packs to the 3.8v/cell storage level and that seems to have worked really well (That was the level that I had done the calibrations at). After around 112 welds on that pack (48 on one side, then 48 on the other, then where I had to weld nickel strip to nickel strip because I left the tabs for the BMS at the wrong end of the pack ), plus the handful of 'test' welds, the only thing that I noticed get hot was the electrodes. The batteries, 10ga wire and XT90s never got warm enough for me to feel. And the welds didn't seem to take anything out of the batteries - they were still up at full storage charge (My ISDT Q8, which to be honest, I rather like and would recommend, wouldn't charge them because they were all at ~3.8v/cell).

After Franks last comment regarding joining the batteries at ring terminals rather than the XT150s, this is what I came up with:

The calibration numbers I got out of it were:
1 Pack
R=2.37mR
Std=19.6
I=1206A
T=26.6C

2 Packs
R= 2.19mR
Std= 25.2
I= 1902A
T= 27.0C

 

[garolittle]

After Franks last comment regarding joining the batteries at ring terminals rather than the XT150s, this is what I came up with:kWeld.jpg

The calibration numbers I got out of it were:
1 Pack
R=2.37mR
Std=19.6
I=1206A
T=26.6C

2 Packs
R= 2.19mR
Std= 25.2
I= 1902A
T= 27.0C

Very nice. Low resistance numbers.

 

[tatus1969]

The calibration numbers I got out of it were:
1 Pack
R=2.37mR

2 Packs
R= 2.19mR

Very nice. Low resistance numbers.

I guess that I need to explain something here The displayed resistance from the SHORT calibration step is not the measured internal resistance of the battery, but the resistance of the output leads. That doesn't change with current of course, and the fact that you get a different reading at higher current is due to limited accuracy of the kWeld's measurement system itself. If it would work 100% accurately, then you'd get the same reading regardless of the current. It doesn't need to be a high precision instrument however, because the actual task is to create repeatable welding results, and it does a decent job there 8)

I think to say that the kWeld is better and more consistent than the Suckko welder I had originally tried is the understatement of the century. VERY happy with the kWeld.

Thanks 8) 8)

 

[garolittle]

Followed the method above (why reinvent the wheel?

The calibration numbers I got out of it were:
1 Pack
R=2.20 mr
Std=29.2
I=1530A
T=32.1C

How do these numbers look? Am I ready to go? Thanks for all the great advice.

 

[tatus1969]

1 Pack
R=2.20 mr
Std=29.2
I=1530A
T=32.1C

How do these numbers look? Am I ready to go? Thanks for all the great advice.

Perfect!

 

[garolittle]

1 Pack
R=2.20 mr
Std=29.2
I=1530A
T=32.1C

How do these numbers look? Am I ready to go? Thanks for all the great advice.

Perfect!

Yes Sir and thanks again for making such a great product. Gary

 

[raming]

I received the kWeld kit last week from Germany and am delighted with the: prompt delivery, excellent build quality and consistent weld performance. The first application was for replacing the five 18650 batteries in an M18 Power Tool pack. The only improvement that I recommend is to Label and Color-Code the Battery Cables to avoid incorrect hookup.
The battery used for powering the kWeld is an

ODYSSEY Extreme PC925 12V Auto Battery (bought via Amazon):
-Sealed AGM with Posts and Stud Connections. Rated for both Starter and Deep Cycle use.
- 925amps PHCA (5 second pulse current rating), 4.3 mOhms rated internal resistance, 1000A Short Circuit current rating, 330 CCA
-26lbs weight
The battery was tested with both Post Clamps and Stud Bolts cables with similar performance (1.1m total cable length). I prefer the Stud/Lug Connected cable set since they are much easier to disconnect.
Open:
-Offset 65
-Std 3.4
Short:
-R 2.38 mOhms
-Std 15.2
-I 1389A
Temp 23.3C
This 4 year old 28 Ahr battery was used for over 100 welds with no significant reduction in capacity (Voltage dropped from 13.1V to 12.9V).

 

[tatus1969]

ODYSSEY Extreme PC925 12V Auto Battery

-I 1389A

That sounds like an amazing option for kWeld, since that battery delivers more than enough current while still being considerably smaller than a regular 65AH/100AH starter battery (at least what I can tell from your picture). Please tell me the amount of current that you get during welding, I'd like to add it to the list of recommendations on my website.

 

[john61ct]

I've used their G-31 size PC-2150 for years, absolutely fantastic as a deep cycling "dual use",

can last 6-8yrs in daily use if cared for

also cranks even big diesels like there's no tomorrow.

Never realized they had little ones in the same series. . .

 

[spinningmagnets]

That's impressive for an AGM. From what I have learned about the internal configuration of batteries that allows high pulse-amps, there was no reason that one like this couldn't have been made a long time ago. There only needed to be a company that thought there was a market for a physically small battery that easily provides high amps. (a quick scan of web info sounds like these are popular for motorcycles, and also race cars of a type that are required to have a self-starting ability)

I am keen on using a 12V battery for future kWeld builds, because the chargers are readily available and affordable. The only downside from my perspective has previously been how large a car battery needed to be to provide enough amps. A big battery is heavy and clunky to move around. Plus, it's always a good thing to have a spare car battery...I am looking forward to seeing your results...

26 lbs is very manageable.
6.7″ L x 7.1″ W... 5.8″ Height

 

[john61ct]

Yah, Enersys engineers invented a lot of the TPPL fundamentals that the industry grew out of, both Northstar and Lifeline were alumni spinoffs.

Originally made the old Sears DieHards, and licensees of the spiral Optima design too, before both got sold down the river to (meh) Johnson Controls.

 

[raming]

That sounds like an amazing option for kWeld, since that battery delivers more than enough current while still being considerably smaller than a regular 65AH/100AH starter battery (at least what I can tell from your picture). Please tell me the amount of current that you get during welding, I'd like to add it to the list of recommendations on my website.

I welded 0.12mm thick nickel strips to Samsung 18650 2Ahr 20A batteries with the following parameters:
30 Joule Setting
18 mS pulse typical duration
1252 Amps typical current
1.0 mOhm typical Resistance

I bought this battery four years ago because of the claim of Deep Charge and High Starting Current in a sealed relatively small format. I carry it to RC Plane Flying fields for charging LiPO battery packs and also use it as a starting battery my V8 Block Inboard skiboat. The only downside is that it cost about $175 which is 2X more than flooded cell starter batteries. However, it has withstood many deep discharge cycles and will probably last at least twice as long.

 

[bentms]

Hi Guys

I managed to screw up my original battery supply Wires (I.e the wires that connect the battery to the kweld, not the actual battery wires) so need to make some new ones.

Could you tell me how long they should be? Or is it the shorter the better?

Cheers

Ben

 

[SubnetMask]

Hi Guys

I managed to screw up my original battery supply Wires (I.e the wires that connect the battery to the kweld, not the actual battery wires) so need to make some new ones.

Could you tell me how long they should be? Or is it the shorter the better?

Cheers

Ben

I think the 'real' answer is 'the shorter the better', but the stock negative cable is about 9cm long, and the positive I believe is about 10cm.

 

[bentms]

Thanks so much, you’re a star

 

[999zip999]

Ok a friend wants to buy one so I'm by pushing this thread to the top thank you everyone

 

[garolittle]

I think I know the answer but I wanted to ask in case I’m wrong. Tonight I was spot welding .25 mm nickel plated copper strips (about 65 - 75 joules). I was curious why I encountered an “over current” message when I used two Lipo batteries in parallel instead of using just one Lipo battery. I have successfully spot welded pure nickel strips (i.e. no copper) many times using one Lipo battery and also using two Lipos connected in parallel but I have never encountered an over current message. My theory is that the nickel plated copper has much less resistance and therefore using two Lipos in parallel triggers the error code. What do you guys think?

 

[serious_sam]

Sounds about right. You could try with partially discharged packs. That might lower the available current enough to get below the limit.

 

[garolittle]

Sounds about right. You could try with partially discharged packs. That might lower the available current enough to get below the limit.

Thanks. I really enjoy learning. I pulled about 1,300A with one Lipo but got the “nearly silent” Overcurrent beep when using the two in parallel.

 

[garolittle]

Sounds about right. You could try with partially discharged packs. That might lower the available current enough to get below the limit.

Thanks again. That actually gave me an idea. Would tungsten electrodes be an option since they have higher resistance? That may put me under the 2,000A limit but still provide enough current to weld copper strips. What do you think?

 

[tatus1969]

I think I know the answer but I wanted to ask in case I’m wrong. Tonight I was spot welding .25 mm nickel plated copper strips (about 65 - 75 joules). I was curious why I encountered an “over current” message when I used two Lipo batteries in parallel instead of using just one Lipo battery. I have successfully spot welded pure nickel strips (i.e. no copper) many times using one Lipo battery and also using two Lipos connected in parallel but I have never encountered an over current message. My theory is that the nickel plated copper has much less resistance and therefore using two Lipos in parallel triggers the error code. What do you guys think?

I'd say your theory is spot on, the copper strips have significantly less resistance and push the current above the limit. I'd also try using partially discharge batteries first.