kWeld - "Next level" DIY battery spot welder

tatus1969 said:
tatus1969

A couple questions to tatus1969:

1. With the kweld the adjustment for different welds is in energy J. I'm just wondering if there will be a difference in the weld energy used in each weld when your battery is fully charged and when the battery is close to the LVC? or will the setting of 20J delivery the same energy from the first weld when the battery is fully charged to the last weld when the battery is as discharged as the system allows?

2. When you change from an approved LiPo battery to an approved SLA battery will the same 20J deliver the same energy to each weld with the two different batteries?

Bob.
 
20 Joules is 20 Joules regardless of battery voltage and welding amps because time is the deciding factor.
As the battery voltage/current decrease, to get 20 joules it will simply increase weld time for instance from 10ms to 13ms.
You can observe this by keeping the pedal pressed after a weld until you see the relevant info.

In my experience it's something like sub 10ms weld times but after 1000 welds it's closer to 15ms the battery also gets warm to the touch.
 
bobmutch said:
tatus1969 said:
tatus1969

A couple questions to tatus1969:

1. With the kweld the adjustment for different welds is in energy J. I'm just wondering if there will be a difference in the weld energy used in each weld when your battery is fully charged and when the battery is close to the LVC? or will the setting of 20J delivery the same energy from the first weld when the battery is fully charged to the last weld when the battery is as discharged as the system allows?

2. When you change from an approved LiPo battery to an approved SLA battery will the same 20J deliver the same energy to each weld with the two different batteries?

Bob.
Just like eee291 describes. But this is only true when the pulse duration is small in general. If current is too low, then the require pulse duration grows and more and more heat can travel away from the welding spot, causing it to cool down. That prevents the heat from staying concentrated and will eventually result in poor welds.
 
Hi. This is my first post so I wanted to say hi to everyone.

Hi everyone.

I bought kWeld yesterday but I still have no choice what batteries to buy. What do you recommend for today? Can it be 4s or must it be 3s?

Thank you all for the answers.
 
djmarco83 said:
Hi. This is my first post so I wanted to say hi to everyone.

Hi everyone.

I bought kWeld yesterday but I still have no choice what batteries to buy. What do you recommend for today? Can it be 4s or must it be 3s?

Thank you all for the answers.
This battery that I recommend on the kWeld product page is readily available and delivers enough power for 0.3mm pure nickel welding: https://hobbyking.com/en_us/turnigy-battery-nano-tech-5000mah-3s-65-130c-lipo-pack-xt-90.html?___store=en_us
 
bobmutch said:
tatus1969 said:
tatus1969


2. When you change from an approved LiPo battery to an approved SLA battery will the same 20J deliver the same energy to each weld with the two different batteries?

Bob.

I believe the answer is no: Yo must re calibrate, Bob. Whenever a ( different) battery is swapped the calibration is carried out. This tells the machine the circuit resistance, and is different between batteries.

I think, perhaps another here can confirm.

Im still trying to give my lipos away, Tatus, Bob. Tried both of you by now, friends. Lol. Noone want them to run a K-weld with. Id love to see others result. I have had nothing but success and good resulting calibrations and high energy quick welds.

I wish we saw more recycled EV battery running these Keenlabs K-Welds. I got a bone to pick with low quality Lipos.
 
tatus1969 said:
djmarco83 said:
Hi. This is my first post so I wanted to say hi to everyone.

Hi everyone.

I bought kWeld yesterday but I still have no choice what batteries to buy. What do you recommend for today? Can it be 4s or must it be 3s?

Thank you all for the answers.
This battery that I recommend on the kWeld product page is readily available and delivers enough power for 0.3mm pure nickel welding: https://hobbyking.com/en_us/turnigy-battery-nano-tech-5000mah-3s-65-130c-lipo-pack-xt-90.html?___store=en_us

If you are using the supercap module, you would do better to go with 2S so you stay within the limits of the super caps. They are 2.8v each and there are 3 in series so...8.4v max. I run mine a bit higher at 9.4v or 3.13v per capacitor. It has been fine and I'm personally comfortable with this voltage after many hours of welding. I run well above spec for the super caps. You really don't want to exceed the specs for these caps very much! If you look back a few pages you'll see a few posts by me where I destructively tested a super cap module by deliberately over volting it. You don't want to what did! AKA...if you are using super cap modules, keep the voltage at the maximum limit or less.

If you don't have the super cap module, you can use anything up to 30v. 6S is a pretty common LIPO pack. That's your realistic maximum LIPO pack size for the KWeld. I tried various 10Ah and 16Ah Multistar packs (10C) in parallel to see if they would work. You can look back a few pages for that, but they could not deliver the short high current bursts needed for welding. I also tried 3, 3S, 750mah graphene packs that I use for an RC quad I built. They had the current delivery for good welds. You basically want high C rate, 2-3S LIPO packs and lots of them in parallel. If you buy just one of those packs Tatus recommended, it will weld, but it will run out of charge rapidly. Spot welding takes LOTS of current and that means you need lots of battery capacity to do very many welds. I'd get at least 4 of them to run in parallel if you intend to do any significant amount of sustained welding.

K-weld%20and%203%2065C%20packs.jpg


My solution may not be something that you want to do. Look back a few pages and you can see what I did. I use a 12v/50a meanwell PSU and super cap modules. The PSU is turned all the way down to 9.4v and keeps the super caps charged up. The caps do all the heavy lifting for welding. I don't need a bunch of LIPOs to weld and I don't have any use for being portable since I do all my spot welding at my desk. I'm not suggesting my solution is the best one, but it works for me. I can weld continuously for hours at a time. The only reason I stop is the welding probes get too hot to hold.

welder%20-%20latest%201.jpg
 
Supercup Is not for me. I do not want it. Look for the best power for supercup. I found something like this. What do you think about it

https://hobbyking.com/en_us/turnigy-graphene-6000mah-4s-75c-lipo-pack-w-xt90.html
 
DogDipstick said:
bobmutch said:
tatus1969 said:
tatus1969


2. When you change from an approved LiPo battery to an approved SLA battery will the same 20J deliver the same energy to each weld with the two different batteries?

Bob.


Im still trying to give my lipos away, Tatus, Bob. Tried both of you by now, friends. Lol. Noone want them to run a K-weld with. Id love to see others result. I have had nothing but success and good resulting calibrations and high energy quick welds.

I wish we saw more recycled EV battery running these Keenlabs K-Welds. I got a bone to pick with low quality Lipos.

I'm welding with the 15Ah Lg chem puches similar to those Chevy ones. I made a few 3S packs with varying capacities, I've calibrated the Kweld once with a 4S lifepo4 battery but have been using 3S4P and 3S2P packs without re calibrating once.
 
For average user high c-rate lipo in 3-4s is ok. Or if you don't have one you can take battery out of your car. You cant drive it anyway when welding :lol:

I don't have clue how many welds 3s 5ah pack can do but when it is "empty" you can just take little recharge break..

BTW: If I use lipo with kweld can I at the same time charge it through balance leads? It would keep it nicely full I think :roll:. I have used small charger with lead acid battery.
 
ossivirt said:
BTW: If I use lipo with kweld can I at the same time charge it through balance leads? It would keep it nicely full I think :roll:. I have used small charger with lead acid battery.
yes, you can.

just make sure your charger is isolated or not in any way connected to any part of the welder.
 
ossivirt said:
BTW: If I use lipo with kweld can I at the same time charge it through balance leads? It would keep it nicely full I think :roll:. I have used small charger with lead acid battery.

The pack and KWeld are isolated so the fact that the RC charger is plugged into AC won't be an issue.

I do something similar for battery packs I've built. I use 4S or 5S balance connectors to connect the BMS to the pack. LIPO packs come like that already and adapting the BMS to use these connectors is easy. It works pretty well and means the BMS can be disconnected from the pack if needed. I like to do periodic health checks of the pack and using the individual balance connectors works well for this purpose. I've had a couple of instances of a cell dropping too low. Access to the balance connector makes it possible to recover the low cell. It doesn't matter that the individual balance connectors are 4S in a string of 20 or 200 cells. What matters is the charger is only connected to a single 4S or 5S connector at a time and otherwise remains isolated from the rest of the pack.

I made these 4S adapters a couple of years ago. The XT60 connects to the banana connectors on the charger. The female JST goes to the balance connector on the RC charger. The male end connects to the individual balance lead on any one packs balance connector.

4S%20balance%20charger%20adapters.jpg


Here's 2 sets of 20S packs on the same charger using 5S adapters. Both 20S packs are electrically isolated from each other. The only connection is the individual balance connectors. From the perspective of the charger and the entire dual 20S strings of cells and the individual active balance connections, there is no connection between them. This means no shorts or exploding electronics. The specific balance connector each charger channel is currently connected to doesn't matter as long as you maintain isolation.

5S%20balance%20charge%20cables%202_zpsznrw7dcp.jpg


For your use case, yes you can do this, but you are talking about using a trickle to fill a reservoir gushing over a dam. The JST balance connector will handle 3 amps just fine. More than that and they get hot and melt down. your RC charger will be limited to the connectors capabilities...or 3 amps. I can assure you that you will be pulling down in a single weld a lot more than 3 amps! The RC charger will not be able to keep up with the discharge rate of the welder. At best, you will prolong the time slightly between recharges of your LIPO packs.

I see something similar to this with my set up. If I get to welding too quickly, the single 50 amp PSU falls behind and voltage on my super caps drops off. I'll soon see that my welds are getting weak. I can go 40-60 welds one right after the other before I notice my welds are getting weak. I have 2 super cap modules in parallel and that helps some for bursts of faster welding. In the end if I keep up a faster welding pace, the PSU will fall behind. I've thought about getting a second 50 amp PSU just so this doesn't happen. For now, I keep my weld speed down to accommodate the PSU limits of 50 amps. Your RC charger won't come close to that 50 amps!


ossivirt said:
I don't have clue how many welds 3s 5ah pack can do but when it is "empty" you can just take little recharge break..

I don't want to stop when my weld leads get hot and that takes a minute for them to cool off. Stopping to recharge, that takes 30 minutes or more...ugg! It's why I use a PSU so I never have to stop for more than a minute. I wait for the PSU to top off the super caps and get back to welding again. My welding probes are getting aerogel fabric on them so I can keep going when they are too hot to hold onto. Just my opinion, but running from battery packs with limited capacity is not the way to go. IF I did this, I'd use a pack with something like 100Ah of capacity.
 
DogDipstick said:
I believe the answer is no: Yo must re calibrate, Bob. Whenever a ( different) battery is swapped the calibration is carried out. This tells the machine the circuit resistance, and is different between batteries.
I think, perhaps another here can confirm.

I know you have to re-calibrate with each different battery you connect to the Kweld, That wasn’t my question.
 
eee291 said:
20 Joules is 20 Joules regardless of battery voltage and welding amps because time is the deciding factor.
As the battery voltage/current decrease, to get 20 joules it will simply increase weld time for instance from 10ms to 13ms.
You can observe this by keeping the pedal pressed after a weld until you see the relevant info.

In my experience it's something like sub 10ms weld times but after 1000 welds it's closer to 15ms the battery also gets warm to the touch.

I understand E=Pt and P=IV. So I used 18J setting for 0.1mm nickel and 30J for 0.2mm nickel. My question was if I change from one approved battery to another will the J setting on the kWelder need to be adjusted or will it be the same. It sounds like the same Energy setting should work as long as the battery is within the suggest specs and it is not discharged to a point where t is to long. thanks.

Another question my battery is a AGM AlphaCell 220GXL with 20hr109Ah, max discharge 900A, Impedance O.005Ohms, and 2800A short circuit, does this battery have to high amps or will it be fine?

 
DogDipstick said:
bobmutch said:
2. When you change from an approved LiPo battery to an approved SLA battery will the same 20J deliver the same energy to each weld with the two different batteries?
I believe the answer is no: Yo must re calibrate, Bob. Whenever a ( different) battery is swapped the calibration is carried out. This tells the machine the circuit resistance, and is different between batteries.

I think, perhaps another here can confirm.
The answer is actually yes, I responded further up. kWeld's energy metering ensures that the results are fairly independent from batter charge status and amp delivery, as long as pulse durations stay low (long pulses cause heat spreading at the spot). Also the calibration only measures at the output of the welder, so you don't need to repeat this when swapping between similar batteries.

DogDipstick said:
Im still trying to give my lipos away, Tatus, Bob. Tried both of you by now, friends. Lol. Noone want them to run a K-weld with. Id love to see others result. I have had nothing but success and good resulting calibrations and high energy quick welds.
I must have missed something :roll:
 
ossivirt said:
BTW: If I use lipo with kweld can I at the same time charge it through balance leads? It would keep it nicely full I think :roll:. I have used small charger with lead acid battery.
There's one risk to that. The inductive kickback from terminating a pulse causes significant voltages for short periods. You can get up to 60V at kWeld's power switch for a few ten microseconds. The amplitude is significantly at the battery terminals but still there. These spikes could potentionally damage the charger's balance if it isn't well protected.
 
bobmutch said:
Another question my battery is a AGM AlphaCell 220GXL with 20hr109Ah, max discharge 900A, Impedance O.005Ohms, and 2800A short circuit, does this battery have to high amps or will it be fine?
Perfect specs for kWeld. How much pulse current does it report?
 
I do not think anything that was deleted today from this thread was off topic.

Every post contained relevance to batteries, welding, and the K-welds construction and operation capabilities.

This is all I have to say here anymore.
 
DogDipstick said:
I do not think anything that was deleted today from this thread was off topic.

Every post contained relevance to batteries, welding, and the K-welds construction and operation capabilities.

This is all I have to say here anymore.
What happened, where's your last post? I can only add that nothing that you wrote did offend me in any way...
 
spinningmagnets said:
There were several posts that were turning into a personal discussion. I deleted them.

Oh darn...I missed the drama...haha.
Thanks for helping us keep this clean.
Or rather...thank GOD I missed the drama!
 
Here is a crazy question for kWeld's creator, Mr. T.
I recently bought a new AGM battery and both of my kWelds units complain of overcurrent when I try to calibrate them.

If I swap the battery and electrode cables with thinner ones, I am able to successfully calibrate them but I know I am getting too close to the upper limit of the juice these great welders can handle, not to mention, I am not able to use all the energy this battery can produce.

So I got to think, what if I buy a solid sate relay really beefy and somehow add it to the mix, either by bypassing the MOSFETs or adding its capacity to the kWeld. This way I would have plenty of headroom in terms of current-handling capability.

This is the one I had in mind (or perhaps a different one would be better) because when you read the specs, it is one of the biggest available in that price range.
https://www.mouser.com/ProductDetail/558-HDC200D160H

The datasheet claims a "Maximum Surge Current (10ms) [Adc]" of 1600:
https://www.mouser.com/datasheet/2/657/hdc-series-dc-panel-mount-1370829.pdf

Is this even remotely doable?
I am OK with gutting one of my welders for testing this. I can always buy a new one to replace it if it dies in the operating table.

Doing this kind of experimentation is in part why I got into building my own battery packs and really enjoy modifying and optimizing my tools.
 
NetPro said:
Here is a crazy question for kWeld's creator, Mr. T.
I recently bought a new AGM battery and both of my kWelds units complain of overcurrent when I try to calibrate them.

If I swap the battery and electrode cables with thinner ones, I am able to successfully calibrate them but I know I am getting too close to the upper limit of the juice these great welders can handle, not to mention, I am not able to use all the energy this battery can produce.

So I got to think, what if I buy a solid sate relay really beefy and somehow add it to the mix, either by bypassing the MOSFETs or adding its capacity to the kWeld. This way I would have plenty of headroom in terms of current-handling capability.

This is the one I had in mind (or perhaps a different one would be better) because when you read the specs, it is one of the biggest available in that price range.
https://www.mouser.com/ProductDetail/558-HDC200D160H

The datasheet claims a "Maximum Surge Current (10ms) [Adc]" of 1600:
https://www.mouser.com/datasheet/2/657/hdc-series-dc-panel-mount-1370829.pdf

Is this even remotely doable?
I am OK with gutting one of my welders for testing this. I can always buy a new one to replace it if it dies in the operating table.

Doing this kind of experimentation is in part why I got into building my own battery packs and really enjoy modifying and optimizing my tools.
My guess would be: no way :) kWeld is basically a solid state relay with 2160A continuous / 8640A pulsed, when just taking the MOSFET datasheet numbers. In reality, I'd guess it can do 300A continuous, and the pulsed limit is known :) And keep the turn-off transition time into account, the SSR's power dissipation is 12V * 2000A = 24kW during that time. For that reason, kWeld has fast switching times. And please make sure to make your wiring as short as possible, or you can easily damage the kWeld. Make sure to read and fully understand the corresponding section in the user manual (inductive kickback).
 
I was afraid you would say that. But it was worth trying.
Thanks for considering the idea!

I have read the whole manual and if I use short wires (around 15cm) made of AWG #4, with the battery I just bought, my kWelds complain of overcurrent.
If I use thinner wire (AWG #8), calibration completes successfully.

So, it is clear that the amount of juice the battery can deliver is just a bit too much for the welders, hence me trying to increase the amount of current they can handle: I don't like running devices so close to their limits.
 
Happy 2020 everyone,

I've spent most of my day reading through this thread when I should be working..... I've just completed order of a full kit and laser cut case, if for no other reason, the support and the work that has gone into this thing is extraordinary.

I am inspired and very impressed, both of which seem to become more elusive as the years pass.

Now I am a soon to be a fully badged member of the KWeld club I have two questions.

Thanks Paul



1) I'm building 3x 20S2P, MOLICEL/NPE INR-21700-P42A 45A 4200MAH FLAT TOP 21700 packs with 60Amp BMS, so the series need to be 30Amp.

https://www.reddit.com/r/electronic_cigarette/comments/b6uyok/bench_retest_results_molicel_p42a_4000mah/

I have been reading about the Nickel/Copper Sandwich for the series connections I'm thinking 8mm 0.15mm Nickel/ 20mm 0.10mm Copper.

https://www.electricbike.com/introduction-to-battery-pack-design-and-building-part-3

How much is know about welding this configuration with KWeld and are there any recommendations?

2) Is their value in this?

Screen Shot 2020-01-03 at 1.47.38 pm.png
 
Back
Top