DIY Electric Car Forums banner
1 - 12 of 12 Posts

·
Registered
Joined
·
6 Posts
Discussion Starter · #1 ·
This may not be the appropriate forum for this but I would appreciate any assistance anyone would be willing to offer. I am trying to spot weld a pack out of 26650 cells. I built my own spot welder using the microwave transformer and some 2 ga. wire. with a timing relay to control the pulse. New to spot welding but I know a little about arc welding. From what I gather I suspect that I do not have enough current. When I get up to about a 400 to 500 ms pulse the electrodes start to melt but there isn't sufficient melting of the nickel strips to the battery. I've tried using thicker electrodes but that doesn't seem to help. With a 500+ms pulse it's putting more heat into the battery than I would like without sufficient welding of the nickel strip to the battery.

Do I need to find a bigger transformer for more amps or is there something else I'm doing wrong?
 

·
Registered
Joined
·
532 Posts
Your spot welder needs to be a capacitor discharge, not a resistance welder. Need to dump thousands of amps quickly into a small spot area versus hundreds of amps slowly in other words.

Lithium batteries make for nice fireworks if you abuse them, so hopefully you're not doing any of this in your mom's basement.
 

·
Registered
Joined
·
911 Posts
Just buy the proper DIY spot welder, designed specifically for this purpose.

Endless-Sphere is the place to read past threads and become close to an expert before it even arrives.

Some things it's better to trod the well-worn path

if your goals are practical, rather than having fun trial and error tinkering.
 

·
Registered
Joined
·
4 Posts
This may not be the appropriate forum for this but I would appreciate any assistance anyone would be willing to offer. I am trying to spot weld a pack out of 26650 cells. I built my own spot welder using the microwave transformer and some 2 ga. wire. with a timing relay to control the pulse. New to spot welding but I know a little about arc welding. From what I gather I suspect that I do not have enough current. When I get up to about a 400 to 500 ms pulse the electrodes start to melt but there isn't sufficient melting of the nickel strips to the battery. I've tried using thicker electrodes but that doesn't seem to help. With a 500+ms pulse it's putting more heat into the battery than I would like without sufficient welding of the nickel strip to the battery.

Do I need to find a bigger transformer for more amps or is there something else I'm doing wrong?

What about something like this?

 

·
Registered
Joined
·
21 Posts
I built one like in this Utube video a couple years ago to make a pack of "c" size nicads for a tool. It worked great!
 

·
Super Moderator
Joined
·
1,342 Posts
I built my own spot welder using the microwave transformer and some 2 ga. wire. with a timing relay to control the pulse.
How many turns did you put on the secondary?

I'd say 2 gauge wire isn't thick enough.

Take a picture of your setup.

An MOT-based spot welder is definitely possible, don't listen to anyone who says you need to throw it away and do capative-discharge, they're wrong. People have built thousands of these and they work just fine.

Random examples:



It's a bit ghetto, but you could also just use your spot welder like a fast, high power soldering iron. Put a tiny coil of solder under the nickle strip, and then pulse the welder while applying pressure, it makes an excellent soldered connection. Some poohpooh soldered cells, but, Tesla packs are soldered.

From what I gather I suspect that I do not have enough current.
Probably. Spot welding also requires significant clamping pressure, unlike capacitive-discharge welding. Especially with heavier nickle strips you should be using (versus what you'll find on laptops), you need LOTS of clamping pressure.

Or, it's possible you have too much. But, probably too little.

Car bodies are spot welded together, so it's not a limitation of the technology. 1000 watts should be sufficient to do what you're trying to do.

Here's a video showing some good or bad techniques:


Do I need to find a bigger transformer for more amps or is there something else I'm doing wrong?
Are you using an especially small MOT?

Show a picture of your setup, I can probably suggest what to improve.
 

·
Registered
Joined
·
911 Posts
To say that the way Tesla fixes their hair-thin fusible links to cells,

with purpose-designed bespoke gear costing multiple millions,

has anything to do with the topic of this thread is. . .

I'm speechless
 

·
Registered
Joined
·
6 Posts
Discussion Starter · #8 ·
So I did some digging thru the trash to find the remnants of the microwave and it looks like I have a 1.2kW transformer so I should be able to get about 1000A out.

So here is my setup:

I believe this would technically be 2 wraps for the secondary? I talked to my EE friend who suggested taking a half wrap out. I get about 1.4V on the secondary.

I have pure nickel strip that is 0.2mm thick. This is the best I can do before I am uncomfortable putting too much heat into the cell. This was using about 0.5 sec pulse which barely gets it to stick to the cell. Much more than 0.5 second pulse and I melt my electrodes!


I'm going to try to measure the supply current and multiply by the voltage multiplier to estimate what kind of current I am getting out. I suspect this transformer may not be the healthiest.
 

·
Registered
Joined
·
532 Posts
It's a bit ghetto, but you could also just use your spot welder like a fast, high power soldering iron. Put a tiny coil of solder under the nickle strip, and then pulse the welder while applying pressure, it makes an excellent soldered connection. Some poohpooh soldered cells, but, Tesla packs are soldered.
😂 What on earth gave you the idea that Tesla modules' wire bonds are soldered? They are NOT.

Jehu Garcia does it, but the stuff he does where EV West is not the one doing the real work is, quite often, (not a "bit", but quite) ghetto....which is ok for an e-bike.

Probably. Spot welding also requires significant clamping pressure, unlike capacitive-discharge welding. Especially with heavier nickle strips you should be using (versus what you'll find on laptops), you need LOTS of clamping pressure.

Or, it's possible you have too much. But, probably too little.

Car bodies are spot welded together, so it's not a limitation of the technology. 1000 watts should be sufficient to do what you're trying to do.
Yeah, car bodies are spot welded...I just dropped 25 large on a 13,000 amp spot welder to put my Humpty Dumpty Model X back together again. With spot welding guns that squeeze down at a quarter to half a ton.

Buy the right tool for the job. It's one thing to make a pack for an e-bike...quite another making one for a car. The cells you plan to use have a polymer spacer under the positive tab...last thing you want to do is cook it with the microwave you pilfered from your mom's kitchen.

What's going on in this thread is a false economy and enthusiasm fueled by bull$hit. Buy the tool john61ct suggested and don't trash $6,000, or whatever, in cells.
 

·
Super Moderator
Joined
·
1,342 Posts
So I did some digging thru the trash to find the remnants of the microwave and it looks like I have a 1.2kW transformer so I should be able to get about 1000A out.
That's a larger MOT. It should be sufficient.

So here is my setup:
Right off the bat, I notice that you're only using perhaps 20% of the space available to you in the core, after you account for insulation and the airgap. So, I'd say that's probably your weak point. Whether that's worth fixing or not, is up to you. You might get away with a monolithic single turn secondary, but your losses might leave you with too low a voltage to be useful. 2 turns is a necessary evil sometimes.

Because you're only dealing with a volt or two, you don't need the heavy insulation. I would strip it off, and then just heatshrink or kapton-tape wrap it afterwards, with the goal of using 2x as much wire. Doesn't have to be pretty, hammer it into shape before you wrap it.

Another thing you could try is knocking out the magnetic shunts that separate the primary from the secondary. The result will be that you run a bigger risk of blowing a breaker (not too likely for your duty cycle), because it'll pull more power.

Another thing you could try is adding some more turns to the primary, if you had some enamelled 14g copper. Coil in the same direction as the existing coil. It'll cut down on your losses significantly.

MOTs are built as cheap and terrible as possible, cutting every corner.

I believe this would technically be 2 wraps for the secondary? I talked to my EE friend who suggested taking a half wrap out. I get about 1.4V on the secondary.
I get his logic, but I think that would be detrimental.

You have a max amount of power in the transformer.

Your electrical resistance is nearly enough zero that an electrical short is not the limitation of your welding current.

I.E. I = V / R ... with an infinitesimal R, gives you something like 40,000 amps even on a couple volts. You won't, because you run into power limits long before resistance limits.

And, since spot welding depends on the amount of amps you can make flow, and the amps are proportional to the turns ratio between the primary and secondary, you could get more amps if you doubled your ratio by halving the number of secondary turns (or a half wrap, or whatever).

But...

- Pure resistance isn't the only thing in play.
- There is a lower limit on the amount of voltage or you'll struggle to make power flow.
- This presumes you'd fill the core to a higher degree with more cross section of copper (i.e. if you put one turn of 14g wire as your secondary, that is not going to improve performance).

Depends on the setup, sometimes adding turns will work better, especially if your clamping pressure is low.

I'm going to try to measure the supply current and multiply by the voltage multiplier to estimate what kind of current I am getting out. I suspect this transformer may not be the healthiest.
I bet it's fine, there's not really anything that can go wrong except a melted wire (which should either short or break entirely, not a lot of in between), but that's good to rule out.

For something fun, test the no-load current too (transformer on but not welding). It'll be abysmal like 7 amps.

...

Judging by your setup, my guess is you don't have enough clamping pressure for how low of a voltage you're using.
 
1 - 12 of 12 Posts
Top