Battery tag welder from China

[robrenz]

I don't know for sure either, just thinking out loud.  Tungstens melting temp is around 6300F so I thought it would probably not alloy with the nickel at weld temp.  Brass will wet to other metals at a relatively low temp as in Brazing.

Take a read of my Sunstone links above. In a spot weld the metal is only supposed to melt at the contact surface of the join. It shouldn't melt where it contacts the electrodes.

Understood, but when you spot weld two .003" thick pieces of metal together do you think the melt can be controled to not reach the surface that is touching the electrode?

[IanB]

Understood, but when you spot weld two .003" thick pieces of metal together do you think the melt can be controled to not reach the surface that is touching the electrode?

I think maybe it can be so controlled, but it requires very precise control of the weld parameters: energy, maximum current, pulse shape, pulse timing, etc. I think a good weld requires about the same amount of precision that you demonstrate in your machining videos... 

[robrenz]

Thanks, yes probably doable but you will need something with more finess than a $200 chinese unit

[nctnico]

I don't know for sure either, just thinking out loud.  Tungstens melting temp is around 6300F so I thought it would probably not alloy with the nickel at weld temp.  Brass will wet to other metals at a relatively low temp as in Brazing.

Take a read of my Sunstone links above. In a spot weld the metal is only supposed to melt at the contact surface of the join. It shouldn't melt where it contacts the electrodes.

Understood, but when you spot weld two .003" thick pieces of metal together do you think the melt can be controled to not reach the surface that is touching the electrode?

If you have precise control over the welding parameters you can make excellent spot welds. I have worked on spot welding control systems which keep the weld invisible on one side for manufacturing cars. Not hobby stuff by far though.

[mzzj]

[/quote]

I did TIG welding myself but I am not sure what the characteristics are with a tungsten electrodes, it could stick on the batteries with high temperature and the arc, brass or copper on the other hand wont melt in to another metal, that's why welders use copper plate if they weld two pieces and things get to hot on a metal work table or in a narrow spaces.
But since you are the real mechanist, I stand corrected if I am wrong  (so... covered my arse if i am wrong)
[/quote]

Copper melts to other metals as well, they even use copper plate/shim for serious high current battery packs. Obviously spot welding copper is royal pain, but doable with right equipment.

I was less than satisfied with pure copper electrodes, too soft and tends to stick to workpiece. Copper-molybdenium and copper-tungsten should be better alternatives but availability is not that great..

Next I was planning to try 4mm pure tungsten TIG electrodes.

Another way around capacitor problem is to use 350v caps coupled trough transformer. Much easier to get required energy this way than  with 12v caps. 350 to 450 volt inverter electrolytics are much easier to find and biggest ones put out serious punch. Also easier for your thyristor/igbt.

Transformer coupling also has its problems, like redesign the windings for short duty pulse and if you use toroid transformer you probably have take care of the core (pre-)demagnetizing before each shot.

[ftransform]

How about using a hazard fraught portable spot welder and making electrode holders to go where the tongs go. Then use controled timing of the on time of the primary AC input to control the weld pulse duration?

or some kind of fast relay? Modding a real welder would be highly favorable

[robrenz]

Solid state relay with flywheel diode with 555 timer for pulse length?

[nukie]

EDIT: Apparently this is no good for materials other than steel. So it isn't very helpful.

This is nice but limited adjustability
http://www.robbe.de/welma-2000-punktschweissgeraet.html

[Marco]

Those big spot welders do work with transformers, but I doubt the ones for the battery tabs do ... a cheap electrolytic capacitor can disintegrate a screwdriver, I think it can make a spot weld on a battery tab.

[mzzj]

How about using a hazard fraught portable spot welder and making electrode holders to go where the tongs go. Then use controled timing of the on time of the primary AC input to control the weld pulse duration?

or some kind of fast relay? Modding a real welder would be highly favorable

Required power levels make direct ac powered battery spot welders impractical. We are roughly speaking about 1 to 10 ms pulse duration with 200-400 Joules of energy. Even the low end gives 20 kW of pulsed power and upper end is 400 kW ! 

[robrenz]

IMHO the portable spot welder will definitely work. It is meant to have intermittent primary energization as that is how you control the spot weld time. It uses a finger controled switch on the primary already so putting a SSR with pulse timer on the primary is not a big deal as long as you don't violate the 50% duty cycle which should not be an issue for sub second weld times. The secondary current with 6" long tongs is 5500 amps and would increase if shorter adapter bars made to hold the small electrodes were used.  As you can see the secondary current is affected by the tong resistance.  To get optimum current for battery tabs It may require using even more resistance than the 18" long tongs. In that case welding cable could be used with the optimum length/Ga to get a hand held probe unit that would be easier to manipulate and have the right current.  Most quality spot welders use a two pulse setup with one light pulse to clean/establish good contact then the actual weld pulse which could be impimented here also.
The weld shown is .006" steel shim stock and the weld time was me just bumping the finger switch as fast as I could.  You can see by the peel test that these are good welds and nothing about this setup is optimal.  Remeber this thing is capable of welding two pieces of .062" steel easily with 12" tongs so I dont think power is an issue.

Edit:
At 42 µOhms per 12" of the .625" diameter copper the current drops 950A per 6" change in tong length (2 tongs x 6" = 12") so it will be easy to drop to even lower currents if that makes the weld easier to control.  But I think a very short higher energy pulse is more desirable as there will be less heat affected zone and heating of the battery in general.

[Marco]

It will certainly work, question is whether it's economical or smart ... the copper in a mains frequency 1.5 kW transformer and the electrodes is probably worth more than the entire machine the OP bought. You still of course have to remove those ridiculous electrodes from the spotwelder as well and bodge something like the pressure activated electrode system from the OP's machine.

For the price OP's machine makes perfect sense to just outright buy ... if you're determined to DIY it's a big enough project mechanically that you can probably put a little more effort into the electronics as well (charging a capacitor to a variable voltage and discharging it through a thyristor and an aircore coil isn't that much harder than modding a spot welder).

[robrenz]

I am not saying what the OP has purchased is bad. I really hope it works and If so it is probably the most economical way to get a tab welding system. 

I was simply suggesting a much simpler approach than the DIY capacitor discharge units mentioned in this thread.  All of the ones I have studied on the web are monumental projects with lots of tweaking and a huge amount of time and money and smoked components involved. I paid $100.00 for the Dayton unit pictured and the Harbor Freight unit is only $160.00 in the US. Add some weld cable, electrodes, SSR with simple pulse timer and foot switch and you have a funtional unit.  All the DIY units I have seen apply pressure to the two electrodes by hand.

[Marco]

Here's a pretty simple one (if the SCR is big enough the inductance of the welding cable is enough to limit dI/dt I guess, with an inductor you could use higher voltage electrolytics though).

http://www.philpem.me.uk/elec/welder/
http://www.rcgroups.com/forums/showthread.php?t=763308

[sorin]

how about using mosfet??

[Flávio V]

Would medium sized super caps(+-500F) be good for the battery welding?

(i have 2 maxwell 350F ones so well...interested)

[IanB]

Would medium sized super caps(+-500F) be good for the battery welding?

(i have 2 maxwell 350F ones so well...interested)

It's not about power, it's about control. For a battery tab weld you need a small amount of energy that is controlled with great precision.

The aim is to make a tiny little weld, not to explode the battery into a glowing fireball 

[Psi]

I've tried welding battery tabs with some maxwell caps and couldn't get it to work reliably.

I think you need PWM control of the current to make it work.

However, i didnt have those battery tabs with the slot cut in them to force the current through the battery metal which would probably make it easier to do.

[robrenz]

In looking at the sunstone units 100 watt seconds or joules is more than enough to do tabs.

The spot welder above is 1.5kVA so it is roughly capable of 1500 watt seconds.

[Marco]

Sorin, a 5$ thyristor can handle a kA surge current and is trivial to drive (you just have to be a little careful with dI/dt). With MOSFETs you need a more complex drive and much much more silicon (IGBTs too, although cheaper than MOSFETs).

Flavio, the balance of voltage/capacitance/ESR for a supercap just doesn't work. If you want high currents you need to charge it to near it's max voltage, at near it's max voltage it takes way way too long to discharge when you short it through the battery tab (destroying the cap in the process). They could work with MOSFETs or IGBTs I guess, 800 amp short circuit current will be able to get things hot, but as I said Thyristors are cheaper and easier so better to use electrolytic capacitors.

[sorin]

5$ thyristor can handle a kA surge current

50RIA20 cost $15

[Psi]

the balance of voltage/capacitance/ESR for a supercap just doesn't work. If you want high currents you need to charge it to near it's max voltage, at near it's max voltage it takes way way too long to discharge when you short it through the battery tab (destroying the cap in the process). They could work with MOSFETs or IGBTs I guess, 800 amp short circuit current will be able to get things hot, but as I said Thyristors are cheaper and easier so better to use electrolytic capacitors.

The maxwell caps i tried to use are rated to 600A max discharge.
I tried various combinations of 9 caps (series/parallel) with the wire length/gauge to the weld point chosen to limit shortcircuit current to 600A / 1200A / 1800A (parallel caps)

[Marco]

50RIA20 cost $15

This one isn't though :

http://www.digikey.nl/product-detail/en/S4065KTP/S4065KTP-ND/967811

[robrenz]

After some more research on the Sunstone welder site:
They make transformer style welders also and 1.5kVA is the rating of their smaller unit and the duty cycle is the same at 50%. The weight of my spot welder is 28 lbs and their unit is 31 so that means the spot welder I have should be a good match (similar weight in copper and laminations). They fire the primary in whole multiples of line cycles from 1 to 100 and also have multi pulse modes.

Tungsten is the higher performance material for electrodes as opposed to copper.

I won't post any more on this thread I will start a project thread when I get the time.

[firewalker]

Nice! Enjoy it! I am also expecting waveforms! 

Alexander.