Build your own CD battery tab welder for about $100.00+-

[pwbset]

GL with your project, but I'd say you should at least bulk up the wires.

Using 8awg "super flex" attached to the 4ga solid 'trodes, but they are like 5' long each side... could trim that to like 2' each I think... would trimming 6' off really help that much? Hmm...

 

[terramir]

Using 8awg "super flex" attached to the 4ga solid 'trodes, but they are like 5' long each side... could trim that to like 2' each I think... would trimming 6' off really help that much? Hmm...

No, no, no not a good idea, you need some room to work with, but I'd think when you have a chance upgrade to 4 gauge wire.
for now just try as is you can tweak later, albeit I'd still say try to somehow drop the charging to 16V. 16.8 is risky
my
$.02
terramir

 

[oldusedbear]

I've been reading the mail here for a while, but this is my first post. New kid on the block.

Hopefully, I can build a successful welder for my limited objective - - tabs on sub-C's for rechargeable tool battery packs. (I do have a store bought e-bike tho - - and had a DIY electric scooter circa 1943 - - during the BIG WAR that was in all the papers. Hope this is sufficient to gain entrance to the forum!)

To date, I've put most of my effort into the mechanical platform. For this I've purchased a small arbor press (about 50 bucks) and am in the process of adding the electrode mechanism. Hopefully, I can figure out how to append some pictures to this post.

After reading more, I'm already concerned that my electode mounting system may be inadequate. I can individually adjust the length that the electrodes protrude from their holders, but once adjusted, they are rigid with respect to each other. Now, I'm wondering if I'm gonna have problems with getting equal, or close to equal pressure between the individual electrodes and the work. Additionally, I'm concerned that even if I could get the initial adjustment good enough, will it stay that way for more than a few welds?

If this is going to be a big problem with my design, I've thought I could mount my present electrode carrier to some sort of spring loaded platform that would allow the electrodes to "self adjust" and equalize the pressure at each tip. Hopefully, some of you more experienced types could comment on this.

 

[Jeremy Harris]

Very nice idea to use an arbor press, wish I'd thought of it. Nice build quality, too.

I made a press for my welder and found that I needed a way to get even pressure on the two electrodes - it was very hard to keep the tolerances close enough for even pressure otherwise. Here's my welder: http://endless-sphere.com/forums/viewtopic.php?f=2&t=2633&start=165#p75859

I used rubber sheet as a simple means of applying some resilience to the electrodes, which worked pretty well.

As a footnote, I eventually took the frame off my welder and reverted to hand-held electrodes, as I found them quicker and easier to use.

Jeremy

 

[oldusedbear]

Many thanks, Jeremy. Your mention of the rubber sheet has fostered a simple idea: Perhaps all I need to do is enlarge the four holes in my electrode mounting plate. Then fit each mounting screw with a rubber sleeve (could be a very short length of rubber tubing), then change out my present mounting screws to a larger diameter truss head. The screws can be left a little loose, and locked on the back side with a nylon lock nut. This should allow just enough freedom of movement for the whole electrode assembly to do a bit of self alignment when pressure is applied for the weld. Hopefully both simple and effective.

For the dedicated DIY types here, I might add that the black plastic parts are made of ABS. For about 20 bucks, you can by a 12 by 48 inch piece of this - - 1/4 inch thick. It is highly polished on one side, and a sort of wrinkle finish on the other. Neat stuff to work with as it can be sawed, drilled, tapped etc. Good insulator too. Best of all you can make really strong solvent welds with ordinary ABS cement as used in plumbing. Really easy to make different shapes with a band saw or jig saw. Circular saws and chop saws create too much heat and tend to make a gummy mess.

BTW, I had initially considered the fabrication of some sort of toggle system to raise and lower the electrodes. When I discovered how cheaply the arbor press could be obtained, I opted for that in deference to the greater range of motion, and a lot less work.

 

[terramir]

I have an idea that might help but it would require a bit of re-machine-ing the holders and the probes. you could make them springloaded that way the probes would exert constant pressure regardless of the wear in the probe tips. the pressure would be equivalent to the springs. you'd just have to thin the tips a little bit and insert a spring and put a little plate on the tip of the holders with a little bit smaller hole.
just an idea, if your not comphrending what I'm saying let me know and I will make a drawing.
terramir
BTW how cheaply did you get that arbor press I'd think I could use one LOL

 

[oldusedbear]

I had thought about spring loading the whole plate that carries both electrodes, as well as spring loading the individual electrode holders. Jeremy's mention of using rubber got me thinking in THAT direction however, because it is so simple.

I had already thought about a testing scheme that is in concept, similar to the way a Brinell hardness tester works. Using some yet to be determined target material, you simply "make dents" with the electodes (no current involved) in the material of choice. Then you examine the material under low magnification to see if the dents look the same. Confess I don't know if that is a valid approach. It would presume that the electrode tips were VERY similar in profile.

In any event, I'm sure a fair amount of testing is in order. I've never seen one of the commercial systems other than pictures online. Clueless as to how they handle this problem.

The arbor press is a Palmgren - - Got it thru Amazon and it was 56 dollars with them paying the freight. I'm guessing it could weigh 15 to 20 pounds, so the free freight is a real factor.

 

[terramir]

little hint many commercial systems have two types of electrodes, both sharp or one flat one sharp. Obviously the flat one is so you make one weld at a time even if you have to press both down on the tab material. Now with our ghetto systems we can achive the same thing by using the side of the conical shape. Albeit this version 3.1 I'm building will be far from ghetto. 3.0 is the test setup I'm planning on using in the next few days, but 3.1 will have a comparitor to make it charge to a predetermined level. So I'll know how many Ws I'm pouring into the weld. Also I will be using hand probes for now, 4.0 might have a press welder attachment .

One thing I really love about your setup is the thinkness of the cables running into your probes, talk about cutting down on losses.
My system will have one thing I have never seen on any of the systems, at least for a try a big 4 gauge air coil in order to protect the scr (thyristor) from on surge current, I designed it to limit the current rise to 250A/us max While I doubt I'll achieve such a rush with 1.22F-1.58F I thought it would be nice to secure my 400 buck original price thyristor(got it for 30 bucks off eBay) dual pack. Anyways I thought spring loading would be a bit neater than rubber.
my $.02
terramir

 

[oldusedbear]

That is 1 AWG welding cable - - Nice and flexible. You've no doubt noticed it is still a one piece loop - - Thought I'd wait to see how I'll be situating the caps and the SCR before cutting the cable.

As regards the electrode pressure question: You may very well be correct, Terramir - - The rubber grommet idea may not get the job done, and I will quite possibly have to go to real springs. For starters tho, I've thought I would cut my present mounting plate vertically - - So the electrodes could move independently from each other. Then I was gonna drill out the mounting holes and install the grommets. If that affords sufficient "rubber loaded" movement of each electrode, it might work. I'm guessing that I won't need a whole lot of travel for the tip pressures to equalize - - or at least come close to equalization.

My next biggest challenge at the moment is the selection of caps and a suitable SCR/thyristor. I've been hopeful of seeing some consensus on which brands of audio caps are a good value. It would be neat if someone could simply say "Buy two of the XXX caps and parallel them for a total of 3 Farads (or whatever I am going to need) and they are about 40 bucks each from YYY." The same sort of advice for the SCR would also be a huge help. I've arbitrarily budgeted about 75 bucks for the caps, and 50 bucks for the SCR. Those numbers are only WAGS tho, and if there are tiebreaker reasons for spending more, I would sure entertain them. It seems that every time someone mentions some hot deal on eBay, it is long since gone by the time I go looking there. I guess I'd rather pay a bit more (if necessary) and go to someone who has more than just one to sell. Input from anyone on selecting these items would be REALLY appreciated.

I'm guessing that maybe three (real) Farads should be enough. PLEASE share your thoughts on that. Additionally, the plan is to use a 12 amp battery charger for my power supply. I also have a good sized Variac I can put ahead of the charger to control/limit the output voltage. My (planned) foot switch will have two microswitches, which I'll configure to switch sequentially. The first one will open a relay that interupts power to the battery charger, and the second one will then fire the SCR.

I'm shamelessly begging for all the advice I can get on the above. Hopefully, my limited application (tabs on sub-C nicads), will simplify the complexity of my needs. I'm trusting that I don't need dual pulse, controlled pulse width etc., for my less sophisticated expectations.

 

[terramir]

hmmmmm, with the caps I really can't help ya cause I got special caps (two main caps out of hybrid caps I basically blackmailed (complained about their 2 farad cap being .6F max) and I went to the office to do it) so they gave me a whole 10F big shiny thing (which was like 5 farads total really but after .6 to .8F the esr rises tremendously (supercaps 6 25F ones in series) and they gave me a spare main cap so I got over a real farad now , but here's the deal you can get a way oversized (which can be a good thing for about 50 bucks from the electrostore.com (there are two inside the pack but the way there configured you can only use one at a time, still trying to figure out how to open it to flip one around for a prezap, welder (but that be version 5.0 LOL) here's the link
http://theelectrostore.com/shopsite_sc/store/html/skkt210-20e-skkt-210-20e-skkt21020e-scr-semikron-used.html As for caps you need at least one farad, preferably a bit more, going to 3Farads is overkill unless your trying to weld A123's. I don't have much experiance but stay away from absolute 2F ones It's about .6F and hybrid caps are useless, unless your just jacking the main cap out of it.
LOL
terramir

 

[Takemehome]

Just a suggestion:
If your going with springs, you could put 2 micro switches in serial to trigger the SCR instead of a foot switch. When enough pressure is applied on both probes, perfect weld. And since micro switches have NC NO contacts, when the pressure is released caps are back on charge.

 

[terramir]

Just a suggestion:
If your going with springs, you could put 2 micro switches in serial to trigger the SCR instead of a foot switch. When enough pressure is applied on both probes, perfect weld. And since micro switches have NC NO contacts, when the pressure is released caps are back on charge.

Ehh I would think that is a rad idea except that microswitches will get burned out quite fast. at at max voltage quite a bit of current that is forced open from the NC position. Unless the microswitch is rated for a minimum of 3 to 4 the current level that the resistor limits it to at max voltage the life span of the contacts will be very short. and since microswitches usually come in 5 or 10 amp current ratings you'd better limit yourself to 3.3 to 2.5amp max charging rate.
a 40 amp car relay holds up fine because it's rated so much higher, but microswitches would cause problems, cause they will inevitiable cause a higher contact voltage drop than a car relay. But that problem could be solved by using a transistor based solution to switch a car relay then the micro switches wouldn't have to carry any load.

My $.02
terramir

 

[oldusedbear]

Well - - My easily derailed train of thought involved a mixture of what both of you are saying. I was thinking of a substantial (15 Amp or so) AC relay on the input side of my battery charger. Using a 5 amp NC microswitch to open the relay contacts when the foot pedal is pushed. When I just LOOK at a microswitch, I can never believe it will sustain the current rating claimed.

As a future improvement (in place of the foot switch), I was considering a small working platform on which the battery pack would sit, and within the platform I'd have the microswitches. Then figured I could support the platform with changeable springs that would determine the electrode pressure when the second microswitch closed. Spring strength to be chosen/adjusted to optimize the weld. For starters tho, I figured it might be best to minimize the number of variables by using the foot switch.

Thanks for the SCR source info, Terramir. That looks like a substantial component. Maybe even if you can't do the Siamese separation thing, you at least have a backup if you toast half of it?

Now if someone can recommend an appropriate brand of audio capacitor or capacitors? I'm guessing that 3 Farads would be ample but could sure use advice on that too!

After considerable rumination, and staring at my electrode mounting system, I'm going to try the thing of separating the mount so each electrode can move independently. Then, enlarge the holes for the mounting screws, and use a little bitty O ring between the screw and the mount - - Hoping this will give me just a little bit of "give" that will allow the tip pressure to equalize. If that doesn't work, I can easily morph the design to a more conventional spring loaded setup. No question but what regular springs would afford a greater range of movement.

 

[terramir]

Well - - My easily derailed train of thought involved a mixture of what both of you are saying. I was thinking of a substantial (15 Amp or so) AC relay on the input side of my battery charger. Using a 5 amp NC microswitch to open the relay contacts when the foot pedal is pushed. When I just LOOK at a microswitch, I can never believe it will sustain the current rating claimed.

You might want to go with a car relay setup instead of an AC relay the low voltages that we use are unlikely to need the arc protection that an Ac relay uses, however the current protection of a 40amp 12V car relay is quite a bit better, and those relay's are cheaper as well I'd recommend bosch relays there quite sturdy and easy to find at autozone, napa, pepboys, kragen or whatever their called at your neck of the woods.

As a future improvement (in place of the foot switch), I was considering a small working platform on which the battery pack would sit, and within the platform I'd have the microswitches. Then figured I could support the platform with changeable springs that would determine the electrode pressure when the second microswitch closed. Spring strength to be chosen/adjusted to optimize the weld. For starters tho, I figured it might be best to minimize the number of variables by using the foot switch.

That might be a good idea if your gonna make several packs and using a spring to trigger the microswitch means that you can use several types of batteries with the setup. But for now get the thing setup and use a footswitch so you can learn.

Thanks for the SCR source info, Terramir. That looks like a substantial component. Maybe even if you can't do the Siamese separation thing, you at least have a backup if you toast half of it?

True true the way it's wired I could use either one but not both together but I dun think I'll be able to fry this part unless I go crazy on the voltage. the contact resistance and the coil I'm gonna use prevents that from happening.

Now if someone can recommend an appropriate brand of audio capacitor or capacitors? I'm guessing that 3 Farads would be ample but could sure use advice on that too!

I wish someone would answer you there, But I can tell you the best approach might be to go with car caps and go local, find a car stereo place that offers various brands, buy one test the capacity complain try another brand. All brands will be significantly lower than the advertized capacity. Stay away from the hybrid capacitors though they have super caps inside and while the advertized esr will be low it would only apply to max the first farad (i.e. the main cap).

After considerable rumination, and staring at my electrode mounting system, I'm going to try the thing of separating the mount so each electrode can move independently. Then, enlarge the holes for the mounting screws, and use a little bitty O ring between the screw and the mount - - Hoping this will give me just a little bit of "give" that will allow the tip pressure to equalize. If that doesn't work, I can easily morph the design to a more conventional spring loaded setup. No question but what regular springs would afford a greater range of movement.

Springs are availible at hardware stores, and usually the best selection will be at the mom and pop stores, how depot and such usually don't have a very good selection in springs.
Hope this helps
terramir

 

[liveforphysics]

I'm just curious, has anyone tried just using a TIG welder to put the tabs on cells? I imagine I could just dial it up to 150-200amps, get the tungsten in the right spot, and blip the current pedal for a split second.

Or... Maybe slide in a round copper rod rather than the tungsten electrode, dial up the current, and just tap it against the strip.

Is the time interval too long with a method like that? Inadquate current intensity to make the weld and not heat the cell? Anyone tried it?

 

[terramir]

I'm just curious, has anyone tried just using a TIG welder to put the tabs on cells? I imagine I could just dial it up to 150-200amps, get the tungsten in the right spot, and blip the current pedal for a split second.

Or... Maybe slide in a round copper rod rather than the tungsten electrode, dial up the current, and just tap it against the strip.

Is the time interval too long with a method like that? Inadquate current intensity to make the weld and not heat the cell? Anyone tried it?

that would make for some cool blow-ups, spectacular I might say LOL That is an arc welding setup, and one of the problems is that the surface area is so small you couldn't put the ground contact on there. while the welder probably could be modified to work, i.e. eliminate the gas replace the tungsten and the ground wires with electrodes you'd basically be back to the same set-up, with two problems too much wattage overall you'd probably turn the two electrodes into a mini jacob's ladder. and B we want a short high amp low power weld here. 125ws to 250ws are pretty much the max, the reason that even welds is because the actual welding period is in the us to lower ms range, so the overall power is not enough to harm the cell, but the pulse is short and high current enough to melt the metal of the tab and either the cathode or anode of the battery to join. your foot can't tap that short.
my $.02
terramir

 

[oldusedbear]

GREAT reply, Terramir. Thanks for going into the details for me.

I ordered the SCR you suggested. Looks like it should have plenty of headroom for my piddly application.

We live in a little podunk coastal town. THE auto stereo store (as in quantity of ONE - - and not very big) stocks 1 cap - - can't remember the brand, but it was about 80 bucks and I think I could get it online for maybe half that amount. I believe I've seen more than one reference to the Volfenhag (sp?) caps on this forum. Wonder if they are OK?? Somehow I'm willing to bet that they don't speak Deutsch in the factory where they're built - - little bit of a marketing ploy there perhaps? Still hoping someone here will respond with THE WORD on the caps. Hopefully some brand they have bought, tested and are using. Ah well.

You're right about the automotive relays - - widely used and cheap. Even for stuff rated at 60 amps or so. Are you suggesting I should switch the DC output of the battery charger? I was thinking of switching the AC input - - nothing sacred about that approach tho - - It is about a 12 amp charger so the car relay would work well I should think.

I'm blessed with spring assortments luckily. In a former life, I owned a little Ace Hardware store. Took a substantial whack of the inventory with me when I sold the store. Wish now I'd taken even more!

Your comments on the foot switch are well taken. Learn first, fine tune later. I see the potential for triggering the weld accidently during the electrode positioning process.

Thanks again for all the input.

 

[rkosiorek]

I have 4 X NOS - Cornell Dublier # DCM105M020DP5D 1,000,000 uF 20VDC capacitors.




using a 1k resistor in series with a 20V supply it takes about 18 to 20 minutes to charge each up to 13.5V. so they are pretty close if not better than the Farad rating. catalog lists the ESR as .003. my ESR meter only goes down to .01 and these read either 0.01 or 0.02R. doing the 5ohm resistor discharge from 10V down to 5V gets a reading of .005 to .010V. so i'm sure the ESR is still plenty good.

3 of these are in very good condition and one has a couple of minor dings. not sure how that affects it. connection is via a couple of 1/4 - 28 screws.

on to the mercenary part. these list at $150.00 to $180.00 each. and they are special order items. no one stocks them (digikey has them for $500.00 each, minimum oreder of 20. but that is just Digikeys way of suggesting you order them from someone else.

they are for sale. $50.00 each +shipping. the dented one is on sale for $20.00 PM me if interested.

also posted in the 4 sale forum. only did it here as well because someone asked.

rick

 

[terramir]

GREAT reply, Terramir. Thanks for going into the details for me.

I ordered the SCR you suggested. Looks like it should have plenty of headroom for my piddly application.

We live in a little podunk coastal town. THE auto stereo store (as in quantity of ONE - - and not very big) stocks 1 cap - - can't remember the brand, but it was about 80 bucks and I think I could get it online for maybe half that amount. I believe I've seen more than one reference to the Volfenhag (sp?) caps on this forum. Wonder if they are OK?? Somehow I'm willing to bet that they don't speak Deutsch in the factory where they're built - - little bit of a marketing ploy there perhaps? Still hoping someone here will respond with THE WORD on the caps. Hopefully some brand they have bought, tested and are using. Ah well.

If you can afford the volfenhag 1.5F x2 you'll get around 2 farads, what the other guy posted probably be better though. from what I've read (and BTW I speak german lol (but born in usa) live in Lost Angeles now). you can probably use those at 19.5V all the time 20 should have a surge rating of 24 or 25, the data sheet didn't mention those particular ones, so 19.5 @ 2F would be like 380.25 Ws which should be more than enough you'll probably end up using them mostly at less than 16V unless you end up making a A 123 battery pack (aluminum welding takes a little more juice).

You're right about the automotive relays - - widely used and cheap. Even for stuff rated at 60 amps or so. Are you suggesting I should switch the DC output of the battery charger? I was thinking of switching the AC input - - nothing sacred about that approach tho - - It is about a 12 amp charger so the car relay would work well I should think.

Yes switch the dc side and protect the psu from any accidents, also I dun think a psu would appriciate being turned on and off on the ac side in rapid succession.

I'm blessed with spring assortments luckily. In a former life, I owned a little Ace Hardware store. Took a substantial whack of the inventory with me when I sold the store. Wish now I'd taken even more!

Your comments on the foot switch are well taken. Learn first, fine tune later. I see the potential for triggering the weld accidently during the electrode positioning process.

Thanks again for all the input.

I'm jelous I have bought out my local ace of the double conical springs #8, and the company now has a new supplier so there ugly not crome and not really conical anymore can't find a single hardware store that stocks the brand . (they have a board on a wall that runs back from another company that got bought up .
Hey no problem I understand the theory of this stuff quite well, and tab welding is easy, I suck at other welding though, today I welded a new pipe to my hatch on my utility body all popcorn looking and such well as long as it holds and well I got enough bond for that, but took forever, bondo will smooth out the weld before I paint over it LOL.
Glad to help
terramir

 

[rkosiorek]

sorry about the Data sheet. when i emailed Cornell Dublier that was the Data Sheet they linked me to for the specs of the capacitor. they also gave me the following break down of the part #
DCMC/DCM = 85 ºC High Capacitance, Screw Terminal, Aluminum Electrolytic.
105 = Capacitance = 1,000,000 uF
M = Tolerance = +/- 20%
020 = WVDC = 20V
DP = Case size = 3.0" diameter X 5.85" high
High Current Low Post
post spacing 1.25"
5 = PVC - Black insulation
D = Terminals = 0.625" diameter, threaded for 1/4 - 28 stud with thermal pad.

the energy these things can deliver is impressive. i accidentaly found out that a 5V discharge will cause a solid copper 16awg wire to vaporize.

rick

 

[rkosiorek]

i was thinking of how to make the length of the output pulse adjustable and how to turn off the SCR in mid-pulse. this method was in an ON Semi APP-NOTE



- in the initial state both SCR1 and SCR2 are off. no current will flow through C1 and R1.
- using a short trigger pulse (~10mS) SCR1 fires and the (-) end of C1 will be connect to 0V and C1 will charge through R1. it will fully charge in 20mS or so.
- the delay circuit will deliver a delayed pulse to trigger SCR2
- SCR2 will discharge C2. but during that time the (-) end of C1 will be below 0V, the Anode of SCR1 will be lower in voltage than the Cathode and SCR1 will turn off.
- once C1 is discharged the current flowing through R1 will be below the hold current of SCR2 so SCR2 will also turn off.

SCR2 only being used to switch a small charge through R1 and C2 can be a relatively small device.

will this work?

rick

 

[caleb7777]

I found this in the thread earlier but it does not say it is an "SCR"

http://cgi.ebay.com/ws/eBayISAPI.dl...2BS%2BIA&itu=UCI%2BIA%2BUA%2BFICS&otn=4&ps=10

Will it work fine?

 

[rkosiorek]

I found this in the thread earlier but it does not say it is an "SCR"

http://cgi.ebay.com/ws/eBayISAPI.dl...2BS%2BIA&itu=UCI%2BIA%2BUA%2BFICS&otn=4&ps=10

Will it work fine?

yes those are SCRs. and it should work.

rick

 

[caleb7777]

Thanks rick

 

[bmazic]

i was thinking of how to make the length of the output pulse adjustable and how to turn off the SCR in mid-pulse. this method was in an ON Semi APP-NOTE

I would appreciate if you can pass on the reference to the app-note you mentioned.

Thanks,
Boris