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

here ya'go

rick
 
Hi guys, I'm pretty noob about welding and I am going to build my battery welder soon.
My question is: why can't I just use a MOSFET instead of an SCR? The famous (here) irfb4110 for example can handle 180A (680A pulsed!!! :shock: ) and if it is not enough I can put several of them in parallel. This way I can modulate PWM, frequency and voltage.
This is what I'm talking about:
http://ultrakeet.com/index.php?id=article&name=cdWelder
watch the video at the bottom! :shock: :shock:
With a lower power / duty cycle you can weld, with high power you can... fireworks! :D

I'm sorry, it's just that I have to put a microcontroller in every project I do. :wink:
 
You can, but you need a large number of them in parallel, plus you'll need to add some transient protection and ensure that the drive circuit switches them all on and off quickly. If you look back through this thread you'll find that Nemo has done just this, although he blew up a lot of 4110s before getting something that worked OK.

There are quite a few advantages for production welding in using the precise control that FETs give, particularly the ease with which a double pulse system can be built. I'm not convinced it's worth it just to weld a few dozen cells together for a one-off pack, even a simple SCR welder will do that job just fine.

Jeremy
 
I am watching a few different caps here on ebay:

http://cgi.ebay.ca/ws/eBayISAPI.dll?ViewItem&ssPageName=STRK:MEWAX:IT&item=230337816823
http://cgi.ebay.ca/ws/eBayISAPI.dll?ViewItem&ssPageName=STRK:MEWAX:IT&item=350216612794
http://cgi.ebay.ca/ws/eBayISAPI.dll?ViewItem&ssPageName=STRK:MEWAX:IT&item=350195039974
http://cgi.ebay.ca/ws/eBayISAPI.dll?ViewItem&ssPageName=STRK:MEWAX:IT&item=230341382594

All pretty much the same except some are 1 can and some are multiple cans. I s there a benefit to either for building a CD welder? Seems like more in parallel may give a little more kick because any losses are shared. Also the first one is the only one 'advertised' as having a low ESR. Is 0.002 ohm low? is this actually going to better than the others or can we expect all to have this ESR?

Just looking for opinions. Which would you get?
 
you want to get a cap with as low an esr as possible. most especially for our uses as a welder. think of ohms law (E=IR) and 2 different capacitors one with an ESR of 0.010 ohms and the second with 0.001 ohms. both are of equal Farads. charge them both up to 10V. if i short out the capacitor with a really big super conductor the capacitors will dump all of their energy. the limit of how fast that happens will be the ESR resistance. so the 0.010R capacitor will deliver 10V/0.010R = 1,000A. the other 0.001R unit will be able to discharge at a higher rate of 10V/0.001R = 10,000A. for a welder you want the higher current.

that being said, most car audio caps in the cheap range have very optimistic ratings. for a 5F car cap i would suspect that hype aside reality and testing would likely be closer to something between 1 ~ 1.5F. don't think there would be much difference between any of them.

a couple of years ago i borrowed a 3F cap (Pyramid) from a local car audio store. using a regulated 20V power supply and a 1K 1% series resistor i was able to charge it to 13V in less than 6 minutes. suggesting the actual capacity was less than 0.4F. same test with a computer grade capacitor 1F capacitor was over 18minutes. (after about 10 minutes i stopped checking as often but i know it was more than 15 and less than 20). the computer grade cap was 1/3 the advertised capacity but actually stored 3X more energy than the car audio cap. the computer cap also took approx. 3X longer to discharge through the resistor verifying that it really did store about 3X more energy. i did the test to because the computer caps were 3 ~ 4X the money of the audio stuff. also the audio stuff was suposed to be 3X the capacity.

not a very scientific test, but if did prove to me that the car audio caps were mostly hype as far as ratings go. and from the reports of the guys using the audio caps in this thread it seems the ratings are still mostly hype.

Rick
 
well yes audio caps are completly overrated. And well one thing is for sure those 5 farad ones that are a single cap size wise there is just no way, as for the power acoustik I would choose the 6F over the 5F volume wise, but since I dunno their quality you may end up with like 1-2 real farad(s) in the end maybe 3 if your very lucky. Volfenhag caps are closer to their rated capacity according to some people I have talked to. I'm not affiliated in anyway with this store, but considering your the second guy I'm sending there I should get points LOL If you buy two of these it's probably cheaper, and my best educated guess is that it it would have more real farads than any of those links you have shown, http://www.millionbuy.com/vofzx22.html the original tabwelder(under $100) builder (not the guy for england) used a 1.5F volfenhag[which he estimated had 1 real F] (BTW there not made in germany LOL) 2 of these 2.2F if their quality hold's up you should end up with more than 2F easy maybe even close to 3F (for 2 and BTW there only about 30 bucks each :D ), I doubt that you would get that close with the ones you have linked to especially the ones that only are one cap inside (the the legacy and the square 2009 Power acuostik). with the 6 farad rated one you may come close or fare even better but it would be a chance I wouldn't be willing to take.
This is my best advice
My $.02
terramir
 
I still have a couple of those Cornell Dublier 1F 20WVDC computer grade caps for sale if anyone is interested, pm me. pretty sure they will be right close to the 1F mark each with a reasonable ESR.

rick
 
rkosiorek said:
I still have a couple of those Cornell Dublier 1F 20WVDC computer grade caps for sale if anyone is interested, pm me. pretty sure they will be right close to the 1F mark each with a reasonable ESR.

rick
Click to expand...
Those are fairly old as people in the for sale forum pointed out. At the prices your asking for them there definetly not worth it.

terramir
 
terramir said:
Those are fairly old as people in the for sale forum pointed out. At the prices your asking for them there definetly not worth it.

terramir
Click to expand...

actually there is no comment in the for sale column. no replies at all. yours is the first such comment.

rick
 
rkosiorek said:
actually there is no comment in the for sale column. no replies at all. yours is the first such comment.

rick
Click to expand...
oops applogize I read a pm of one of my friends wrong, it wasn't in the for sale thread it was here:
http://endless-sphere.com/forums/viewtopic.php?f=2&t=2633&p=147530#p147530 and so on, if these things have been sitting around for years there is no way to dermine besides using them how long of a useful life they will have at the prices you asked for car audio caps will definetly be a better deal.
My $.02
terramir
 
i guess you also missed the post about actually charging and discharging them several times to verify the capacity. did it with 1k sesistors for a slow charge and with 0.25R for rapid discharge. not very scientific tests, but they do verify that they do hold a charge, are near to 1F actual capacity and do have a low ESR.

in this case your .02 worth is inflated.

rick
 
rkosiorek said:
i guess you also missed the post about actually charging and discharging them several times to verify the capacity. did it with 1k sesistors for a slow charge and with 0.25R for rapid discharge. not very scientific tests, but they do verify that they do hold a charge, are near to 1F actual capacity and do have a low ESR.

in this case your .02 worth is inflated.
Click to expand...

Well
how many hours did they hold a charge before you timed a discharge curve and got a greater than 90% of charge? Big electrolytic capacitors of that grade should hold a fairly decent charge for days as for fast discharge, .25R is not what were dealing with in capacitive spotwelders more like .06 to .006 which would allow the circuit a current flow of 250A to 2500 A @ 15V. need heck of a contact to get down below .03 though but 250A to 400A is feasible the question is how long those caps of yours would stand up to that kind of abuse, especially because the density of those caps you have mean they could be quite dangerous if faulty, warn anyone you sell em to to put em inside a safety box, newer car caps may vent on top, but they don't really have the density of the caps you got there. @ 20V the max current you can push through a resistor is 80A (remember at 15V that would only be 60, 10V 40A,... etc.
I'd just think it would be prudent to understand while caps can last 40+ years if treated right the application you are advertizing them for maxes out the specs quite frequently, while in other long term uses any engineer would ensure that the use would call for max 50% of capacity if the thing is supposed to be fail-safe beyond 10+ years. Considering this thread deals with the major abuse of caps, ergo stresses them to the limit old stock may not be such a good idea, If I built a welder with those, I would use at least two and limit the voltage to about 14V which would be about half the overall capacity of these caps. (196Ws vs 400Ws @ 20V 2F). Just to be safe since they are old.
My $.02
terramir
 
gip_mad said:
*aehm* did anyone see this link I posted early?
http://ultrakeet.com/index.php?id=article&name=cdWelder

Take a look at it, especially the nice video at the end of the page... Then tell me what you think about it! :D
I was going to follow his example - no offense, your system is very good, works well and is simple and cheap, but I'm a control freak! :D
Click to expand...
Very very cool after I do the plain all the plain battery tabs I have to do(over 1000x2) I might just work on moding it into something like that, however I would need to still be able to use it as a just plain spot welder And well for a plain spot welder the code would need to be modified to output just one or two pulses that are longer in frequency , so a switch to vary from cut mode to single or dual pulse mode would be great. But I can't say anymore till I truely looked at the schematics, because this guy looks like he has an aweful (yes I'm in awe pardon the pun) lot of power 8) either he got really lucky with that car cap or he's supplying part of the welding power from his supply (talk about toasting a supply if the electrodes fuse on) :twisted: .
My $.02
terramir
 
with the .25R discharge i was trying more to measure capacity and verify the estimate i got by charging. i just didn't want to wait a half hour to do it. also i don't have any decent sized power resistors of a lower value. i did discover that a 16AWG wire makes a nice bright fuse though.

started at 13V a couple of days ago and still over 11V today. so i think i've done eough to verify basic function. these have not leaked or dried out. the ends have not bulged and they don't make any of the funny noises or smells that big electrolytics make before they fail.

now if i had a couple of SCR's bigger than 80A handy i might try a welder. might try it anyway. i still want to try the circuit using a parallel fet to deliver a timed pulse. be nice if i could find the electrodes i made a couple of years ago as well.

i did check out the video as well. takes a lot of juice to get those tungsten electrodes to heat up red hot that way. really impressive considering the simplicity of the hardware. i wonder if one of his unpublished software features is a shutdown routine in case of a short.

rick
 
rkosiorek said:
with the .25R discharge i was trying more to measure capacity and verify the estimate i got by charging. i just didn't want to wait a half hour to do it. also i don't have any decent sized power resistors of a lower value. i did discover that a 16AWG wire makes a nice bright fuse though.

started at 13V a couple of days ago and still over 11V today. so i think i've done eough to verify basic function. these have not leaked or dried out. the ends have not bulged and they don't make any of the funny noises or smells that big electrolytics make before they fail.
Click to expand...
Not saying they won't do for a welder, hey just put em in a box,(safety first) but the price is a bit hefty, considering the audio caps while very overrated if you get the right one's you'll get it done for less
rkosiorek said:
now if i had a couple of SCR's bigger than 80A handy i might try a welder. might try it anyway. i still want to try the circuit using a parallel fet to deliver a timed pulse. be nice if i could find the electrodes i made a couple of years ago as well.
Click to expand...
I used 4 gauge stereo cable and 4 gauge bare wire from the home shmoo. also I used two hollowed out pens for the grips :p little shrink wrap and a little length of copper pipe for a solder sleeve all set. There are some people out there that are using several parallel mosfets to do that, it's doable. I was thinking I might go that route myself at a later date, however after seeing what this guy is doing I'm definetly gonna wait till he's published the stuff cause that is friggin great I could use such a cutter, but I'd sure wanna wear very dark glasses when using that in cutting mode. 8)
rkosiorek said:
i did check out the video as well. takes a lot of juice to get those tungsten electrodes to heat up red hot that way. really impressive considering the simplicity of the hardware. i wonder if one of his unpublished software features is a shutdown routine in case of a short.

rick
Click to expand...
Yeah well I actually think the main part of his power comes from the supply itself. I'd be worried about overheating that whole setup as well, he definetly needs some short and overheating shutdown routines, I'd tried to e-mail him but it's bouncing.
 
parallel FET's ?? i should have had a coffee first. i meant parallel SCRs as suggested by the ON Semi appnote.

6 years ago i assembled a welder exactly like the original www.philpem.me.uk drawings and using caps (8X 100,000 uF 20WVDC) scrounged from some early 80's era minicomputer. i used some elecrodes i machined from 3/8" diameter copper bar stock. it worked well enough to assemble a dozen or so 48V 13Ah NiMh packs. after a couple of years of non-use i traded it for something at a ham radio swap meet. the guy i traded it to modified it somewhat and got some new electtrodes made from ELKONITE that he was using to weld copper braid to phosphor bronze buttons for some antique radio restoration projects. Since then he has passed away and his welder has disappeared.

what i want to find and make my electrodes from is the elkonite he made his electrodes from. he gave me a few 8" lengths that i just can't find. i'd also like to find his schematic for his version of the welder. it did weld copper to bronze.

rick
 
I ended up ordering this capacitor today

http://cgi.ebay.ca/ws/eBayISAPI.dll?ViewItem&item=220439843596

It's 10 farad and only a few more dollars than 4 or 5 but if you read the small print and go WAY down the page it comes in a steel case! perfect project enclosure for this and has clear front so I could still monitor what volts i am at.

My plywood and 2x4 can stay in the shed this time :)

Now I just need to buid a nice electrode set like RLT
 
many big caps like those are made up from series/parallel strings of lower voltage/higher farad caps. the ESR may not be as low as a more traditional cap because of the large number of caps in series to get the voltage requirement.

i just saw the guts of one that was marked as 20F @24V it was made up of 10 series groups of 28 supercaps in parallel for a total of 280 super caps. each one was rated 6.8F at 2.5V each parallel group also had an unmarked power resistor in parallel across it. i guess the resistors were used as a balancing string. super caps like this are usually around 100mR per unit so as a rough calculation the 28 parallel 10 series network will work out to about 37mR ESR. at 14V that works out to a 380A pulse that will last for aprox 3/4 of a second. because of the high ESR that particular (very expensive) super cap would not be a good choice for a welder. on the other hand it would be good as a stiffeneing cap for a car stereo because it can sustain a fairly high current for a longer period of time.

can't wait to see what your cap looks like. hope it doesn't disappoint.

rick
 
caleb7777 said:
:(

I hope it is 2 caps
Click to expand...
The way that thing looks I suspect it's 2 0.6F to 1.5F stiffening caps and well about 4 farads worth of supercaps 6x25F in series in the middle. rip out the supercaps, and you'll have 1.2 to 3F of real low esr capacitance, this is what my welder is based on two parallel main stiffening caps out of a hybrid cap very simular to that one.
Just mine onlty had one in originally they realized that one stiffening cap won't do so the second generation of hybrid caps have two, and now there actually comming closer to the rated vaules overall, still after those two main caps are drained the esr goes up to 30 ohms so for welder purposes rip em out and put em in a fast charging model airplane LOL
my $.02
terramir
 
I'm reading through the posts and it seems like you guys haven't even build one pulse width control welder yet and you are already talking about 20 Farad capacitors...I don't understand why you guys want to go that big?....Going that big will not increase the current(heat) but it will increase the time the weld will take, and yes it will give you deeper and better welds but very expensive to control. By increasing the capacitance(the time), the components will have to be bigger and more expensive to stop and control this weld. By increasing the voltage on the caps,(more than 60V will kill) you can discharge it into a pulse step down transformer and get a cheap and easy controlled pulse width at a very high current on the secondary side.(to weld your train wheels to the tracks...just kidding :) ) For me, any weld 600Ws and more, heat should not be a problem and you can use a microwave transformer for your welds. I'm in the process of building a mosfet controlled SCR dual pulse welder http://www.youtube.com/user/Fritz9111
 
seen your closed source welder, you definetly spent more than 100 bucks on it. It's cool however it's definetly more than 100 bucks. You definetly got lucky on that car cap that it can go to 20V is astounding, most car caps can't. I think your welder is a bit overrated capacity wise since I doubt that cap has the full capacity advertized. (never seen a car cap that wasn't way over rated, then again I haven;'t seen your schematics, if your power supply is supplying part of nthe current during the weld you might just have that capacity. Waiting for your scematics to come out then I'll reserve judgement on the rest. But either way it's cool 8)
terramir
 
Terramir, I have a Capacitor tester for my caps and this cap measures 2.97 Farad. I also look at the discharge times
with an oscilloscope and I can calculate the exact size of the cap.This cap is rated 20 volt and have a surge of 24volt. At 20 volt It has a current leak of about 3 amps but it works. It might have cost more than a $100 but it is still less than $6000 :roll: I have posted an updated schematic about a week ago.
 
Fritz said:
Terramir, I have a Capacitor tester for my caps and this cap measures 2.97 Farad. I also look at the discharge times
with an oscilloscope and I can calculate the exact size of the cap.This cap is rated 20 volt and have a surge of 24volt. At 20 volt It has a current leak of about 3 amps but it works. It might have cost more than a $100 but it is still less than $6000 :roll: I have posted an updated schematic about a week ago.
Click to expand...
Okies what brand of car cap is that there about to get a whole lot more business!!!!!!!!!!!!!!!!!!!!!! :eek:
cause all the car caps we have seen so far have grossly exaggerated ratings! can you do a manual 1rc test over a voltage constant source to verify you capacitance meters accuracy? and BTW where the heck did you get a capacitance meter that is capable of more than 4000uF most of us are stuck with something that is not even as good as that, many multimeters will even just go to 400uF max.
as for rated 20V 24V surge in the community we have had caps rated 20/24V that blew a gasket at 17.5V


terramir
PS:Looked at the schematic and I still dun see any way your stopping the scr from discharging there is nothing connected to short the scr, etc as far as I can see right now :S
 
You must log in or register to reply here.

Similar threads

O
Battery Build Planning
Replies
1
Views
604
Hummina Shadeeba
H
Lucifer
Want a battery spot welder that works for £20?
2
Replies
26
Views
7,003
Daijoubu
D
J
Cyclone 3000W Aluminum custom frame build...
Replies
3
Views
3,237
Jim Day
J
nutnspecial
Parts/build list
Replies
6
Views
3,559
nutnspecial
nutnspecial
Back
Top