DIY Arduino timed spot-welder, FET-switched, adj. pulse

I went to a makerspace this evening to try measuring the mosfet avalanche time with their storage oscilloscope. When I got there, I found the anode tab had broken off the schottky diode (from bumping around in my basket for 13km) so I wasn't able to use it. The people there helped me use the DSO (I got there late and didn't have time to let them show me how to use it myself). I did the test with no protection diodes and got ~150μs of avalanche.

I then decided to test the TVS by itself (I wasn't worried about destroying it as I can get a new one when I buy another schottky). I did the test but the TVS diode did not blow up like I was expecting (only using one 12V battery). Also the current didn't stop, leading me to believe the mosfets had somehow failed. I was unsure why they would fail and hadn't had time to test it yet. Talking to my father later I suddenly realised that the closed circuit was probably the failure mode of the TVS diode when the avalanche energy isn't enough to blow it up. I've just tested it and confirmed that it was only the TVS that failed.

The above test with the TVS diode only resulted in about 20-25μs avalanche even though the diode eventually failed. The dotted yellow horizontal line in the photo is 20.4V and the solid line is 0V. You can see the voltage falls back to 5V (12V minus voltage drop across the TVS), not 0V as it should.
 
From ES member Tesseract:

https://endless-sphere.com/forums/viewtopic.php?f=14&t=81400&start=525#p1209998

So, we want to concentrate on changes/improvements that will reduce the importance of selecting the right size battery, for example. Too small a battery and you'll be limited to welding very thin metal, or not at all, while too big a battery and you might exceed the peak current rating of the MOSFETs, or their avalanche energy rating (if no freewheeling diode is present on the MOSFET board - not dangling in the middle of the welding cables somewhere... ahem).

In that spirit, then, I would say that adding a freewheeling diode is mandatory, but the input capacitors can probably be dispensed with, especially if the suggestion I'll give below is implemented.

For the freewheeling diode, part number 100BGQ045 part number that okashira suggested a few posts back is superior to the MBR8020R I suggested, both in peak current rating and cost.

The original spec'ed MOSFETs - (AU)IRF1324 [the AU prefix means Automotive rated and is optional] is a really good choice, though I'd probably bump that number up to (8) in parallel so there isn't such a worry with the size/ampacity of the battery.

I would design the board so that it bolted onto the negative post of the battery and used a short-as-possible jumper to the positive post comprised of 5-7 #14AWG insulated wires wire-tied together. Doing this, rather than using a single larger diameter wire, will put the stray inductance of each wire in parallel, which will reduce it by the inverse of the number of wires paralleled (ie - down to 1/5th or 1/7th), and the lower this inductance the less need - perhaps no need at all - for decoupling capacitance (though as I mentioned before I consider this bad engineering regardless). Also, 5-7 #14AWG wires will be cheaper and easier to obtain than a single #4-#6AWG wire, and they will be easier to solder (in individual holes) to the PCB.

I would use a stronger gate driver IC - I'd like to see at least 4A peak here - and lower the value of the gate resistors accordingly. For example, MIC4420, a 6A driver for $0.95-US, and 22 ohm resistors for each MOSFET (whether 6 or 8 are used).

One of the minor changes I would make is to get power to operate this circuit from the 12V battery by using a blocking diode and a bulk storage capacitor so the circuit still functions while the battery is delivering welding current. There are undoubtedly a few other things that I would do differently, but this is already turning into a proverbial "Homer's catalog of ships" (re - The Iliad) so I'll leave off for now.
 
Thank you flangefrog and everyone else who has contributed to this design. I'm sure there will be several versions of this produced (each with some minor improvement), but we are now quickly closing in on a design that will work well, and will be very affordable to produce.
 
flangefrog said:
Here is a new prototype spot welder probe. Using 6AWG silicone which is probably fine but I'm going to try to buy some 4AWG silicone. It's much more flexible than the neoprene rubber insulated cables I was using.

That's a fantastic looking Probe, may a ask how you made it? or where you can purchase parts to make it. Well done anyway, it looks very professional.
 
I used these Hakko clone soldering iron handles: http://www.ebay.com/itm/232021180398

The iron is just drilled out enough so that the copper can be press fit into it.

Copper rod I used is here: https://endless-sphere.com/forums/viewtopic.php?f=31&t=80315&start=50#p1215403
 
Hi flange, i ordered the mouser parts list from the aus thread and have built up another welder that used the same components as yours. Did you ever get the tvs and schotkey diodes up and running properly? I see you had some mechanical issues but was wondering if you had progressed any further. I am wondering if the 2 components are worth adding?
On another note, using the copper bar makes it far easier to build the bottom board. I had no problems building it this way compared to past welders
 
By keeping this "open source", nobody can start producing them for sale at an outrageous price. The design is laid out, and anyone can produce a few of these for sale from their garage. It may not seem like that big of a deal now, but I see this as a link in sparking a revolution. ebike kits will boom soon in North America, and there will always be customers who will gladly pay more for a turn-key battery pack. However, if there is a trade war with China, motors and battery packs will be very difficult to get.

If they remain easy to get, then I was worried for nothing, but also...if they remain easy to get, then spot-welders will not be in much demand....only a handful of ES enthusiasts making a few bucks producing custom-shaped packs in their garage for low production market needs...
 
Agree with flangefrog. I have been using the "stanard version" for a while now with no problems. The main reason i became a battery diy enthusiest was because of battery shipping regulations and the flexibility of being able to build any shape/ configuration you want. You also know exactly whats in your pack so you don't get ripped off with someone selling you over rated packs
 
Flangefrog, did you ever compare the difference the alternate tvs and flyback diodes made? I'm Auckland based and have access to a Digital scope if you wanted to do some testing.
Do you have any spare PCBS for the spot welder?
 
Flangefrog, I know you are a busy guy, but...do you feel we are close enough to order some PCB's to be ordered? and also publish a parts list?

I would like to get the ball rolling and order a batch of parts to make ten units. I'd like all the "open source" info published so anyone can go their own route, but I also want to take the next step and make a parts kit, and also a few finished spot-welders made available. If enough parts are ordered at the same time, the price per unit goes down. I'd like to spend some money to make this happen. I am certain that I can break even on a mass buy and distribution of kits, whether in parts or as a finished unit.

If this has a glaring error in design, we need to know now...if it is the best possible design from a bang-for-buck perspective, I'd like to verify that ASAP. Help me to make this happen...
 
calebleemcd said:
Flangefrog, did you ever compare the difference the alternate tvs and flyback diodes made? I'm Auckland based and have access to a Digital scope if you wanted to do some testing.
Do you have any spare PCBS for the spot welder?

Hey There, If Flangefrog can't hook you up with any Spare PCB's i've got some spares. Happy to help anyone out. PM me
 
Got my frog mods on the arduino. Seems to work. So far no noticable difference in performance from the original design but i think it will be able to handle more amps than i am testing with and will be more reliable. Will put a 3rd battery in parralell on the weekend to see if it goes bang. I am able to weld .2mm nickel easy, having problems with .1mm copper though.. i really want my next pack to have copper connections to handle the amps i want to through at it.
 

Attachments

  • IMG_2131.JPG
    IMG_2131.JPG
    83.1 KB · Views: 9,589
Just finished my spot welder and found a problem:
My arduino freezes when I use it together with the schottky diode.
The status LED does not turn off anymore, the LCD digits stop alternating and no further pulses can be fired. A reboot fixes it until the next pulse.
Without schottky everything works beautifully.

Any idea what can cause this behavior?

TVS Diode is also installed - so it is at least better than without modifications I guess.
 
First thing i would check is the voltage on the pins of the mosfets with a multimeter while its hooked up to the battery. Make certain that the pin from the mosfet drivers are not 12v. Ie, 2 pins should be 12v the other should register pretty much noise. If this is all ok i would change out the diode. I would also change out the nano- i have found that once they get fried they will always keep giving you problems.
Post some close ups of the top board and base so i can make sure there is nothing out of place
 
Here are some pictures.
The Nano works still perfectly fine after removing the schottky.
I removed it and was able to spot weld my first 4S2P Liion pack fpr my electric drill.
With the schottky attached the Arduino did not just stop responding but the weld was also a lot weaker than without.
But for now I just hope the TVS is already sufficient for the low number of welds I am planning.

2016-11-14 09.12.16.jpg
2016-11-14 09.11.23.jpg
 
Your diodes are different from the ones i am running- not sure if that could be the problem. The ones I ordered were from mouser as seen below. I have used the standard version of this welder with no mods for a long time and its never missed a beat. Only reason i built the flange version was to see if i could weld copper reliably. If you are welding up to .2mm nickel i wouldn't hesitate to ditch the diodes. I have put 1000's of pulses through mine.
 

Attachments

  • IMG_2159.JPG
    IMG_2159.JPG
    40.5 KB · Views: 9,421
  • IMG_2160.JPG
    IMG_2160.JPG
    44 KB · Views: 9,421
I used the diodes that flangefrog recommended in this post due to me ordering from reichelt in Germany (they did not have the originally recommended ones)
https://endless-sphere.com/forums/viewtopic.php?f=14&t=81400&start=550#p1219916
He also said
I think two schottky and three TVS diodes would be the max you need. Even just one of each or only a schottky will help.

I used three 1N6275A and two MBR 4045PT diodes.
After noticeing the fault with the schottky I just removed it and left the TVS as you can see in the image. Worked like a charm for my first Liion pack with 0.15 Nickel strips. I was just wondering if someone could explain the reason for the weird behavior with the schottky. That is why I initially asked.
 
I just finished building this welder and although it all looks ok - it sparks when the foot pedal is depressed but I am struggling to weld 0.15 nickel to a cell.
I'm using soldering iron tips as probes and the best I can manage to do is get these to stick to the nickel when i try to weld.
I've tried turning the pulse up to 20ms but i'm not sure it seems to make a lot of difference.
Anyone able to offer any suggestions on how I might start troubleshooting this?
 
Back
Top