Apartunis
100 W
The usual Florida way is to vent to ambient with no protection for workers. So yay?
Texas, same.
Pure nickel vs nickel plated & copper sheet?
- Thread starter markglos
- Start date
The usual Florida way is to vent to ambient with no protection for workers. So yay?
Texas, same.
I've been doing test welds using the same 40 18650 cells that's been Dremel-stoned a few times. The Ni plating long gone, the steel is thinner so the welds are not representative of a new cell anymore. I need to find more virgin 18650/21700 cells to practice on.
Threaded terminals will make life a lot easier. When I was looking for cells to build my moped battery, nothing else would fit and fill the battery box and provide the V and Ah needed as well as the Gotian 32135's.
Dug out my old Tektronix scope from storage to have a look at the waveforms of my AWithZ. Some results are not what I expected. I'll write about it tomorrow.
Threaded terminals will make life a lot easier. When I was looking for cells to build my moped battery, nothing else would fit and fill the battery box and provide the V and Ah needed as well as the Gotian 32135's.
Dug out my old Tektronix scope from storage to have a look at the waveforms of my AWithZ. Some results are not what I expected. I'll write about it tomorrow.
My soldering welding station used to be in garage, but being Florida with a South facing door and west facing cinderblock wall radiating stored heat internally, and a hot car engine radiating heat well after sun down means sweating balls and swatting mosquitoes a majority of the year.
Soldering/welding station is indoors now, so HEPA and activated charcoal air filtration at point source on desktop, as well as nearby was needed for peace of mind. My lungs have seen too much particulates already.
The point source filter's charcoal filter turns white from soldering smoke.
12vdc nominal forms a basis for much of my ventilation and lighting, and lead is nearly dead.
My 22ah UB12220 AGM battery is well aged, and the last time I tried to use it to help jumpstart a vehicle it was obviously weak, as expected for its age, even when treated as well as one can treat lead acid.
The Gotion cells at battery hookup in a 4s2p configuration were the most economical option, but now that I know I cannot weld to their terminals with my P20B, they are not an option anymore. Especially at 12v nominal as excessive voltage drop means I use fat cabling, and piddly 0.2mm copper was not going to satisfy my desires even if I could weld it.
The SAFT VL30PFE cells with screw terminals will allow me to make FAT busbars and 4 of these power cells in series will likely outperform a healthy group 31 AGM in starting my engine, when the agm inevitably fails.
Charging below 0c/32F is unlikely to ever be a factor for me, and I can control my alternator's max target voltage manually.
Not sure I will ever install it in my Campervan as the starter battery, but I want it to be able to perform that task.
I am not trusting drop in Lifepo4's build quality or the BMS'w ability, so a DIY 28ah 4s1p of the Saft Lifepo4 cylindrical power cells and a 300 amp dumb BMS have been ordered to finally get into Lifepo4. I was waiting for my Lead acids to die first and they are getting close.
Looked at the waveforms of my AWithZ UF20B on an oscilloscope. The Intermittent setting (1-20ms, 1ms increment), does increase the delay between the Preheating (0-50ms, 5ms increment) and the first welding pulse. With Intermittent set @1ms, the delay is 120ms, with intermittent @20ms delay is 140ms.
My UF20B has Number of Continuous Spot Welding of 2 (@sternwake I think your P20B has 3?).
With Preheating set to 0ms, I see only one welding pulse on the scope but hear 2 beeps. With preheating set to anything other than 0ms, I see 2 welding pulses and hear 2 beeps. Preheating settings between 5ms to 50ms does not seem to change the timing of the 2 pulses.
Gear setting vary the duty cycle, thus the power:
Gear 1 pulse width 1.25ms 6V
Gear 10 pulse width 7ms 6V
Gear 50 pulse width 32ms 6V
Gear 99 pulse width 60ms 6V
Duration of 1st and 2nd pulse is the same.
My UF20B has Number of Continuous Spot Welding of 2 (@sternwake I think your P20B has 3?).
With Preheating set to 0ms, I see only one welding pulse on the scope but hear 2 beeps. With preheating set to anything other than 0ms, I see 2 welding pulses and hear 2 beeps. Preheating settings between 5ms to 50ms does not seem to change the timing of the 2 pulses.
Gear setting vary the duty cycle, thus the power:
Gear 1 pulse width 1.25ms 6V
Gear 10 pulse width 7ms 6V
Gear 50 pulse width 32ms 6V
Gear 99 pulse width 60ms 6V
Duration of 1st and 2nd pulse is the same.
Interesting results.
I just set my p20b to 0.00MS preheating and dual pulses at gear 90( a good setting for 0.15mm pure nickel), intermittent 02ms, and the unit beeps twice and I can feel it hit twice and the leads jump twice.
I have the "interval of continuous spot welding' set at 0.2 seconds.
This "interval" setting is on a different sub menu in the 'Settings" Different than the Main menu with the gear and number of continuous spot welds, and is perhaps the most obvious noticed change when making a weld, although I cant say whether 0.1 seconds or 0.2 seconds is 'better' or forms stronger welds.
This "interval of continuous spot welding' should be on the main menu screen, not hidden in the 'settings' sub menu in my opinion.
I do like that battery hookup capacitor spot welder where one gets to set the duration of either pulse and the delay between them as opposed to 'gear level'
I was not aware of this setting. Just found it, it was set to 0.1 seconds. I will look at it later with the scope and see what it does.
My scope tests were done with a 22 ohm resistor as a load across the probe tips, not doing actual welds. I will see if I can set the scope up again while doing actual welds and see if the behavior with 0.00ms preheating changes.
I have used 0.1 seconds 'interval' on most of my welding but switched to 0.2, not that I noticed any difference in weld strength results, but that I feel same power spread over longer would cook less electrolyte on the inside of the cell under the electrodes.
Feels does not equal facts though.
Honestly, it is difficult to see feel much difference in the welds on tear off strengths when changing preheating or 'intermittent' or interval values. Gear has by far, the biggest effect.
Opening up the interval from 0.1 to 0.2 seconds is obvious in that the pulses are more separated, but again, I can see no difference in weld results.
I have not even found much difference between a single weld pulse and dual weld pulses on the same gear setting.
It kind of seems like a single pulse at say gear 400 hits harder than each individual pulses at gear 400 when set to 2 pulses.
My p20b is capable of 3 pulses and I have used triple pulse when trying to find upper limits welding 0.3+mm copper, but again I cannot definitively say that 3 pulses are doing what 2 or 1 at the same gear cannot.
The oscilloscope results are appreciated.
Whether the 'intermittent' or preheating or 'interval' changes the resistance of the spot welds themselves is a whole other can of worms.
Feels does not equal facts though.
Honestly, it is difficult to see feel much difference in the welds on tear off strengths when changing preheating or 'intermittent' or interval values. Gear has by far, the biggest effect.
Opening up the interval from 0.1 to 0.2 seconds is obvious in that the pulses are more separated, but again, I can see no difference in weld results.
I have not even found much difference between a single weld pulse and dual weld pulses on the same gear setting.
It kind of seems like a single pulse at say gear 400 hits harder than each individual pulses at gear 400 when set to 2 pulses.
My p20b is capable of 3 pulses and I have used triple pulse when trying to find upper limits welding 0.3+mm copper, but again I cannot definitively say that 3 pulses are doing what 2 or 1 at the same gear cannot.
The oscilloscope results are appreciated.
Whether the 'intermittent' or preheating or 'interval' changes the resistance of the spot welds themselves is a whole other can of worms.
With 0.00ms preheating, doing a weld instead of the 22 ohm resistor load, I heard 2 beeps and saw 2 pulses on the scope. So the results with 22 ohm load was not valid, at least for 0.00ms preheating.
With 22 ohm load, the pulses on the scope were nice clean square waves, not so when welding. I started at gear 10, then I went up to gear 50 on a piece of 0.3 nickel, press the foot pedal, saw a big flash melted a hole through the nickel which fused on one of the tips. Had to clean it up with a diamond file.
It was much easier and safer with the 22 ohm load. I am not going to continue. I think the timings from the tests done already is correct other than @ 0.0ms preheat.
It was much easier and safer with the 22 ohm load. I am not going to continue. I think the timings from the tests done already is correct other than @ 0.0ms preheat.
There are too many settings on the AWithZ IMO which makes it more complicated than necessary. Doesn't help the manual does not explain clearly what the settings do.
I agree. The manual is basically non existent and of no use in determining where to start for a given material thickness.
I have been very frustrated when trying to determine the effects of increasing or decreasing preheating, or the delay between preheating and first weld pulse, or the interval between weld pulses.
I was also a bit annoyed as I initially thought 'intermittent' was the interval, but 04ms shoukd basically feel like one welding pulse, but there were obvious defined pulses.
Later I discovered the partially hidden 'interval' option and changed that to 0.2 seconds and then the weldi g oukses were seoarated more, bjt I coukd detect no difference in weld quality from 0.1 to 0.2 seconds interval.
I've largely settled on 0.15ms preheat, 04ms intermittent, 0.2second interval between double weld pulses, and just raise or lower the gear.
But I an mostly using 0.2mm copper under 0.1mm stainless with flux and gear 385 on my p20b yields very solid very consistent welds.
Overall I think it is a solid performing spot welder and the bells and whistles can largely be ignored. Set it to single pulse and keep raising the gear in stages until both welds are same size and strength when tearing them off.
I've largely targeted copper stainless sandwich so perhaps the bells and whistles settings have more of an effect on pure nickel or copper nickel sandwich.
Once I used stainless, my nickel and nickel plated steel were neglected.
I have been very frustrated when trying to determine the effects of increasing or decreasing preheating, or the delay between preheating and first weld pulse, or the interval between weld pulses.
I was also a bit annoyed as I initially thought 'intermittent' was the interval, but 04ms shoukd basically feel like one welding pulse, but there were obvious defined pulses.
Later I discovered the partially hidden 'interval' option and changed that to 0.2 seconds and then the weldi g oukses were seoarated more, bjt I coukd detect no difference in weld quality from 0.1 to 0.2 seconds interval.
I've largely settled on 0.15ms preheat, 04ms intermittent, 0.2second interval between double weld pulses, and just raise or lower the gear.
But I an mostly using 0.2mm copper under 0.1mm stainless with flux and gear 385 on my p20b yields very solid very consistent welds.
Overall I think it is a solid performing spot welder and the bells and whistles can largely be ignored. Set it to single pulse and keep raising the gear in stages until both welds are same size and strength when tearing them off.
I've largely targeted copper stainless sandwich so perhaps the bells and whistles settings have more of an effect on pure nickel or copper nickel sandwich.
Once I used stainless, my nickel and nickel plated steel were neglected.
This morning I tried again on my AWithZ UF20B maxed out on gear 99. Took a video this time.
0.1mm 304 SS over 0.2mm Cu + flux over the infamously hard to weld 32135 LFP cell with a 1.6 mm thick copper tab. It was a weak weld but a weld nevertheless.
AWithZ P20B is not powerful enough per tests done by @sternwake. Maybe higher power welders such as P30 and higher will work? Won't know till someone try it.
0.1mm 304 SS over 0.2mm Cu + flux over the infamously hard to weld 32135 LFP cell with a 1.6 mm thick copper tab. It was a weak weld but a weld nevertheless.
AWithZ P20B is not powerful enough per tests done by @sternwake. Maybe higher power welders such as P30 and higher will work? Won't know till someone try it.
Pretty sure Zeus-FL used gear 700 on the P90C to weld 0.2mm copper to those same cells, with flux but without the SS sandwich in his video.
Ill GUESS that would be require gear 550 with stainless on top.
The p90C is rated at 56.8kw
55% of 56.8kw is 31.24kw
The AwZ p30c is only 24kw.
The AwZ p60f is 29.4kw
There's is no awz welder between the 60f and the 90c, and the 90c is a 650$ spot welder.
BlueSwordM
1 kW
- Joined
- Jul 23, 2018
- Messages
- 498
Yep. The main trick about the P60F vs P90C is that the P90C has a much beefier PCB. The larger supercaps also make a different, but not by that much.
Otherwise, it's not that different from the P60F, which is why it's a much better bargain.
I've not tried to open( pry off rectangular sidecap?) up my p20B and have a look inside.
I have thought about having a 10mm wide 12v fan attached to the side, and a filtered intake opposite
The temp gauge rises pretty quickly when I am doing a lot of welds at higher power levels.
BlueSwordM
1 kW
- Joined
- Jul 23, 2018
- Messages
- 498
If I'm not wrong, the P60F uses a 3S2P array of 1000F supercaps, the P90C uses a 3S array of 3000F supercaps and a stronger PCB, and the P120D uses a 4S array of 3000F supercaps with an even stronger set of MOSFETs and components.
BlueSwordM
1 kW
- Joined
- Jul 23, 2018
- Messages
- 498
IMO, the P90C and P120D are clearly into the realme of "You buy this because you need it for your business or you want it and because you can't afford one of our laser welders yet".
P90C 56.8 kw, P60F 29.2 kw. P90C has 48% more power than P60C. Why do you say it's not that different?
BlueSwordM
1 kW
- Joined
- Jul 23, 2018
- Messages
- 498
Specifically because the main difference seems to be down to them not speccing out the more capable PCB on the P60F to not encroach on the P90C. I'd love to know about the difference in supercap specs to see if the P90C truly has 94% more power available, or is part of the upselling reliant on just the PCB differnces.
Also, it's not 48% more power, it's 94% more power.
You are right!
Source?
Not sure if this table is 100% accurate. Feel free to correct if you see any errors. Looking at the data, P60F and P30C are the two that seem to be not very different from each other, not P90C and P60F.
| P120D | 69.2 kw | 10.5 v | 10.3 lb | |
| P90C | 56.8 kw | 10.5 v | 6.6 lbs | |
| P60F | 29.2 kw | 9 v | 3.6 lbs | |
| P30C | 24 kw | 9 v | 3.6 lbs | |
| P20B | 14.6 kw | 9 v | 3.13 lbs | |
| UF20B | 10.5 kw | 6 v | 2.76 lbs |
BlueSwordM
1 kW
- Joined
- Jul 23, 2018
- Messages
- 498
Oh no, I'm not talking about the differences in build. It's obvious that the P90C is bigger, beefier and uses larger caps. I'm just saying that if the P60F had the larger beefier PCB and components of the P90C, that the difference wouldn't be as large as 94%.
Also, I think this table is wrong. The P90C uses 3x supercaps in series with a max voltage of 9V, so 9V 3000F.
The P120D is where it gets very very interesting. It uses 4 caps in series, but either underdrives them from 3V to 2.7V, or they're using different supercapacitors entirely, which are more power dense, but less energy dense.
That would explain the huge weight jump from the P90C to the P120D: larger beefier PCB and components, a 4th supercap, lower energy density caps, etc.
braunschweiger
10 mW
any recommendation for a soldering flux to solder balance wires onto the SS-304?
Nix Liou
100 W
Since I've been building some larger packs, I've figured out a pretty good setup and work flow for spot welding cells:
With my P60F, and nice copper bus bars utilizing Ni/Cu weld points, it's quite consistent when welding to standard nickel plated steel cells. I'll post my exact settings next time I'm at my workbench, but I have it set to gear 2 out of 999, when welding the .1mm weld points (35 microns Cu on bottom, 65 microns nickel on top) to 21700 cells. With such a short duty cycle, I can really crank through a ton of welds without overheating or damaging the welding needles. I found the settings on the preheat pulse on the P60F to mostly help with the diameter of the weld point, so I found a nice middle ground between a wide weld point and it melting and spraying droplets of nickel around...
Maybe I should do some testing myself on Alu cells though?
Here are some examples of the bus bars. I determined the thickness of Cu and Ni on the weld points by carefully sanding the copper off until it was only silvery, and measuring it with my calipers, so it's kind of an approximate. Maybe it is 50/50.
I calculated what effect on series resistance this thinner weld point would have, as well as the thermal losses at up to 50A from a single cell, and it's pretty much irrelevant. Since I can get these bus bars with up to 1mm thick copper, I can counter losses from the thinner weld point with that.
BlueSwordM
1 kW
- Joined
- Jul 23, 2018
- Messages
- 498
So, I just got the AwithZ P60F.
It's clear I don't know what strength settings to use, because my probes are sticking and nuking either the probes or my test cells.
Any recommended settings for 0.15mm Cu + 0.1mm steel? This video kind of mislead me TBH:
It's clear I don't know what strength settings to use, because my probes are sticking and nuking either the probes or my test cells.
Any recommended settings for 0.15mm Cu + 0.1mm steel? This video kind of mislead me TBH:
Last edited:
Can you be more specific please? Which column(s) and which model(s) is wrong? I believe the kw and V columns are correct since they were gleaned from published sales ads. Weight was not found for all models in sales ads. Some were obtained via Google AI search.
