120V Flux Welder, 2000W Inverter. What?!!

[12VoltInstalls]

I did something wrong today.

I had to weld a couple small things and was about to plug into the generator when! I thought it would be inter-cool-resting to see what happened if I tried to use the solar system.
I have a 2000W psw QZRELB/Reliable inverter currently sourcing from only two 140Ah 12V LiFePo batteries in parallel.

Welder is a Lincoln 120V spooled with fluxcore. It ‘can’ and has run on a 15A breaker but long welding trips it after a bit. 20A breaker and it will run continuously until duty cycle if not on a long extension cord.

I made my five small welds on “C” (3rd of 4 positions) and 2 wire speed. On the 2kW inverter. Surprises me that it worked! Don’t know what actual amps were unfortunately, but I might try the 120V mig next time.

Pretty cool little experiment to me. I’m quite entertained. Very pleased with myself LOL
I plan to get a QZRELB 3000W version at some point. For little fab stuff and sheet metal it appears I won’t need the generator, and the headroom of the 3000W should be sufficient if the dang welder didn’t fault out the 2kW unit. I feel so giddy and juvenile right now

 

[Steve_S]

I've used my MIG with Gas & Flux (pends what I'm doing) without any issues, it a 120V/15A model BUT one of the first lines in the manual is to put it on a 20A circuit _and_ if using an extension cord, no more than 5' & industrial type with 10g wire. Anything thinner and they melt... ask me how I know, you likely can figure it out ;-)

Note, I first used my MIG with a Yiyen APC-24-3000W Low Frequency Inverter, it handled it but wasn't terribly pleased (heated up fast and the fans were in overdrive). Then my Samlex EVO-4024 Low Frequency Inverter didn't even batt an eye... It warmed up as expected but not overly so.

BTW, if you have a smart BMS with a phone app, watch what the BMS does when you pull that MIG Trigger... that 1st surge pull is a doozie !

 

[12VoltInstalls]

that 1st surge pull is a doozie !

Apparently electrode positive has a higher surge. The mig would trip out the inverter 8 or 9 out of 10 trigger pulls. I think a 3000W will fix that.

 

[Hedges]

If these welders have a non-PFC diode/capacitor front end, I would expect them to pull higher RMS current than what wattage indicates, more heating of wires and tripping of breakers. House wiring ought to be adequately protected by the breaker, both of which experience I^2R heating. Undersized extension cord only adequate for 1800W resistive load would feel the stress.

My pool pump used to operate continuously on 15A breaker when it was split-phase induction motor. Now same HP 3-phase motor fed by non-PFC VFD, full speed trips 20A breaker after a while.

I would expect those gulps of current to be a challenge for small inverter.

See top (green) trace of third screenshot:

Battery ripple current

After reading of someone tripping battery breakers at well below rating, I realized that could be due to pulses of current drawn by the inverter. Ideally capacitors would smooth out current draw so high frequency switching pulses and 60 Hz draw all came from the caps, and current from batteries...

diysolarforum.com

 

[12VoltInstalls]

would expect them to pull higher RMS current than what wattage indicates, more heating of wires and tripping of breakers

I have (well, connected at this time; I ‘borrowed’ the 200Ah chins to run a pump to water the lawn from the brook) two 140Ah LiFePo with 100A BMS’s. The 2000W inverter has a 250A fuse, batteries have 125A Class T’s, current 120VAC subpanel has a 30A breaker with 15A to the outlet. Breakers didn’t trip, but the inverter tripped instantly (assuming inverter; I cannot imagine the QZRELB 2kW inverter actually exceeding the BMS’s but that is an unlikely possibility and the fuses didn’t blow).
So that link was an interesting read though a bit above my pay grade. What bothered my head is that the ‘electrode negative’ fluxcore caused no issues on “C,” yet the mig machine wouldn’t run on a “B” or “C”setting. (Mig process in case someone doesn’t know is ‘electrode positive’ with C25 shielding gas). FYI the Champion 6250 inverter generator ran it fine even on ‘D’

My suspicion is that the fluxcore wire starts an arc into plasma easily with the polarity, while with the mig process you can watch the wire light up ever so slightly before the arc establishes bright and white through the auto-darkening helmet I use. That visible pre-arc heating- though only a split second- must be a significantly higher amperage and for a longer time (in milliseconds) so the inverter dumps its “surge” caps before the amperage can settle.

No matter: I didn’t actually expect the fluxcore “test” to work on 16.x amps so no biggie. If/when I buy the 3000W version and I can’t do 16ga …that would disappoint me a little.

 

[OffGridForGood]

Hum, Mig on the portable MPP 2724 -24v system with a single 304Ah LiFePO4 battery - about 85% charge at the moment.
Sure why not try it?
The MPP has it's own outlets right on the side of this inverter, rated for 20A, plug the Mig directly to the inverter, no extension cord.
Some 1/4" plate, setting "D" 2.5 wire feed speed - five minutes straight, no sign of trouble, no faults.
Lets try this again with a Kil-O-Watt plugged in, to try to catch the power consumed and peak (well at least whatever the P3 can pick up fast enough)
Test #2 same plate and settings: five minutes of welding - no fault no signs of trouble. P3 doesn't record a peak, only accumulated kwh =0.08kWh
Okay try this a third time, set my phone on record to catch the peak amps while welding!
Lets try a bit higher setting, "D" and 3.5 wire feed speed.
LOL should have done this before 5:00PM could have let someone else run the welder and I could catch the P3 reading without the I-Phone!
Result: okay so start up on the Mig, not welding, just "on" the P3 shows 0.35Amp
Started welding, the amperage jumped to 17-18-19 bounced around between 18-19.6 for about a minute then the inverter shut it off - damn. Now I don't know if the Inverter was heating up after the first two tests, or if the higher feed rate boosted the amperage to tip it over the limit, LOL.
I guess some short light-welds can be done with the MPP 2724 paired with a 304Ah DIY battery - this one has the Daly 8S 150A BMS.

 

[OffGridForGood]

okay let the MPP cool off for a bit - actually I had some supper and wondered about the Mig, so came back to the shop to run a 4th test:
Set up the welder, and the I-Phone to capture the P3 amperage while the welder is running.
since setting 'D' seems to run close to the 20A max output on the inverter, lets' try the lower settings.
Setting C feed at 2.5, amps bouncing around between 14.5 and 16A
Setting B feed at 2.5, now the amps are between 12- 14A
Setting A feed at 2.0, now the amps are between 9-11A
Seems like the lower settings you could run ok and not trip the inverter, with the small welder.
I repaired a gate hinge last fall using a minigrinder & the welder on this same MPP 2724 small portable inverter and battery set up, but never thought to plug in the Kil-O-Watt until now. Interesting.
If you used a 3000W inverter, assuming you can pull 25A from it continuous, you should be able to run a Mig on "D" - at least until things heat up inside the inverter.
A 2000W inverter, (16A) it would seem is going to be 'right on the edge' for setting C, maybe short runs only.

 

[Hedges]

Does Kil-o-watt report power factor?

 

[OffGridForGood]

Does Kil-o-watt report power factor?

Not that I can see, it is pretty basic: Volts, Amps, Watts, Hz, kWh accumulated, Time.
Can we calculate PF from these?

 

[Crowz]

I have worse luck killing the kil-o-watt meters vs what I'm testing...

I end up popping the thermal overload on these like popcorn :

https://www.amazon.com/gp/product/B00009MDBU/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&th=1

P3 P4400 Kill A Watt Electricity Usage Monitor​

 

[Crowz]

Now I stock replacement thermal fuses for the things and just swap them out when they fry.

 

[Hedges]

Not that I can see, it is pretty basic: Volts, Amps, Watts, Hz, kWh accumulated, Time.
Can we calculate PF from these?

Try Watts / (Volts x Amps) = PF

If we read volts and amps with separate meters and multiply, that is apparent power.

Old watt-hour meters were some kind of motor that I would guess used voltage through a resistor to make magnetic field, current to make magnetic field, and accumulated the product.

I use a digital scope, have it multiply V x I sample by sample, then average.

 

[12VoltInstalls]

2000W inverter, (16A) it would seem is going to be 'right on the edge' for setting C, maybe short runs only.

Just to clarify- considering the colloquial use of “mig”- that was ‘gas welding’ not fluxcore wirefeed? Another Lincoln?

But ya, I didn’t expect the flux machine to work and was surprised it did fine. On “C” with fluxcore. (I have two identical Lincoln 120V models).
FWIW I was on a 50’ 12ga cord. Once I bury the two runs of 3/4” pipe I will have 10ga thhn feeding the shop which should help.
And I’m curious about your setup

fyi it was merely an attempt for the sake of entertaining myself, not a requirement. But if 2000W sorta works 3000W should let me do common things like sheet metal or fab up stuff which often is intermittent welding only a minute here and there. “Actual” welding for anything I’ll fire up the generator; plus next year I might set up a separate split phase aio system with 48V batteries for the shop.

 

[OffGridForGood]

Just to clarify- considering the colloquial use of “mig”- that was ‘gas welding’ not fluxcore wirefeed? Another Lincoln?

But ya, I didn’t expect the flux machine to work and was surprised it did fine. On “C” with fluxcore. (I have two identical Lincoln 120V models).
FWIW I was on a 50’ 12ga cord. Once I bury the two runs of 3/4” pipe I will have 10ga thhn feeding the shop which should help.
And I’m curious about your setup

fyi it was merely an attempt for the sake of entertaining myself, not a requirement. But if 2000W sorta works 3000W should let me do common things like sheet metal or fab up stuff which often is intermittent welding only a minute here and there. “Actual” welding for anything I’ll fire up the generator; plus next year I might set up a separate split phase aio system with 48V batteries for the shop.

Oh yeah, this was the small wire feed Lincoln Mig-10 - it is gas, but it can run fluxcore too.
I will swap it out today (Saturday, I have the whole shop to myself) and I need to re-run that last test now to get the data to calc Power Factor, per @Hedges notes. I need to get the data from the P3 while it is running in order to see the amperage and voltage and total to do the calc.

Try Watts / (Volts x Amps) = PF

If we read volts and amps with separate meters and multiply, that is apparent power.

Good idea, I can set up two meters and I-Phone to 'watch' the current and voltage while I run a weld. (pretty hard to watch the meters with a welding helmet on !)

Then I can report back what the data showed.

I have worse luck killing the kil-o-watt meters vs what I'm testing...

I end up popping the thermal overload on these like popcorn :

I hadn't thought about the current limit on the P3! guess I am just lucky that it took the 19.7A with the welder on "D" LOL!
Another good reason to drop down to "C" for today's test. Thanks for pointing this out, before I blow it up!

 

[OffGridForGood]

Try Watts / (Volts x Amps) = PF

If we read volts and amps with separate meters and multiply, that is apparent power.

Old watt-hour meters were some kind of motor that I would guess used voltage through a resistor to make magnetic field, current to make magnetic field, and accumulated the product.

I use a digital scope, have it multiply V x I sample by sample, then average.

Had a few things to do this morning and then had some time to play with the welder running on the MPP 2724.
Taking a good look at the meter I realized the meter does directly measure VA and PF you just need to select those settings! *who knew
So I set up the MPP 2724 on the same DIY 304Ah battery as yesterday, charged to about the same SOC again.
Put a spool of 0.9mm flux core on the Mig and took off the shield gas, set the Mig to "C" and feed rate to 2.5

Ran a few beads while the I-phone recorded the PF reading on the P3.
Voltage started at 118.5v under idle load, dropped to 117.9 while welding, bounced a bit between 117.9 and 118.5v

With the welder "on" but not doing any welding, the PF reading was 0.5, until the arc was started.
Welding it bounces around a bit but hangs out mostly at 0.85 most of the time, saw a few dips to .80/.75 one to .7 for a second.
When I stopped welding the PF dropped to 0.5 for a few seconds, while the welder is at idle - ie about 30W load.

So, @Hedges - what does it tell us if the welding results in a PF of 0.85?

 

[Hedges]

The 0.5 pF was just tiny sips of power to keep circuits running, so insignificant heating.

The 0.85 pF under large load means apparent power is 1.176x actual, and I think RMS current that much higher that it could have been with 1.0 PF load like a resistor. That doesn't seem too bad, not even as much as the recommended 25% margin for breakers.

 

[12VoltInstalls]

this was the small wire feed Lincoln Mig-10

Cool. So basically a newer version of my two mid 1990’s Lincoln 110’s.

Good little machines. I have no idea how many 10# fluxcore spools I put through the first one, probably well into the hundreds in ~30 years. It doesn’t have the guts to do 035 to my expectations but it runs 030 great.
The mig setup i bought ~2016-ish to do sheetmetal and exhaust. It was essentially unused just old. I just couldn’t bear the idea of switching out flux-to-mig constantly so I watched for a decent ’twin’ to come up and got lucky to buy the mig version including a full tank for $225. I run 0.023 exclusively.

I’ll be interested to see what you come up with mig-to-flux for numbers.

 

[OffGridForGood]

Cool. So basically a newer version of my two mid 1990’s Lincoln 110’s.

Good little machines. I have no idea how many 10# fluxcore spools I put through the first one, probably well into the hundreds in ~30 years. It doesn’t have the guts to do 035 to my expectations but it runs 030 great.
The mig setup i bought ~2016-ish to do sheetmetal and exhaust. It was essentially unused just old. I just couldn’t bear the idea of switching out flux-to-mig constantly so I watched for a decent ’twin’ to come up and got lucky to buy the mig version including a full tank for $225. I run 0.023 exclusively.

I’ll be interested to see what you come up with mig-to-flux for numbers.

They are handy, use them mostly for thin stock and tack-welding assemblies prior to stick-welding.
The Mig-10 is a very handy size for moving around, even outside far from the main power system since it will run on a genny.
Good idea to run two, instead of swapping around - I have an older one in a shelf that could be used for that. Thanks!
If you got the full set up for $225 you scored a good deal!

 

[OffGridForGood]

The 0.5 pF was just tiny sips of power to keep circuits running, so insignificant heating.

I thought this may be insignificant while the Mig is just idling.

The 0.85 pF under large load means apparent power is 1.176x actual,

So here, if we see current at 15Amps = the real current draw is x1.176 = 17.64A?
This is the power the inverter is actually subjected to, and heating up it's internals?

and I think RMS current that much higher that it could have been with 1.0 PF load like a resistor. That doesn't seem too bad, not even as much as the recommended 25% margin for breakers.

Now I am itching to try the test one more time, but with my 19kW main 48v system, and then the grid as supply to see if there is any difference.
I expect the PF a function of the load only - will make no difference what supply is used for the welder, the PF will remain unchanged?

 

[Hedges]

Yes, I expect PF to depend on load, not power source (so long as a sine wave.)

Whatever RMS current draw is measured is the actual current draw (assuming true RMS meter).
To deliver 1200W at 120V into a 1.0 PF resistive load would require 10.0A
If 0.85 PF load, 11.76A

The heating of wires, inductor windings, transistors (MOSFETS saturated, a few milliohms on resistance) is simply voltage across the resistance multiplied by current through the resistance, which is I^2R
11.76^2 / 10.0^2 = 1.38 times as much power dissipation (and temperature rise.)

Don't know if it is significant, but non-sinewave current like the clipped sine wave charging capacitor through diode when voltage of line goes above voltage in capacitor contains higher frequency harmonics. Those harmonics cause circulating currents in inductors (induction heating) and are squeezed toward surface of wires (skin depth), which would raise power dissipation further.

For occasional use equipment, just need to oversize system.
For inverter-drive motors like minisplit or my pool pump, would be nice to have PF corrected power supplies. I understand such products are more readily available in Europe, so I've been interested in adding external PFC front end.
I got some 1000W Lambda PFC modules but haven't used them yet because have to assemble support circuitry. We used some 1500W ones for an experiment at work (more highly integrated, just needed EMI filters and capacitors), had an intern design/build/test.

https://product.tdk.com/en/system/files?file=dam/doc/product/power/switching-power/ac-dc-converter/catalog/pfh_e.pdf