The welder selection decisions?

[Buffalo21]

The welder you linked to is DC tig, with scratch start, you need high frequency ac for aluminum.

Greg

you can DC tig aluminum, NASA and the aerospace industry does it all the time. It’s a lot harder, more demanding, that why almost everyone else uses high-freq AC.

 

[graham-xrf]

you can DC tig aluminum, NASA and the aerospace industry does it all the time. It’s a lot harder, more demanding, that why almost everyone else uses high-freq AC.

With modern inverter switcher electronics, they are switching at high frequency anyway. That is what allows the smaller, lighter ferrite inductors and components to replace the big heavy 60Hz transformer. With transformers, the surge current is high, but once going, they can be efficient.
Once you have switcher inverter technology in use, then configuring to make it AC, or pulse-width modulated DC, of either polarity is possible.

Assisted scratch start, with smart anti-stick-on and overload detect tricks may be the feature for some. I think RF ignition may be something else. It is a value feature that would normally be mentioned separately, instead of just being there.

Weld procedures used by NASA are subject to insane-level procedure and scrutiny and testing, where they know if it is done a certain way, with certain kit and procedures, the result is extreme reliability predictable. They often cannot take advantage of more modern kit without having it be explored, tested for every extreme scenario they can dream up, and certified for use. The same goes for most other kit. By the time it is qualified to fly, "new" versions are out there.

 

[graham-xrf]

Maybe European amps are different than this side of the pond, but my 180 amp tig at max will trip a 50 amp breaker on 220 v. Its transformer technology but can't see that would matter.
The welder you linked to is DC tig, with scratch start, you need high frequency ac for aluminum.
Greg

I assure you the amps are the same both sides of the pond!

A transformer switch-on surge certainly can trip a breaker, especially if it is of a type that was intended to supply resistive loads, such as heaters, stoves, lights, kettles, etc. Also, many transformer types have an adjustable core power control, where part of the magnetic laminations core is moved in and out to adjust the mutual inductance coupling, as a means of power control.

Just before switch-on, the transformer core has no magnetic field in it. The core is empty of flux. It's a transformer. There are coils of copper in a primary winding which have a magnetic inductance, which limits the current if it is changing all the time, as in 50/60Hz AC. Once running, the energy from the ever changing field does not need so much primary inductance, and to save the copper and iron costs, and weight, many transformers simply don't include it.

At switch-on, into a core with no field, the only thing limiting the current inrush is the (low) DC resistance of the primary coils. What inductance is there slows down the rate of rise of the current, but without enough copper and iron, that surge can be 10x or 15x more than the average running current, but only briefly. Once the magnetic fields are established in the core, building then receding, then reversing and repeat, as with AC voltage, the current drops to very low, taking only a tiny "magnetizing current" to keep the field going in the core. Only when you then take energy out of the (lower voltage) secondary winding, and start welding, does the current rise as needed.

The trip switch may see the first milliseconds of that surge, yet not trip out. Magneto-thermal trip switches intended for inductive loads (transformers, motors, etc.) in UK, use Type-"C", taken from A, B, C, D. The trip will allow a surge, without popping out, for a certain delay, in milliseconds, to give enough time for the mains cycle to get past the surge. A short-circuit will trip the magnetic part after that. An overload current. though not short-circuit will trip it out from the thermal part after a good many more seconds. Trip switches can become touchy, and simply let go at a lower current than printed on their rating, especially if they have had cause to be re-set often.

This may not be the case with your welder. Once the AC is switched on, the field should be OK, unless the transformer is inefficient.

 

[addertooth]

Graham-XRF,
It always pleases me to see people paraphrasing Lenz laws in a functional way, and I agree with your summary.
It would be remiss of me not to mention inrush current on Switching Regulator (IGBT) power sources as well.
Switching supplies have the raw AC, which is rectified to pulse DC, and then uses that to charge a large bank of capacitors
to smooth those pulses to a more consistent DC current. A well designed switching supply will have Inrush Current limiters
which keep that huge gulp of initial power from tripping breakers (or popping fuses on older homes). However, even
with inrush limitation on switching supplies, the start up current always exceeds the operating current of those supplies.

Summary: Both transformer based, and switching based supplies can draw gobs of current when first switched on.
Loads are often described as Resistive or Reactive, with reactive being either inductive or capacitive. With reactive loads
you can do things to tweak the "power factor" on them to make them "look" more like a resistive load, but that is
a much longer topic.

Where switching supplies shine is their higher efficiency in conversion, as the typical figure for transformer efficiency
is only around 80 percent (for 50/60 Hz transformers), and high efficiency switching supplies can hit up to 98 percent (and custom built can exceed
99 percent efficiency of power out versus power in). As with all electronic devices, any electricity which is not used to
perform "work", is given off as (potentially) destructive heat.

 

[f350ca]

Your responses prompted me to do some research, didn't realize inverters were more efficient than transformers. Just the same I'll stick, pun intended with my old school transformer welders. My Miller tig is good to about 140 amps output, after that it trips the breaker. Guessing it uses a higher arc voltage than my 250 amp mig which never has tripped the breaker, mind you it never needs close to full output with .035 wire. Again my ancient Miller stick has never tripped the breaker, can't remember how high I've set it though, think its a 250 amp machine as well.
Yes you can weld aluminum with DC, tried it one day, repaired the skeg on a friends outboard. Couldn't get a nice arc, didn't sound right, weld wouldn't flow but got it done. Took a little more flap disk sanding than normal to make it look pretty but he was happy. Went to turn off the machine and realized it was set to dc out after welding steel. Hard slap to the fore head fixed that. lol

Greg

 

[graham-xrf]

Where switching supplies shine is their higher efficiency in conversion, as the typical figure for transformer efficiency
is only around 80 percent (for 50/60 Hz transformers), and high efficiency switching supplies can hit up to 98 percent (and custom built can exceed
99 percent efficiency of power out versus power in). As with all electronic devices, any electricity which is not used to
perform "work", is given off as (potentially) destructive heat.

Phew! Thank you - I am glad I got that right!
Re: Energy storage in IGBT inverter power supply systems..
Yes - there is stored energy in capacitors, but with modern switched mode power supplies, it no longer has to be enough to hold up the voltage to a relatively low "ripple". The energy is stored cycling between an inductor, and a capacitor operating at many kHz. Upstream, where the DC bus has the rectified mains with some capacitor storage, the 60Hz ripple may be large, but the control loop of the switching regulator is easily fast enough too ride up and down the residual AC to still provide a regulated pure DC, or other waveform as needed.

The better ones have what amounts to a "pre-switcher" stage to do power factor correction, so that the shop supply sees what looks like a resistive load, with the current taken being in phase with the voltage. Even more upstream, expect some relatively heavy inductors and capacitors to capture the harmonic energy from switching transients - the EMC filter.

The biggest switch-modes I ever had to deal with were 2 x 70kW servos pumping aviation fuel, which could be "back-driven" by 60HP propeller turbines (RATS). The electricity coming back could no longer be dissipated in resistors, for which I had used re-purposed house heating radiators wired in series-parallel during testing. The unused returning electricity would only drive the bus voltage up to destruction. The only choice was to use two drives in an arrangement to return the energy to the grid, in effect have it end up helping supply all else at that site. Now, with PV panels on roofs, many folk have exactly such electronics in their homes.

A small point about surge. Transformer inrush we understand. What happens with inverter welders is different. Even when it is on and running, there is a transient grab at input current when you strike the arc. The voltage drops, and the regulator slams it's PWM to near 100% for some milliseconds until the feedback lets it get a grip. The bandwidth of the control loop - the speed of response to the load transient, is what decides the temporary gulps of current that might take out a trip switch. The more real weighty hardware in inductors and capacitors for hold-up energy storage there is, the less the hit on the supply current. That would add to the weight and expense of the more "quality" kit.

I know it's odd, but I know much more about IGBT switching power supplies than about actual welding and available welders, for which read "relatively zilch"!

 

[addertooth]

I am surprised they use insulated gate bipolar transistors for the low voltages of all form of welding. The old classic Hex-FETs are more robust at the high current of welding than IGBTs. I would bet there must be a rather high counter EMF spike associated with arc welding (specifically on the make-break initiation of it), which is why they have to use IGBTs as versus HexFETs.

Certainly energy can be stored in inductors (capacitors do it in a more easily managed fashion), I have used inductors for the storage which is released when the field is collapsing... which can be fairly substantial. However, the huge gulps of power needed to weld can be problematic during the zero-crossover point on 60hz applied power (if you want consistent power delivery), which happens twice each cycle. And yes, I must admit, most of the power supplies I engineered and built had insanely low ripple figures, as they powered instrumentation which worked with nanoVolt level signals. Most of what I designed were boost or buck regulators. They critical ones had carefully calculated inductor (and core) values to achieve insane efficiency levels (due to them being used in very remote locations).

But yes, like you, I am very familiar with classical transformer-based linear power supplies, as well as switching supplies. Both certainly have the reasonable purposes, advantages, and disadvantages. I suspect both of us know far more about power-supply design, than welding.. but every craft can be learned.

 

[graham-xrf]

One thing about switch-on surge I did discover when I encountered (big) spot welders was in the solid state relays, and zero-crossing switching. This is about the SSRs that switch at the instant the gate demands, as opposed to those which wait for the AC voltage to be crossing zero. The latter are just fine for resistive loads like furnaces, and controllers can deliver burps of a counted number of cycles as needed.

Blunting the switch-on surge tripswitch smack
Getting to the instantaneous switchers, as used to be operated to deliver parts of a phase cycle, with all the problems of EMC, and distorting the supply voltage, like SCR devices that could switch on at the controlled instant, but would have to deliver the rest of the cycle, only able to switch off when the polarity reversed. Instantaneous FET or transistor devices can be switched on or off at will, though as you point out, the consequences can be very high voltage stored energy back emfs that can destroy semiconductors unless one captures it. So we get to switch-on surge trick.

If the instant one switches is not at zero voltage, (or current with a different timing), but chosen instead to be about 15% past the peak, then the voltage driving the surge is already on it's way down toward zero. It massively blunts the surge. By then, the core field is established, inductance is there, so as the voltage crosses zero and reverses, the transformer is functioning. This trick can turn a x5 overcurrent surge into a little 20% bump.
- - - - - - - - - - - - - -
Related to actual welding of aluminium:
What I am getting is that it is possible with all types of welder, though with some you need special rods, etc.
I see there are even special rods for "welding" aluminium with MAPP gas that looks in style a bit like brazing.
I also get it that for aluminium, TIG is considered best, and within that style, AC is a standard option, but if DC is used, the polarity has to be positive workpiece, negative electrode.

The TIG welds look so beautiful, and yes, at some stage I may well want to weld aluminium, but I get the impression that it can be a mistake to plunge straight into TIG with a dedicated machine. The advice is surely to start with stick welding, and learn how. HM members may have opinion on this. You can see why I was tempted by the Polish machine, so I could dabble with any.

 

[Aaron_W]

One thing about switch-on surge I did discover when I encountered (big) spot welders was in the solid state relays, and zero-crossing switching. This is about the SSRs that switch at the instant the gate demands, as opposed to those which wait for the AC voltage to be crossing zero. The latter are just fine for resistive loads like furnaces, and controllers can deliver burps of a counted number of cycles as needed.

Blunting the switch-on surge tripswitch smack
Getting to the instantaneous switchers, as used to be operated to deliver parts of a phase cycle, with all the problems of EMC, and distorting the supply voltage, like SCR devices that could switch on at the controlled instant, but would have to deliver the rest of the cycle, only able to switch off when the polarity reversed. Instantaneous FET or transistor devices can be switched on or off at will, though as you point out, the consequences can be very high voltage stored energy back emfs that can destroy semiconductors unless one captures it. So we get to switch-on surge trick.

If the instant one switches is not at zero voltage, (or current with a different timing), but chosen instead to be about 15% past the peak, then the voltage driving the surge is already on it's way down toward zero. It massively blunts the surge. By then, the core field is established, inductance is there, so as the voltage crosses zero and reverses, the transformer is functioning. This trick can turn a x5 overcurrent surge into a little 20% bump.
- - - - - - - - - - - - - -
Related to actual welding of aluminium:
What I am getting is that it is possible with all types of welder, though with some you need special rods, etc.
I see there are even special rods for "welding" aluminium with MAPP gas that looks in style a bit like brazing.
I also get it that for aluminium, TIG is considered best, and within that style, AC is a standard option, but if DC is used, the polarity has to be positive workpiece, negative electrode.

The TIG welds look so beautiful, and yes, at some stage I may well want to weld aluminium, but I get the impression that it can be a mistake to plunge straight into TIG with a dedicated machine. The advice is surely to start with stick welding, and learn how. HM members may have opinion on this. You can see why I was tempted by the Polish machine, so I could dabble with any.

TIG is its own thing most closely related to oxy/fuel welding, but still quite different if that makes any sense. Being good with stick or MIG is of limited value when learning TIG.

I really wanted a MIG welder (need is debatable), but I got the Multi-process so I could dabble with TIG. I have not regretted that decision, but I sprung for a higher end AC/DC machine, if I was going for TIG I was going in with both feet.

In reality I have so far used the machine as an overly expensive MIG welder because I am a passable MIG welder, but not a very good TIG welder. I hope with time though to get better and use the machine to its full capacity. I will never get better at TIG without practice and I can't practice without a machine (or at least that was my thought process when throwing the budget out the window). I have projects I needed a MIG for, once I'm past that then I can start to find uses for TIG.
I bought the welder with a 15-20 year growth plan in mind, so not using it to capacity in the first year of ownership is not a failure in my view, in fact I'd be worried if I was already reaching its full capacity.

 

[DavidR8]

TIG is its own thing most closely related to oxy/fuel welding, but still quite different if that makes any sense. Being good with stick or MIG is of limited value when learning TIG.

I really wanted a MIG welder (need is debatable), but I got the Multi-process so I could dabble with TIG. I have not regretted that decision, but I sprung for a higher end AC/DC machine, if I was going for TIG I was going in with both feet.

In reality I have so far used the machine as an overly expensive MIG welder because I am a passable MIG welder, but not a very good TIG welder. I hope with time though to get better and use the machine to its full capacity. I will never get better at TIG without practice and I can't practice without a machine (or at least that was my thought process when throwing the budget out the window). I have projects I needed a MIG for, once I'm past that then I can start to find uses for TIG.
I bought the welder with a 15-20 year growth plan in mind, so not using it to capacity in the first year of ownership is not a failure in my view, in fact I'd be worried if I was already reaching its full capacity.

I love having both MIG and TIG capability.
MIG feels rushed somehow whereas TIG feels like meditation.


Sent from my iPhone using Tapatalk