A rosette weld (aka a MIG 'spot weld' aka a 'plug weld') is done, at a basic level, just like any other weld.
Set/adjust the weld parameters to get a 'proper' weld. That means enough 'power' to melt properly into the piece(s) being welded together without instantly just blasting a hole right through things.
A rosette weld is pretty much just a circular form of a lap weld. So, just like a 'regular' lap weld, adjust the weld parameters so that you can achieve the desired penetration into the layer/sheet underneath and not instantly blast a hole right through it and still have enough power/heat to melt the edge of the hole in the upper layer/sheet.
The only real complication you have introduced into this situation is by having different thicknesses in the two sheets/layers. If they are just a little bit different in thickness it's not too bad, it gets a bit more difficult to do when the thicknesses are very different from one another (say, 20 ga sheetmetal onto 1/2 inch plate takes some practice
).
Hole size in the top sheet/layer can be anywhere from about 1/8 inch to 1/4 inch or so. That usually isn't so critical.
OK, you say the top sheet/layer (with the hole) is the thinner one. That's good, as you should be able to set the weld parameters for the thicker bottom sheet/layer and just let the puddle then wash onto the edges of the top sheet/layer.
You also have to make
SURE that you clamp or otherwise hold the two sheets together when you want to make the weld. Line everything up and then clamp it (or stick a heavy weight onto the top layer if gravity will let you use that to hold the layers together). You really can't weld air.
You can use a relatively small hole in the thinner top sheet/layer, and then just make a weld puddle right in the middle of the hole onto the slightly thicker bottom sheet/layer. That will probably let you get a puddle that first melts/fuses/penetrates into the bottom sheet/layer and quickly fills the hole and then melts the edges of the hole. Done.
If you use a slightly bigger diameter hole (1/4, 5/16 or 3/8 is plenty big there) in the top sheet/layer, then you have to do more of a lap weld around the edge of the hole and then once that is all melted/fused together you do the swirl towards the center of the (formerly there) hole and fill it in. Done.
I kind of like doing rosette welds. They seem kind of 'fun' to me.
Step one with
ALL manual welding is, you have to be able to
SEE the weld puddle! Not the bright light of the arc itself, but the puddle of molten metal being formed
by the arc.
Step two is to use the correct weld parameters. 100+ amps on sheetmetal is just going to quickly blow a hole right through things. 30 amps is barely going to be enough to melt/fuse into 'thicker' sheetmetal (16 gauge?), although the MIG filler wire electrode will still melt pretty much just fine.
Which is why MIG (GMAW) is
notorious for making a weld that
looks OK, but is just sitting on top of the base metal. Classic "inadequate penetration", aka "inadequate fusion" aka "cold lap". The filler wire melted just fine but there was not enough heat to properly melt
into the base/parent material. So the pretty little weld 'bead' is just sitting on top of the base metal and not melted into it.
Clean the metal (no rust, grease, oil, paint, mill scale, mustard, etc, etc), clamp the joint, use 'proper' parameters, make weld while watching the weld puddle and not the bright arc.
C25 shielding gas and 0.023/0.025 wire should be fine to weld sheetmetal with a 120V MIG welder.
For parameters, step one is find out the thickness of the thicker metal piece. Then use the welder's parameter chart and set the machine. Do a practice weld or two and then maybe do a little bit of fine adjustment is desired.
For the voltage setting, too low of a voltage (for the selected wire feed speed) and the wire will 'stub' into the work and push back the gun. Too high of a voltage (for the selected WFS) and the arc is unstable and there is excess spatter (parameters may be nudging into globular transfer mode instead of short-circuit transfer mode). Set voltage (for that WFS) halfway between the two extremes.
For the WFS, more WFS = more amps = more heat/power. Turn the WFS too high and you may not have enough voltage for that WFS and then you get wire stubbing (gun pushes back from the workpiece as you try to weld). Turn the WFS too low (for the voltage setting) and you hvae too much voltage for that WFS and you get an unstable arc and too much spatter.
For GMAW, the parameters all have to work together. The shielding gas type (C25 or plain CO2 for steel), the wire electrode diameter (0.023/0.025 vs 0.030. vs 0.035 solid steel wire are most common for the 'small' machines), the voltage setting, and the WFS setting all have to work together.
Do some practice welds
FIRST.
