Hi, Scott
Good news
I checked like you have said, using paper and pen, like I did before when I was practicing, and the trace left by the ink was similar at every point of the test. I also run the Gcode three times and observed how the z axis moved up and down with no sudden move like it happened on the "O" before. I did a little study on the code itself, trying to figure out how you guys understand it and I noticed the values related to z were consistent every time went up and down. Good practice. I think I've learned a little thing right there.
I went back to the sheet of aluminum I was using and I found out why this accident happend. At one moment, I was trying to adjust the depth cut and I made a mystake, so the bit went thru the sheet leaving a hole in it. I never thought of that hole as a factor to create uneven on the engraving surface, so the next time I placed the sheet to try a different depth value I had the bit going thru the sheet again. I believe now it was because the uneven created by the hole on the sheet.
Well... I went back to the experiment. I placed a good peace of sheet on the bed surface, making sure it is flat even (pressing down with my fingers, though). Previously, I went back to that "stupidity" of making the depth cut value a negative one (-0.012), I created the Gcode, I opened it in Mach3, sent it to engrave with a simple prey ( Just " whatever") and VOILA! The thing printed the characters evenly. I don't give you guys the picture because the printing is not what I need to do, but I will post the pics. No doubt about it.
Thanks for your help guys and I'll be back