Chris, you make a good point re. greater mass of the laser head unit when upgrading just that element in the CNC system. I upgraded twice, from 3W (single diode) to 12W (dual diode), and from 12W to 24W (quad diode) over the past 4 years. In each case there was more than a doubling of mass due to heat sink and fans. Fortunately I have a CNC machine that is capable of handling the added mass and then some.
re. durability - my latest laser unit is the Neje E80. So far so good. I like the near doubling of power output from the A40640 dual diode, clearly makes for faster run times. In the previous laser units, I kept the max power output to 85% or less to keep the diode temp cooler, which I understand to be key to long duration/life of diode(s). Neje has a temp monitor built in, and I find with the E80 that it runs 10degC cooler at 85% than did the previous model, and at 100% pwr it’s still 8degC cooler than previous. The mass of the heat sink and dual fans likely explains that.
to note- For 3-5mm BB and luan ply, although the E80 is capable of cutting through in 1 pass, I find multiple passes at higher speeds to be better, it results in less pitch/tar condensate on the underside of the cut, both on the wood and on the metal bed shield.
Lastly, one down side is, the beam divergence at the focal point seems to be greater based on the kerf width at full power vs. less pwr. below 70% and diminishes width as pwr decreases. Perhaps someone more familiar with optics can better explain the physics. I found this post and video to be quite revealing on what this phenomenon may be: 👁 EXCELLENT READING - Russ talks about Laser Lenses