You don’t lose much at all from quality gold-over-silicon mirrors. American Photonics claims 99.6%. That’s so trivial, don’t worry about it.
Molybdenum mirrors are listed as 98%. However, they are nearly indestructible due to high thermal conductivity. I like to use molybdenum on the #3 mirror, the one most likely to get dirty from smoke, at our makerspace because this machine is 225W and a dirty silicon mirror will get so hot it bubbles off the gold plating. And, being a makerspace, most users are casual and don’t inspect and clean so this does happen.
Molybdenum mirrors get hot in operation due to the losses, but they can absorb a lot of wattage without overheating the front face. It’s literally a polished, uncoated face of a solid molybdenum puck, and molybdenum doesn’t oxidize.
I did have some lenses explode from the power density. Like the beam bored a hole through the zinc selenide and the user reports it was cutting poorly… like, yeah, no lens. The #3 mirror was moly but had such a baked-on coating of brown crud I couldn’t clean it. But, the part of the surface I could get the brown crud to come off of appeared undamaged.
In the 4 mirror setup I have seen, it was an HPC*, I think, you keep mirror 2 next to mirror 1 and mirror 3 is on the ‘top’ rail, next to the tube, shorter beam path.
Set up is trickier with 4 mirrors but losses are minimal.
Benefit (theoretical?) is mainly that you have a much shorter unsupported rail so less vibration at higher speeds.
*almost identical to a K40 until you spotted the extra mirror.
this is great info, i noticed the gold ones were lighter, but had no idea on the thermal and therefore durability issues with any soot on them. deal breaker for me, so you just saved me from aligning with gold mirrors for no reason.
Gold over silicon mirrors are still a fine choice.
Actually what I’ve come to notice is the simple loss by absorption isn’t the significant part of it. It’s scattering. I’ve seen more significant power loss that had me boggled as to where it’s going since nothing was getting hot. I have a Coherent power meter I can get accurate digital numbers live with.
One thing I’ve gotten a lot of return on is a thermochromic “mood ring” sheet. It’s useful for seeing the beam shape in some ways, but really difficult to avoid damaging the sheet of course.
That led to some confusing readings because #1 was scattering like 10% of the wattage, but my meter’s sensor is a couple of inches wide so that energy still lands on the sensor between #1 and #2 so I wrote the mirror down as good. But that 10% scattering misses the #2 mirror so the sensor shows the energy is missing right after the #2, but it’s not being lost at #2.
The thermochromic sheet could be put beside the #2 mirror and test fire the beam beside it. I could see a haze of IR energy all around, not really concentrated near the spot where a pulse would burn the tape. Neat to visualize.
I’d like to test various mirrors from various mfg for scattering some time, I think it’s way more important than we realize.
It’s still pretty rare to have a failure of gold-over-silicon mirrors. I did get this spectacular failure with 200W in a pretty narrow transit beam. Funny thing was, this mirror was in service with no prob for a long time with no prob, but once it decided to go bad, it went really bad fast. That is a hole melted in the silicon substrate and the gold plating is blistered across the surface. And I don’t see any evidence that the mirror had been smoked up badly beforehand. Maybe it could have and this damage hid the contamination, I could believe that. Never seen anything else quite like it.
