Quote:
Originally Posted by tazio
What were you making? Why were 65 pound ingots suspended ~11.5 meters (Did I do that back of the envelope math right? 15m/s final speed divided by 9.8m/s/s gives the time it fell; multiply by that by an average speed of 7.5m/s to find the distance.) above two foot diameter drums and how did that waste a multimillion dollar tool?
|
Look up solar and semiconductor wire saw processes for details. Meyer Berger saws. It didn't waste a multimillion dollar tool, it was 10k plus to repair for each failure and replace the spools and clean up the bird's nest of piano wire. We were not aligned with the home company on process, we were doing the R&D. They were not cutting monocrystalline ingots in Germany, but we were in Oregon and Washington at the time. They knew there were resources in the area doing semi wafer cutting for decades on mono crystalline ingots and tried to tap in. They got the wrong people 2 miles down the road from Intel making processors, not wafers.
We were cutting a lot thinner than semis, and trying to run semi fast, at 2x the wafers per ingot. Heat failure was an issue in the early development phase of trying to instantly become 24/7 manufacturing. I was the grinder engineer, they blamed me for surface roughness issues making it too smooth or too rough ad nauseum. It was a dept of chemical engineers running a machine shop. I was one of two mechanical engineers. Subsurface damage was a big issue for yield. Grinding produced 7 um of subsurface damage versus 30 um from the wire saws. On each side of the wafer, that was only 120 um thick. They never listened.
Silicon is tough to handle at that level, lot of yield loss. Some of it can be recycled. Mono square cutting had us putting 30% back into the mix as a matter of routine of the shaping process. Always recycling 30% of your material is a business plan that isn't getting us to grid parity via renewables.
Back on point, I drove my FRS today and it was good. This application is a no fail process via the strut bars.
Solar epoxy in this case is one that's made to fail to release the wafers from sacrificial glass plate. The trick was running the saws cooler than the release temperature. Cutting at 15m/s wire speed cutting 1300 wafers with ~100um wire is tricky. Math is likely off, but we were producing 1300 or so wafers per 500mm ingot. And yielding only 80% from that.
Silicon solar reprocesses way too much material. I think they should tap into the reclaim semi wafers and reprocess the waste stream. And to be fair, still don't see a honey comb packing of the circles to process into panels, sill making squares.