ASML’s EUV Tools Have A Throughput Problem, But Lyncean Has The Answer By Scaling EUV Light Source Power An Order Of Magnitude
A TSMC fab with 10 EUV tools running N3 wafers would only be capable of ~15,000 wafers per month at current power/uptime.
Great read, thanks for sharing! I have so many questions...I understand that more power would lead to shorter exposure times, and that the Lyncean system would also simplify maintenance, so these are the 2 main drivers for increasing throughput, is this correct?
A higher power could potentially have a side-effect on resist performance, for example?
For the sake of the argument, if the Lyncean system was commercially available today with a cost of 30M€ (their approximation), would that make ASML's NXE cheaper or more expensive?
Sub component cost for ASML EUV machines isn't really public. Lyncean tech isn't exactly ready for high uptime quite yet. Lots of issues with mirrors and reflectors.
Light source downtime including collector mirrors is currently biggest bottleneck on uptime and power on throughput. If power were to rocket up, could reengineer machines for much shorter exposures.