Beyond the Qubit podcast

What Should Actually Move Cold in a Quantum Computer? | Stefan van Waasen, IceCirc

0:00
43:40
Reculer de 15 secondes
Avancer de 15 secondes

What should actually move cold in a quantum computer? The surprising answer is: less than you might think.


In Part 2 of my Beyond the Qubit conversation with Stefan van Waasen, co-founder of IceCirc, we explore one of the central challenges in scaling quantum computers: deciding which control electronics should operate inside the cryostat and which should stay at room temperature.


Stefan's philosophy is straightforward: do not move electronics cold unless doing so creates a clear system-level benefit. Every cable entering the cryostat brings heat, but every transistor operating inside it also consumes power and generates heat. The challenge is finding the right balance.


We discuss three important functions: biasing, multiplexing, and RF control. Stefan shares a concrete example from semiconductor spin qubits, where 12 cables could be reduced to 2 by moving biasing functions into the cryogenic environment. We also explore why multiplexing can reduce wiring and hardware, why high-frequency signal generation should generally remain warm, and how selected RF processing can move closer to the qubits.


Beyond these three functions, the conversation moves into local control, RISC-V processors, cryogenic memory, and the challenges of scaling beyond 10,000 qubits.


For anyone following quantum hardware, the key takeaway is that cryo-CMOS is not simply about moving more electronics into the fridge. It is about designing the right architecture and deciding where every function belongs.


💡 In this episode, we cover:

  • Why moving more electronics into the cryostat is not always better

  • How biasing circuits can reduce cabling from 12 lines to 2

  • How multiplexing allows control resources to be shared across multiple outputs

  • Why multiplexing only makes sense when its benefits outweigh the added complexity

  • Why RF signal generation should generally stay at room temperature

  • How RF shaping and frequency distribution can be performed closer to the qubits

  • Why the 4 Kelvin stage is important for cryogenic control electronics

  • How RISC-V processors could support local quantum control

  • Why quantum computers may need distributed, local memory

  • The potential roles of SRAM, DRAM, and memory-in-compute

  • What could limit cryo-CMOS adoption and scaling

  • The technical milestone Stefan would like to see by 2030

🔗 Resources / Links

  • IceCirc: https://icecirc.com/

  • Stefan van Waasen: https://www.linkedin.com/in/stefan-van-waasen-2ab3007/

  • Forschungszentrum Jülich: https://www.fz-juelich.de/en

  • Beyond the Qubit: https://www.youtube.com/@BeyondtheQubit

    https://open.spotify.com/show/7HZpSCz1w7a782e1B26MYA?si=6bccdb8c73f34519


🎧 Listen to all episodes → https://www.youtube.com/@BeyondtheQubit

https://open.spotify.com/show/7HZpSCz1w7a782e1B26MYA?si=6bccdb8c73f34519


If you work in quantum computing, semiconductor technology, or deep-tech investing, share this episode with someone interested in the engineering challenges behind scaling quantum computers.


Subscribe to Beyond the Qubit for more conversations about the technologies, companies, and business models shaping the quantum industry.


Disclaimer: This post is shared on a personal basis. I do not represent any company. This is not investment advice.


D'autres épisodes de "Beyond the Qubit"