We had a Mooney land at Newcastle last year with suspected induction icing. The pilot had overboosted the engine trying to maintain power. It lost most of its oil through the turbocharger and I believe most of the cylinders needed to be replaced. The conditions were perfect for normal carburetor icing. The Mooney was in VMC. The engine died on the runway.
I think people rule out icing far too quickly. Have people read the articles on clear air ice formation under the anvil of large thunderstorms in turbine engines?
My theory here is that induction icing formed at the butterfly.