I would have been impressed if he made one that actually measured an hour.

I don't think that's possible because, even if he used a large enough glass and enough mercury, it doesn't flow evenly like sand does, so can't be used to measure time accurately.
The air in the bottom has to bubble up through the mercury to equalize the pressure (instead of flowing evenly through the sand) and get the flow going again, then it instantly becomes un-equal in pressure again which slows the flow to almost nothing, and repeat. Maybe that's predictable...but it would be some seriously difficult math because each round deals with different pressures and volumes.