Well, there are a few issues with the question..
The earliest unambiguous evidence of hourglasses is this 1338 fresco and there are other definite references from that century but not earlier (ref 1,2). This means that the hourglass is in fact younger than the first mechanical clocks. (they were not yet common at that time though)
As the name implies, an hourglass measures a fixed amount of time such as an hour. There's a little bit of a problem with that in the middle ages though, because in those days an hour usually wasn't a fixed amount of time.
People kept track of the time by the sun, dividing the night and day into 12 hours each, from sunrise to sunset. But those things are only the same twice a year; So the hours of daytime would be shorter in the winter and longer in the summer. You had what's called 'seasonal hours', which is what was used for civil timekeeping. Sundials were not the only instrument used for this, you had (among other things) quadrants and astrolabes to measure the position of the sun more precisely.
But they still had the concept of 'equal hours', measuring from one point (sunrise, sunset, noon) of the day to the next and dividing by 24. (That goes back to the Babylonians) Which is of course the (within error) fixed intervals of time that a hourglass measures. That was mainly of interest to astronomers and astrologers looking at the movement of heavenly bodies, and to mariners who couldn't use sundials on moving ships. But it wasn't until the proliferation of mechanical clocks in the 15th and 16th century that ordinary society switched to that time-keeping. (ref 3)
I haven't found anything on how early hourglass makers calibrated their glasses, although the methods available were imprecise, an open question to me at least is whether there's any reason to believe early hourglasses were particularly precise. Ref 1 quotes Richard Polter, who in the Elizabethan era, well after mechanical clocks were widespread, wrote (in The Pathway to Perfect Sayling, 1605) about hourglass makers:
'who indeed careth but little what error more or lesse, is delivered by those glasses in 24 houres, nay in halfe an houre, with whom (as the proverbe is) an inch breaketh no square'.
(the proverb essentially means 'Close enough. That's fine')
So most of them were not very precise, but they did not need to be - they were mainly used for measuring things like the lengths of sermons or work shifts, keeping track of a fixed interval of time, but not one that needed to be exact. Therein lies Polter's problem because at the time they'd started using chip logs to measure ship speed in conjunction with an hourglasses. Knowing their course and speed was integral (pun intended) to figuring out where they were using dead-reckoning. Although it was used since at least the early 1500s, Mathematician William Bourne was first to explained the method in print (A Regiment for the Sea 1574) Using "either a minute of an hour glass, or else the known part of an hour".
But by that time clocks were fairly common and clocks were used to calibrate. Here's a picture from a bit later (and in Nuremburg) of a hourglass maker in 1698 (in Abbildung Der Gemein-Nützlichen Haupt-Stände) who on the following pages is described as using an 'Eich-Uhr' - a calibration clock.
Sources:
(1) A.J. Turner (1982) ‘The accomplishment of many years’: Three notes towards a history of the sand-glass, Annals of Science, 39:2, 161,
(2) R.T. Balmer, The operation of sand clocks and their medieval development. Technology and culture, v. 19, Oct., 1978: 615-632
(3) Clocks and Timekeeping in Glick, Livesey, Wallis, Medieval Science, Technology, and Medicine: An Encyclopedia
