I once performed a solo concerto with orchestra:

During the depression, there was federal money for all sorts of unlikely projects to keep people in work. Some musicians got a grant to build a Hurdy-Gurdy, the crank-driven instrument you've seen pictures of, that accompanies Italian Organ Grinders and monkeys. The grant covered writing out parts for one of Haydn's Hurdy-Gurdy concertos, and programming the Hurdy-Gurdy to play the two solo parts. (That's right, there were two solo parts, not one.)

The musicians on this project must have been a little puzzled about what they were doing, because they built the wrong instrument. The Hurdy-Gurdy of old is a violin with a round, spring-drive mechanism that continuously "bows" the strings, and a set of push-down pegs that play notes. It sounds pretty sickly, but it's a lot easier to manage than a real violin. Haydn wrote concertos for pairs of thiese old Hurdy-Gurdies. It was child's play to program the new windup H-G to play two notes at once, heck, it can play a lot more than that.

The conductor of the Columbia University undergrad orchestra, Howard Shanet, specialized in digging up unusual music, and this find was a humdinger. The orchestra parts were easy to play, and how hard is it to play a H-G? You just turn the crank handle at the right speed, and the notes come out. Shanet decided that the librarian of a music collection in Philadelphia, who had often loaned us orchestra parts, would get the reward of performing the concerto in concert. For rehearsal, Columbia orchestra's own librarian would play the H-G.

The first rehearsal was a disaster. It was almost impossible to crank the H-G at a "wrong" speed, but our librarian could not stay in synch with the orchesra. For some reason, the problem was immediately obvious only to me, and in between "takes" I tried to coach the librarian, but he just didn't "get" it. Finally, an impatient conductor snapped at me, "You know so much about it, you do it." I toook over the H-G part and all was well.

The problem was finding the downbeat. You had to crank the H-G once, all the way around, for each measure. It seemed to be "human nature" to crank DOWN for the first beat -- that would be six o'clock if you imagine the crank going around a clock circle -- but the first beat was actually on the way up, around ten o'clock. I played H-G for all the rest of the rehearsals.

The day before the concert was a tense time for me. I wanted to perform the concerto, and I thought I really ought to. What was the chance that the Philadelphia librarian would find the downbeat without a single rehearsal? If I may be frank, I would say that my tension came from this: I could not decide what ached me more: the risk that the Philly librarian would make a fool of himself; or my great desire to be the soloist.
Hours before the concert, Howard Shanet informed me that the Philly librarian would not be able to make it. I still wish I knew what he told the librarian during their phone call. Anyway, I had a ball.