It has been done.
Several years ago Allen Whitmore and I conducted some experiments involving three full length Aerotech delay grains. They were separated as exactly into two pieces as possible and he burned three halves at his altitude (about 400’ MSL) and I burned the other three halves at my altitude (over a mile MSL). In this way we were both burning sections from the same full length grain, without any concern about being from different lots.
We timed the burning from the time one end ignited until sparks emerged from the other end. I used an app called Video Stopwatch which allowed me to put markers down and then see the time elapsed between markers.
The delay grains burned 8% longer at my altitude than at his. The difference was very consistent. Allen published the results as part of a report on variability of delay grains that Allen submitted to the Tripoli Report and the results (more accurately than my memory can do) are there for all Tripoli members to read.
I believe that at your altitude you would see a similar increase, perhaps 12-13% longer than at sea level.
But, what this really highlighted is that even though the delay material burned at a very consistent rate, there are a lot of other things that affect the duration of the delay grain when used in a rocket.
First, the burn duration of most rocket motors is affected in the same way. At higher altitudes they burn at a slightly slower rate. Now that ambient pressure difference is quickly made insignificant by the high pressure within a rocket motor, but its effect is not zero.
Also, the burn rate of a delay grain varies enormously during the “life cycle” of a rocket motor. In a BP motor (which is an end burner) the delay grain doesn’t ignite until the propellant has completed burning. It’s ignited by the tail end of the propellant.
But almost all APCP motors have a core as part of their geometry which means the delay grain is burning the entire time the propellant is burning. The high pressure of the rocket motor means that the portion of the delay grain that burns while the propellant is burning burns at somewhere around five times the rate that it burns once the motor internal pressure has returned to approximately ambient. If something happens to cause a rocket motor to burn 1/5 second longer, that’s a full second that is removed from the portion of the delay grain that’s intended to burn after the motor burns out.
Similarly, if the motor burns faster than designed, the delay will burn longer afterwards.
Finally, if the delay grain isn’t ignited at exactly the same time as the the propellant that can really throw a wrench into the works. That sometimes happens because a person still has some assembly lubricant on their fingers when they handle the delay.
The fact is, that it’s amazing that delays burn as consistently as they do in rocket motors. I don’t think a person has any practical reason to expect that a delay grain will be closer than one or two seconds of the targeted duration, but that really is close enough.
The other thing is that I have noticed that my results with delay grains are now much more consistent than they were when I first started out. I remember being upset about bonus delays. I don’t see that happening for myself anymore. I believe my techniques are more consistent than twenty some years ago.
Finally, if you truly want accuracy in timing, use an altimeter to control deployment.