Have I said I hate spoilers?
So true about spoiler lift up! I recall flying my Airtronics Cumulus and Sagitta 900 at “high" speeds between thermals and seeing the spoiler pop up. In those dark days I was using torsion bar springs, dental rubber bands or small magnets to hold down the spoilers. Combined with draw strings to pull open the spoilers. These were horrible setups to keep adjusted.
When I was last at Aloft looking for programing help, Wayne was a bit surprised that I was using a 7 point curve for my spoiler to elevator compensation curve. Normally the instructions given with spoilers is to add "up" as they are deployed. Nominally this is correct. But once I started to use positive control set ups for the spoilers I noticed that the airplane does not respond in a linear fashion to their deployment. I found that at low Cl numbers (like dropping into the landing pattern) the aircraft will actually balloon as the spoilers are first starting to be deployed. I think this is because the spoiler is lifting the air that is still attached. This results in the air acting like it is working against a thicker airfoil (A+ in the drawing) with more camber. With the lift from the "added" camber the plane will pitch up needing a bit of down compensation to the elevator spoiler comp curve (-2 in the drawing).
As the lift and drag of the wing are changing rather abruptly early on in the deployment I find that the overall shape of the comp curve is asymptotic. The elevator comp reaches its full value rather early. Then as the spoiler reaches its full height, the disruption to the lift is rather constant but the drag keeps going up. So to keep from stalling I take out some of the up compensation add earlier.