From J Prakt Chem, 1933, pp339.
A solution of 40 cc of nitromethane and 100 g. of trimethoxybenzaldehyde in 200 cc alcohol is cooled to 0 deg C and while it is stirred mechanically there is introduced a solution of 45 g. pure potassium hydroxide in 45 cc water and 90 cc methanol at the rate of about one drop per second, care being taken that the temperature does not rise. Fifteen minutes after the addition is completed the solution is poured into 500 cc concentrated hydrochloric acid mixed with sufficient fee to assure its presence throughout the slow addition and to maintain the temperature of -10 deg C. The precipitated nitrostyrene is separated by filtration and washing and may be purified by recrystallizing from 700 cc alcohol. The pale yellow plates which melt at 120-121 deg C are obtained in a yield of approximately 78% of theory.
Doesn't use cyclohexylamine, ammonium acetate, or GAA. Just keep the basic addition at -5C. At first, much of the aldehyde may precipitate from the alcohol, but as soon as you add about a mL of your base, everything will go back into solution. After the addition of the NaOH solution, you may notice the nitroalcohol solidify creating a slushy mixture - that's okay.
Also, to help make it easier in the dehydration step, dilute the HCl so that you can make stirring easier.
If you're determined to try it your way, go ahead and report back. If you want something that is easy and guarenteed to work, do it as above.
hest - mescaline isn't too bad. Some of those substituted-alkylthio-aldehydes can be a pain though! But in general SWIM doesn't believe in a "universal nitrostyrene catalyst". Every aldehyde has it's own unique character.