Terbium has run the AmCl/HCHO reaction. I have run the hex/HCl reaction. There might be slight differences in the temperatures between these two reactions because the hex/HCl variation has a bit more AmCl equivilant than the AmCl/HCHO version.
The way I ran this reaction to test out Terbium's suggestion was over the course of three days. As each day progressed, I would turn the heat up a bit more. After stirring at 104C for a day, nothing happened. I went up to ~104C vapor temp (this is 110-115C liquid temp) and the mixture started bubbling as I normally saw when hitting this temperature faster (mostly gas bubbles, some vapor bubbles).
Keep in mind that the "work up" of this reaction requires the distilling pot to be boiled dry, so except for time spend at each temperature, there was no difference in the overall method. There was also no difference in the approximate yields.
The odd thing is that both Terbium and I stand by our recommendations. I think the best piece of advice is to simply "go very slow" with heating.
Honestly, this is a pretty hard reaction to fuck up. Eleusis did it on a hotplate with a frying pan and still got some sort of product that could be used for amination. (*but do not do this, your product will contain significant amounts of inactive N,N-dimethyl- product*)