It's reportedly much faster and cleaner to do the paste route. The paste should look like paste: It should be hydrated but not watery, and all of the DMT-fumarate should be completely dissolved--essentially melted. It should be stirred thoroughly and allowed time to soak and react (about one hour). It should be completely dried before adding acetone. Stirring in anhydrous magnesium sulfate will help not only to dry the paste but to more thoroughly dry the acetone as well.
Three thoroughly stirred acetone washes on the powder should be sufficient. Evaporate the acetone on a dish at room temperature. To confirm that the mixture has been completely exhausted of its DMT, just do another wash and evaporate on a clean dish.
The result should be yellow and white crystals, unless there's a significant jungle-spice contamination (usually if you did naphtha pulls before the xylene pulls), in which case the result will be dark and oily. If the latter is the case, It would be advisable to leave everything on the dish and pour around 50mL of warm naphtha directly on the dish and scrape within the naphtha to loosen and break everything up. Then just decant or pipette off the naphtha in to a container for freeze-precipitation. After everything has precipitated out, decant, repeat and just continue to precipitate in the same container without scraping out the crystals. Do this three times before collecting and drying the crystals and then to perform the process at least once more to gauge whether the product has been exhausted.
If SWIY's having trouble with water contamination when evaporating, there is the option of drying the puddle with magnesium sulfate, scraping it up and trying again directly from the resulting powder, or by mixing that with the original powder and extracting.
SWIM should really get going on the section of the handbook regarding this. He'll post a thread in the FASA section linking to that and to the FASA section of the handbook when its complete, so keep an eye out.