From looking at the little of Sino-Tibetan: A Conspectus that Google Books would allow me to read I was able to work out that aspiration in many Tibeto-Burman languages operated in terms that plosives at the start of words, either just voiceless ones or both voiceless and voiced ones, were aspirated by default and were only deaspirated by many prefixes coming before them (but not all prefixes, e.g. m- and ʔǝ- in Tibetan still allowed aspiration, and r- and s- in Burmese also allowed aspiration). Note though in some languages voiced stops became unaspirated voiceless stops, while in others voiceless stops often became voiced stops.

The system I am going to use is going to be relatively conservative, in that voiceless stops will become aspirated word-initially except after prefixes except after some of m-, ʔǝ-, r-, and/or s-, except that they will be deaspirated in affixes. Voiced stops will be preserved but deaspirated.