Is it me or does this suggest that HMPE's are a form of reverse reaction.
We know non titration followed by WBD produces an excess of HMPE's,
so does using less catalyst (non titration) produce a marginal reaction, that can be, to some extent,
reversed if then heating for extended time at high temp (WBD).
Could it be that the esters are not properly bonded together, so are easier to split the methanol molecule from them,
but still bonded enough to pass a 27/3 test.
I'm no chemist, so this is only what my simple mind has come up with.