Using \partcombine with explicit voices

[Urs Liska]

Hi,

I'm trying for the first time to engrave partcombine-d parts, and I'm 
running in an issue for which I don't see anything in
http://lilypond.org/doc/v2.18/Documentation/notation/multiple-voices.html#automatic-part-combining.

Consider the attached example. As it it it compiles fine. But if you 
exchange the comments and use the lines the explicitly instantiate the 
voices for the music variables it doesn't work - lily complains about 
clashing note columns and gives the attached output.

It seems that \partcombine doesn't work when the two music expressions 
are enclosed in explicit voices. Is that true? is that expected? is that 
necesary?

This poses problems for me because in the current context I'm forced to 
enclose a number of consecutive music variables in one Voice context in 
order to be able to write spanners or slurs that cross variables (i.e. 
start in one and end in the other).

Is there a solution for my problem, either by making some settings so 
\partcombine works or by massaging the Voice context to spit out a 
"bare" music expression again?

TIA
Urs

voiced-partcombine.ly
Description: Text Data

voiced-partcombine.png
Description: PNG image