Re: extendedLagrangian for non-scalar colvar

From: Jérôme Hénin (
Date: Thu Dec 10 2015 - 11:18:39 CST

Hi Jerry,

That's correct, extendedLagrangian only works for a scalar colvar. You can
access individual Cartesian coordinates as "distancez" colvar components,
setting the axes to x, y, and z, respectively.


On 10 December 2015 at 17:45, Jerome Karp <>

> Dear NAMD users,
> I am trying to use the extendedLagrangian feature for a non-scalar colvar
> (specifically, the "cartesian" colvar new to NAMD 2.11). When I do this,
> though, I get this error message: "colvars: Trying to assign a colvar value
> with type "scalar number" to one with type "n-dimensional vector"." Is
> this because the extendedLagrangian only works for a scalar colvar? If so,
> is there a way to use an extendedLagrangian for each coordinate of an atom
> (i.e., add a fictitious particle corresponding to each of the x, y, and z
> of a given atom)? It seems like the only way to get coordinates of an atom
> as a colvar is using the cartesian colvar, or use the distanceVec colvar
> with one of the groups being a dummy atom; in either case, the colvar is a
> non-scalar, so that wouldn't work.
> Thank you for any help you can provide.
> Sincerely,
> Jerry Karp

This archive was generated by hypermail 2.1.6 : Tue Dec 27 2016 - 23:21:38 CST