This would mean I have to write my entire worksheet in a single execution group, and split it where I want the compilation to end.  This isn't really viable for the sort of worksheets I write.  Thanks though

This would mean I have to write my entire worksheet in a single execution group, and split it where I want the compilation to end.  This isn't really viable for the sort of worksheets I write.  Thanks though

@Axel Vogt 

Well, all my equations are of the form a*X + b*Y + c*Z etc =0 (lower case represents constants, upper case represents variables).  There are no constant terms on the other side of the equals sign.  Obviously one trivial solution to this is that all of my variables are zero, and this is the solution LinearSolve seems to pick out - every time I run linearsolve I just get a column vector of zeros. 

Also, its not actually the variables themselves that I'm interested in but rather the relationship between the 'constants' a,b,c etc.  I'm studying electromagnetic waves in a plasma, and its these 'constants' that give me a dispersion relation, which right now is what I'm after.

@Axel Vogt 

Well, all my equations are of the form a*X + b*Y + c*Z etc =0 (lower case represents constants, upper case represents variables).  There are no constant terms on the other side of the equals sign.  Obviously one trivial solution to this is that all of my variables are zero, and this is the solution LinearSolve seems to pick out - every time I run linearsolve I just get a column vector of zeros. 

Also, its not actually the variables themselves that I'm interested in but rather the relationship between the 'constants' a,b,c etc.  I'm studying electromagnetic waves in a plasma, and its these 'constants' that give me a dispersion relation, which right now is what I'm after.

Ok, perhaps we can look at this from another perspective.  Among the constants in the matrix is a parameter I have labelled tau, now I'm only working to first order in tau, which means I can ignore any terms with tau^2, tau^3 ,tau ^4 etc in them.  When dealing with smaller matrices I've been able to calculate the whole determinant, and have then worked to order tau using the

convert(series(Result,tau,2),polynom);

command.  Now I'm pretty sure that if I could get maple to use this approximation within the Determinant() procedure it would be able to calculate it, but I'm not sure how to go about doing this.  I guess I'm going to have to write my own procedure to calculate the determinant and put it in manually.  If anyone has any other ideas I'd appreciate it!

Ok, perhaps we can look at this from another perspective.  Among the constants in the matrix is a parameter I have labelled tau, now I'm only working to first order in tau, which means I can ignore any terms with tau^2, tau^3 ,tau ^4 etc in them.  When dealing with smaller matrices I've been able to calculate the whole determinant, and have then worked to order tau using the

convert(series(Result,tau,2),polynom);

command.  Now I'm pretty sure that if I could get maple to use this approximation within the Determinant() procedure it would be able to calculate it, but I'm not sure how to go about doing this.  I guess I'm going to have to write my own procedure to calculate the determinant and put it in manually.  If anyone has any other ideas I'd appreciate it!