check your initial conditions. Hard to tell what is wrong without seeing the model, why not load it up?

@Volker Lehner 

Agreed. Do you have experience with other programs where you would say that Units are implemented as they should?

needs attachment

it takes longer to type the formula into Maple than doing this by hand. Simplify the square bracket.

@taro 

sometimes it is easier just to manually rewrite your expression the way you want to have it. 

@Ovle 

sorry, I don't know. Have you tried the Library search? https://impact.github.io/#/

@mathstutor 

I'm not an expert for this topic, but there is a new community forum driven by Maplesoft, which is dedicated to Maple TA and the new learning plattform Möbius. You should try it out and ask your question there: https://mapletacommunity.com

.

You probably cannot do much, the bug does not appear in the current release Maple 2016

dgl:=26*(diff(y(x), x$4))-50*(diff(y(x),x$2))-2*y(x) = 0;dsolve(dgl);odetest(%,dgl);

0

@Vliegkikker 

solve does not contain eq Md, but Mb which is commented out.

.

@Mariusz Iwaniuk 

great, Maple rocks with Mathematica as engine :o)

.

Maybe this is a misunderstanding and they downloaded the free but limited Maple Player from the Maplesoft website? 

.

@Carl Love 

Additionally, to avoid the float evaluation of Pi (if wanted):

restart;
kernelopts(floatPi = false);
W := t*(62.8+0.0945*d/t)/(Pi*d);
eval(W,t = d*TD):expand(%);

@Pascal4QM 

strange, this does not seem to work as equality:

`%=`(`%-`(2*x-5, 2), 3*x-6);

value(%);

the constants are not simplified completely

@Pascal4QM 

not bad. One can even nest operator form to prevent simplification on one side of the unequality, i.e.

`%>`(`%-`(2*x-5, 2), 3*x-6)

Not sure how to evaluate it back to simplified form, though.

@rlopez

I'm wondering why there is a reference to a physics property in that help page. How is the time variable different from any other  independent variable?