Problem with odesolve in Mathcad Prime 1.0 - Need Help

[MBuck2]

The attached MC Prime 1.0 worksheet uses a method, I implemented in MathCad 15.0. It worked fine in MC 15 but for some reason I cannot get the MathCad prime version working. I have tried changing initial conditions and the function slightly but nothing seems to work. Conceptually what this ode is designed for is to assess the temperature of a beam exposed to elevated temperature - in a fire for example. The cs function is the change of the specific heat of steel with temperature which no doubt is some type of curve fit to experimental data. I know the function looks strange but I have checked it against my MathCad 15 implementation (which works) and it is identical. Can someone show me what I am doing incorrectly MC Prime seems to work somewhat differently than MC 15. Thanking you in advance.

Regards, Mark Buckton

[MikeArmstrong]

For some reason your worksheet will not open in my Mathcad Prime. Can you try and reduce the size of the worksheet by just posting the solve block problem?

Mike

[MBuck2]

Mike;

Thanks for looking at this for me. I have deleted the plots of some of the functions now the sheet is only 32kB.

Regards, Mark

[ValeryOchkov]

It will be good to use units by ODE solving.

See one example:

[MBuck2]

Valery,

Thanks for this it is certainly my intention to put units into the calculation but I would like to get it working first. What I cannot understand is why it worked in MC15 and now does not work in Prime 1.0?

Prime's syntax seems different but I maybe it is just that I have made a stupid mistake but even if I simpify the cs(t) function by making it a constant it still dosen't work?

Regards, Mark

[AlanStevens]

You had Tf as a function of t, and didn't have the Ts that is an argument to cs as a function of t. Also, I had to rewrite the solve block before it would accept the changes! See attached.

Alan

[MBuck2]

Alan;

Thank you for this but I not sure I understand if what you have done maps to reality? I agree that Cs is a function of Ts i.e. Cs changes with the steel temperature but Tf (aka the fire temperature) is also a function of t i.e. time. For example heat transfer from a real fire as in this case or a furnace. I looked at my MC 15 worksheet (see image file attached) and this is how I did it and it appears to work. In other words, in my view the solve block / differential equation should take the change in fire temperature into account i.e. conceptually the heat from the fire is transfering heat to the steel beam ii.e. both the fire and the steel are increasing with temperature as a function of time i.e. they increase in temperature as time goes on but the steel temperature lags the fire or furnace temperature - this being the real word situation. Is that being considered with your reformulation?

Regards, Mark

[AlanStevens]

Mark Buckton wrote:

Alan;

Thank you for this but I not sure I understand if what you have done maps to reality? I agree that Cs is a function of Ts i.e. Cs changes with the steel temperature but Tf (aka the fire temperature) is also a function of t i.e. time. For example heat transfer from a real fire as in this case or a furnace. I looked at my MC 15 worksheet (see image file attached) and this is how I did it and it appears to work. In other words, in my view the solve block / differential equation should take the change in fire temperature into account i.e. conceptually the heat from the fire is transfering heat to the steel beam ii.e. both the fire and the steel are increasing with temperature as a function of time i.e. they increase in temperature as time goes on but the steel temperature lags the fire or furnace temperature - this being the real word situation. Is that being considered with your reformulation?

Regards, Mark

Sorry, didn't read the whole file properly, just went on the fact that you'd assigned a single value to Tf in the solve block. Will have another look.

Alan

[AlanStevens]

Just removing the redefinition of Tf as a single value, and making Tf a function of t inside the solve block does the trick - see attached.

Alan

[ValeryOchkov]

Mark Buckton wrote:

Valery,

...but I maybe it is just that I have made a stupid mistake...

May be it is a "stupid Mathcad Prime"

One example!

You can change in my task about car moving the ODE from ...+...=N on N=...+... and get... an error message