Find/Given using programming toolbar - it only works intermittently

[ptc-6685094]

I'm using code off of a project partner where he laid out a find/given function using the programming feature. In most instances it works, however when I change a variable value or add a variable to the equation, it goes "undefined" and will no longer give me an answer. This seems work well when we did the original calculation sheet, however, when I go and try to make revisions to it there are problems that always occur.

Even when I just copy and paste the parts to another worksheet it stops working.

I didn't want to redo his entire worksheet using another method to find the answers. Does anyone know why I would keep having problems with this method? Is there some syntax that is very discrete that I'm missing? If not solving through this method, do you have a recommendation?

Thanks,

Nick

[Werner_E · Accepted solution]

When using units you often run in problems with precision when evaluating a solve block. If you change CTOL higher to something like 0.01 your sheet works.

I wonder why don't you calculate F.V.d directly by multiplying A, N and sigma.

If you have to use a solveblock for a more compicated formula, you may just write F.V.d:=Find(F.V)

[MikeArmstrong]

Werner Exinger wrote:

When using units you often run in problems with precision when evaluating a solve block. If you change CTOL higher to something like 0.01 your sheet works.

You beat me to it

I did forget to mention CTOL though.

[Werner_E]

Mike Armstrong wrote:

Werner Exinger wrote:

When using units you often run in problems with precision when evaluating a solve block. If you change CTOL higher to something like 0.01 your sheet works.

You beat me to it

I did forget to mention CTOL though.

And I missed minerr(). 😉

Interestingly enough minerr() seems not to suffer from the problems Find does when using units.

[MikeArmstrong]

Werner Exinger wrote:

Mike Armstrong wrote:

Werner Exinger wrote:

When using units you often run in problems with precision when evaluating a solve block. If you change CTOL higher to something like 0.01 your sheet works.

You beat me to it

I did forget to mention CTOL though.

And I missed minerr(). 😉

Interestingly enough minerr() seems not to suffer from the problems Find does when using units.

That is strange, even when CTOL is at it's default value.

[ptc-6685094]

Yes, CTOL change worked. Thanks!

The reason why we were starting to use this method was because the formulas get very large. I didn't give a fair representation about the formula complexities. I can't simply manipulate the formulas to get an answer.

I will see if there is a better method we can use to come up with our solutions.

Thanks for the help.

[Werner_E]

Even with more complicated equations it may be worth a try to let the symbolics solve it:

[MikeArmstrong]

Have a look at the attached and see if it helps.