You would have to post the file for us to examine.
Which Mathcad version and which maintainance release?

It would be better to trace the bug, if it is a bug anyway, down to just a few lines of code and not sending a huge worksheet. I guess only very few people would be willing to wade through a huge sheet they are not acquainted with to trace down what could cause that effect.

Again - which version and maintenance release of Mathcad are you using? Current version is Mathcad 15 M030. (Prime being so deficient to that very high degree up to uselessness can't be considered a Mathcad version and guessing from your screenshot you are not using Prime).

I assume that autocalc is turned on (Tools/Calculate/Automatic Calculation)? If not, thats the problem.
You can recalc a single expression with F9 as opposed to CTRL-F9 for the whole sheet.
It may have to do with the inline evaluation. What happens if you do assignment and evaluation in 2 separate regions? Does the problem persist?
I also noticed that in the left region of your screenshot there is a mutliplication dot between the number and the unit while in the right one there is none - not sure at all if that has any meaning, though.

In this forum we cannot edit posts we already sent, right? Or am I missing something?

Anyway, the reason I was asking for the exact version you use is a bug which was not fixed until Mathcad 14 M030 (the last but one version of MC14 and the last we can recommend using as M035 had a severe bug).
From the release notes for MC14-M030:
"SPR 1437427 Fixes in-line evaluation to update display to match available value. "

Hi rmix
Its seems that I have the "EVIL VERSION" wich is known for this bug.
V14 MO20

Yes, it looks like its this bug which is already fixed. So you either have to upgrade or place assignment and display/eval in two separate regions to avoid the display of wrong/old results until the next recalc.

BTW, MC14 M020 wasn't that "evil", but the 14ers before are, suffering from real bad, severe bugs. That was the reason PTC decided at those times to give M020 to all customers for free, even those without maintainance.

Is V.15 free of this problem?

Quote (mar2805)

Is V.15 free of this problem?

As far as I know the bug was not reintroduced after it was fixed in MC14 M030
Mathcad 15 F000 was originally planned to be Mathcad 14 M040, but then PTC marketing took over. So there is not that big change under the hood.
Find attached the release notes of the current MC15 M030, listing all bugfixes since MC14 M035. The bug in question is not listed as it was already fixed in MC14-M030. Can't find the release notes of MC14 M035 but you can google for them (as I did before) - its not a pdf but a couple of html pages. OK, found it at PTC here: http://www.ptc.com/WCMS/files/98347/en/Mathcad_14_M035_RELNOTES.HTM

As I can't edit my previous post, here is the attachment I forgot

• http://files.engineering.com/getfile.aspx?folder=21cd2164-b20f-4ae2-9808-bc

M14 was a pretty good state of recovery from the debacle that was M12.

TTFN
FAQ731-376: Eng-Tips.com Forum Policies

Need help writing a question or understanding a reply? forum1529: Translation Assistance for Engineers

Yes, I agree. MC12 was the worst version ever (OK, Prime. But Prime doesn't count as Mathcad for me
Many consider MC11 the best version ever. A good reason for this is Maple as symbolic engine instead of muPad.