ACCEPTED SOLUTION
Accepted Solutions
Your programs don't show what you intend to do if the function argument x is not exactly in the list of nominal diameters.
The programs (with Terry's fix applied) will return just the last value of j in this case (15).
If, as I assume, that's not intended, you could simply place the return value you want to see in case of an invalid x in the last program line, like:
To achieve what you obviously intend to do (take the nominal diameter ND, lookup the corresponding outer diameter OD and use it to calculate the cross-sectional area) could more easily be done using the "lookup" function:
To avoid the error in case x is not in the list you could add your own return value in case lookup fails:
I think that returning the index j as well was done just for debugging. But in case you really need that index you could use the "match" function (see attached sheet).
You may also just use the next larger value in ND in case x is not found in that list
Similarly you could also decide to use the next smaller value or the nearest value (see attached sheet).
And of course a further option would be to use linear interpolation to deal with x-values not in the ND-list.
Prime 11 file attached
Hi,
The comparison equals is a bold equals sign and is entered by the following keys together [CTRL][equal sign].
Cheers
Terry
Your programs don't show what you intend to do if the function argument x is not exactly in the list of nominal diameters.
The programs (with Terry's fix applied) will return just the last value of j in this case (15).
If, as I assume, that's not intended, you could simply place the return value you want to see in case of an invalid x in the last program line, like:
To achieve what you obviously intend to do (take the nominal diameter ND, lookup the corresponding outer diameter OD and use it to calculate the cross-sectional area) could more easily be done using the "lookup" function:
To avoid the error in case x is not in the list you could add your own return value in case lookup fails:
I think that returning the index j as well was done just for debugging. But in case you really need that index you could use the "match" function (see attached sheet).
You may also just use the next larger value in ND in case x is not found in that list
Similarly you could also decide to use the next smaller value or the nearest value (see attached sheet).
And of course a further option would be to use linear interpolation to deal with x-values not in the ND-list.
Prime 11 file attached
Are you aware that a 'return' statement stops the execution of a program?
So your program stops as soon as 'Detector' gets equal or larger than QSP.1. This is already the case right the first time and so your program already stops at i=0.
BTW, it's not good programming style to alter the loop variable i of a for-loop. You should avoid the i <- i+1 assignment.
Here you see in variable YY all the results from i=1 up to i=6. Detector is larger in all cases:
It may not have to do with the problem you encounter but I think you made a mistake when you defined function Pvh.
Your argument rho could be omitted as its not used in the function definition. Maybe rhow was meant to be just rho?
As Terry already explained is the problem that Prime does not allow to add a unit to a text string.
As an alternative to the text output you may use an 'impossible' return value (such as 0 in or a negative length), or you may return a NaN (not-a-number) with appropriate unit or you may create your own error-message:
Well, you kind of can add units to a text string via the format function.
(Mathcad Express 11, hence the convoluted expressions. Prior workings hidden to avoid hysterical laughter at my misunderstanding of what's going on)
Stuart
@StuartBruff wrote:
Well, you kind of can add units to a text string via the format function.
Hmm, I would say that's rather adding units IN a string, not TO a string.
I wonder if it would be possible to write a function "unit2str" ?
Of course I would strive for a method that does not consist of treating all possible units individually, like
@Werner_E wrote:
@StuartBruff wrote:
Well, you kind of can add units to a text string via the format function.
Hmm, I would say that's rather adding units IN a string, not TO a string.
I wonder if it would be possible to write a function "unit2str" ?
Of course I would strive for a method that does not consist of treating all possible units individually, like
WE: "I wonder if it would be possible to write a function "unit2str" ?"
No. Not at all. Impossible, I tell you, absolut unmöglich!!
- I've needed to convert strings to symbols and vice versa on many occasions, but never as much as I needed it for this task! Yes, of course it's a long-standing feature request
- When assigning values to multiple variables, which happen to be contained in an array, it is a real pain not being able to select them all and change their Math Styles in bulk. Given my very low attention span, setting multiple instances of 24 variable names in a vector to Unit style is not my favourite way to pass a few idle minutes.
- I wish PTC would fix the Text Block so that it only covers the left-hand page, rather than extending off to infinity on the Draft pages. Perhaps there is some use case that many people have for it, but I'm afraid I can't see what it might be. IIRC, Mathsoft Mathcad also had the ability to toggle between a Text Block and a Text Box -- very useful.
- I'm bored now and in need of coffee, plus I have no work ethic, so I'll maybe complete a few more completely factored derived units and tidy up the worksheet before posting it. The conversion is largely powered by brute force and ignorance, so expect no startling insights.
Stuart
(Mathcad Express 11 displays numbers less than 10-15 as 0. Changing the zero threshold is a Premium feature)
Thank you, but, no, just the one cup. It required less tidying up than I thought,
Although I did accidentally delete some work I'd done on the reverse transformation str2unit.
I'm half sure I did something similar a long time ago, but I couldn't find the worksheet, so I thought it was quicker to recreate it from scratch. I've added a few more derived units to the worksheet and created a very inelegant str2unit.
I might brood on str2unit for a little bit longer, as there is at least one bug in it that I need to fix, and a few more derived units to add. Or, apart from the bug fix, I might not, as I have a shedload of other worksheets that I'm working on Expressifying.
You might notice one significant omission from the vector of frequencies, and therein lies my bug.
Stuart
After dealing with a number of non-Mathcad related tasks, I got around to sorting out the problem. I now have a lot of tidying up to do, but won't be able to get around to for at least a couple of days due to other stuff getting in the way.
OOIC, does anybody know of a way to break out of a fold?
Stuart
