PTC Mathcad - Symbolic Solver Demo

[@rebel_jedi]

Thank you !

[@PatJHeffernan]

I hope it is helpful. Good luck.

[@RoyWickrama]

Thanks for the video. Today I thought of using symbolic solver for a equation of which the variables and constants have units. Hope you could shed some light on this or provide me with some comets.

[@PatJHeffernan]

@@RoyWickrama An interesting thought. I’ve never tried to integrate. Obviously every variable has units associated with them, and for the equation to work, the units need to be compatible… but I’ve never used units with the Symbolic Solver to further manipulate. Clearly, as you simplify or manipulate, the units of whatever you derive can be determined by whatever variable are left.

[@sergeydukman5832]

Very usefull presentation. Thank you.

[@fluna90]

Great video! Thank you very much!

[@johnschindler9641]

This is a start. I have a problem similar to Omar's. I have the equation I = C * exp(-(ms*S+(1-S)*mmx)*d*t) (Beer's law of a binary mixture) I had to type this, copy/paste from Mathcad leaves a horrid mess, but this is basically correct. I have values for all the variables, and I calculates correctly. I want to invert it by solving for S. I've tried :=, =, CNTRL =, explicit, and everything fails saying that S is a "bad solve variable". I have tried making it conform to the assumption that the Depression equals zero by moving the I to the other side of the expression. It can be done on an HP Prime calculator and the result checks out numerically. I'm hoping to get something I can paste into another document.

[@johnschindler9641]

Well, shoot. Experimenting a bit, it seems explicit doesn't really help, even in very simple cases. I have to keep a copy of the equation above the variable settings and another below them for numerical evalutions with different variable values.

[@121-f3v]

It would be great to attach the Mathcad file to this video.

[@PatJHeffernan]

I wish I could, but I don’t believe there is a way built into KZbin to do so. That said, it would be easy enough to recreate based on the video. Cheers,

[@121-f3v]

@@PatJHeffernan you can attach a link to google drive ))

[@PatJHeffernan]

@@121-f3v Hmm… I will look into that. I don’t know much about Google Drive, but i can look into and use it when desired.

[@PatJHeffernan]

Here is a link to the file on my OneDrive... see if that works for you. The real value is in "creating" the worksheet. 1drv.ms/u/s!AkKQQeo9NFT_gdEemlhOmMvZ0zG2_A?e=YMdmS6

[@121-f3v]

@@PatJHeffernan i downloaded it, thanks a lot

[@ahbushnell1]

Can I save the results of the symbolic processor for future calculations? I know I could copy the results and paste but it won't automatically update.

[@PatJHeffernan]

Andy, do you have an example of what you are trying to do? I can play with it to see what is possible. It’s not a thing I’ve needed before so I’d have to look into it.

[@ahbushnell1]

@@PatJHeffernan Thanks for getting back to me. I will do circuit analysis. I will generate a complex matrix say 3x3 or 4x4. The matrix has comlex s values for the frequency domain. I will invert the matrix (with symbols) and multiply by a voltage vector. This gives me the analytical answer. the way I do this now is to copy the answer and paste to the next line. If I change the input I have to redo the copy and paste. Thanks Andy

[@PatJHeffernan]

One way I've been able to save results from the symbolic solver for numeric use after is to use functions. For example, if you have a polynomial and assign it to g(x), you can use the symbolic solver to assigns its integral to a new function f(x). You can now use f(x) for normal calculations. Not sure if this is useful to you. Good luck.

[@ahbushnell1]

@@PatJHeffernan That won't work. I'm having Mathcad to algebra. I produce a set of equations in the Laplace domain (i.e. functons of s). Then I do a matrix inversion thats symbolic and multiply a vector to get a vector this algebraic. I have to copy the results to a function of s to do the calculations. If I change the matrix I have to repeat the process. Thanks for the thought.

[@Omarfa25_]

In case i have to solve symbolicly an equation that inlcudes a 0.85 number for example, i want it to mathcad keeps it explicit, but when i use explicit, it reduce it in to unwanted number of decimals. How can i solve this

[@PatJHeffernan]

Hi Omar. I’d be happy to look at your problem, but I don’t understand exactly what you’re trying to do. Do you have a way to share an image of your problem?

[@Omarfa25_]

@@PatJHeffernan Hello, i think i am unable to share an image tru this way. But what im tryn to do is solve an equation like this As*fy=0.85*B*c+A'*f'y Where the only guess is c, i want it to solve it symbolicly for c, but the program doenst leave the 0.85 explicit like this 1/0.85, instead it reduce it in to unwanted number of decimals, like 1.1764705882352941176

[@PatJHeffernan]

Love concrete... so you're using the symbolic solver to solve for c (depth of NA) but you would prefer to keep the 0.85 isolated and not integrated. Try rewriting your equation using a variable for the 0.85. Make it a function of that variable. Using the symbolic solver to solve for "c" should give you a solution as a function of the undefined variable. Now just substitute 0.85 into your function and you will get the plain answer. Control the number of decimals displayed on the math formatting tab (kzbin.info/www/bejne/iX3Fp2CPhbKFg7c). Hope this helps.

[@Omarfa25_]

@@PatJHeffernan yeah thats exactly i want to do, keep the 0.85 isolated. Actually i did that you just said, but was wondering if there was other way to do it. Thanks anyways for you help, Grettings from Panama