The Origin Forum
File Exchange
Try Origin for Free
The Origin Forum
Home | Profile | Register | Active Topics | Members | Search | FAQ | Send File to Tech support
 All Forums
 Origin Forum
 Origin Forum
 Ternary diagram smoothing with lnx instead of TPS?

Note: You must be registered in order to post a reply.
To register, click here. Registration is FREE!

Screensize:
UserName:
Password:
Anti-Spam Code:
Format Mode:
Format: BoldItalicizedUnderlineStrikethrough Align LeftCenteredAlign Right Horizontal Rule Insert HyperlinkUpload FileInsert Image Insert CodeInsert QuoteInsert List
   
Message:

* HTML is OFF
* Forum Code is ON
Smilies
Smile [:)] Big Smile [:D] Cool [8D] Blush [:I]
Tongue [:P] Evil [):] Wink [;)] Clown [:o)]
Black Eye [B)] Eight Ball [8] Frown [:(] Shy [8)]
Shocked [:0] Angry [:(!] Dead [xx(] Sleepy [|)]
Kisses [:X] Approve [^] Disapprove [V] Question [?]

 
Check here to subscribe to this topic.
   

T O P I C    R E V I E W
VGL Posted - 01/12/2021 : 2:17:24 PM
Hi all!

I'm currently using OriginPro in order to create ternary phase diagrams. Since I only have a limited number of data points, I'm using the smoothing function for interpolation. Origin offers a smoothing option for ternary diagrams which uses Thin Plate Spline (https://www.originlab.com/doc/Origin-Help/PD-Dialog-ContourInfo-Tab#Smoothing). However, since I'm working with phase transitions I'd like Origin to interpolate using an ln-function or something that better represents eutectic points (basically sharp minima and broad maxima).

Maybe this picture helps to understand what I'm trying to do:


Is there an easy way to accomplish this within the ternary diagram plot?

OriginPro 2020b, Win10
9   L A T E S T    R E P L I E S    (Newest First)
VGL Posted - 01/18/2021 : 10:38:59 AM
quote:
Originally posted by YimingChen

Hi,

We discussed here and we feel it is still difficult to define the negative log function. To define each negative log function for each region, boundary has to be predetermined. We couldn't find anything useful related to this on the internet.

James



Then I will keep working with the given smoothing method. Thanks anyway!
YimingChen Posted - 01/18/2021 : 09:59:06 AM
Hi,

We discussed here and we feel it is still difficult to define the negative log function. To define each negative log function for each region, boundary has to be predetermined. We couldn't find anything useful related to this on the internet.

James
VGL Posted - 01/15/2021 : 12:33:20 PM
quote:
Originally posted by YimingChen

Hi,

In the ternary phase diagram in the link, the contour was divided by three boundary curves. Do you know how to determine these boundary curves? Once that was done, we can smooth each region using the current method respectively.

Yiming



Hi!

The boundary lines are a result of the "domes" meeting each other. The intersection of these shapes defines the boundaries. The "domes" can probably be described by a negative log-function starting at the maximas of the 3D field that is calculated by TPS. A customizable parameter for the steepness of the log-function would be amazing.
Thanks!
YimingChen Posted - 01/15/2021 : 11:01:26 AM
Hi,

In the ternary phase diagram in the link, the contour was divided by three boundary curves. Do you know how to determine these boundary curves? Once that was done, we can smooth each region using the current method respectively.

Yiming
YimingChen Posted - 01/13/2021 : 12:05:53 PM
Thank you. This is helpful.

James
VGL Posted - 01/13/2021 : 11:57:29 AM
Hi James,

Thanks for the heads-up and for creating a ticket regarding this.
I don't have a corresponding paper at hand, but I have some 3D-visualization of what I'm looking for.
Here is a 3D-plot of a ternary phase diagram for melting temperatures:
https://docplayer.org/docs-images/43/3662271/images/page_19.jpg

As you can see, the peaks are broad while the minima are discontinuous (see the image in my first post).

EDIT: Here is a paper featuring the kind of phase diagrams I'm trying to produce in origin. Note how the minima converge into lines (labeled ternary eutectic) in Figure 2 instead of building an area: https://www.tulane.edu/~sanelson/eens212/ternaryphdiag.htm
YimingChen Posted - 01/13/2021 : 11:42:10 AM
Hi,

Currently Origin doesn't support such smoothing/interpolation method. I have reported to our developer see if we can improve. You can use the jira ID (ORG-23087) to check the update with us. By the way, do you have any reference article/paper of doing such or any software/package that support this type of interpolation?

Thanks
James
VGL Posted - 01/13/2021 : 09:44:42 AM
Hi James,

Thanks for your help. You are probably right about it being no smoothing problem (at least not exclusively). I'm trying to change the way Origin interpolates my data. I think I need to clarify what my graph looks like:


It's a ternary diagram of the XYZ(Z2) type. Z2 is my height data. Circles labeled 1-10 are my data points. The height profile in-between is being interpolated by Origin. Now what I need to adjust is the way the mountain profile falls off between my data points. Currently it's looking like an n-polynom, I would prefer a log-function with broad maxima and sharp minima. Is there a way to adjust the function Origin uses for interpolating the data.
However, the smoothing option adds fictive data points all over the grid and I assume their distribution is following the Thin Plate Spline function.

Any ideas?
YimingChen Posted - 01/13/2021 : 09:26:51 AM
Hi,

It looks not a "smoothing" problem. If you just want to connect two points with a curved line, you can try the curve tool which can plot line with adjustable curvature.


James

The Origin Forum © 2020 Originlab Corporation Go To Top Of Page
Snitz Forums 2000