cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Try the Materials Informatics Toolkit, which is designed to easily handle SMILES data. This and other helpful add-ins are available in the JMP® Marketplace
Choose Language Hide Translation Bar
Mytom
Level II

Mixture sum

Hi. 

 

I tried to make a custom designed mixture experiment with only two different ingredients with varying procents, all other ingredients is helt constant. I tried to use the "mixture sum" option to sum up the two ingredient at the maximum range but it does not work. The last ingredient gets changed into the value stated in mixture sum but that is not what I am after. Any suggestions on what could work? 

 

Example: 

 

Ingredient A have ranges 0.5 to 30%, B has 0.5 to 13%, Mixture sum is therefore 43%, but non of the ingredients should alone have up to 43%. 

2 REPLIES 2

Re: Mixture sum

I think more details are needed.

 

From what you stated you are only wanting to change two components, A and B. A and B should add up to 43%. However, the only way for you to achieve that is if A is at its maximum and B is at its maximum. So there is only one trial: A=30% and B=13%. If you want less of component A, B MUST go higher than 13%.

 

Finally, note that with only two components changing, you really could just make a non-mixture design with just one of the components. The other just varies to reach the total.

Dan Obermiller
Victor_G
Super User

Re: Mixture sum

Hi @Mytom,

 

As stated by @Dan_Obermiller, we are missing some informations to help you further and there might be a problem in your constraints.
Please note that if the total of your 2 ingredients is fixed in your formulation, you can also use one numeric factor in your design that will correspond to the ratio of your 2 ingredients (A/B), and specify high and low levels.

 

I hope this answer will help you,

Victor GUILLER

"It is not unusual for a well-designed experiment to analyze itself" (Box, Hunter and Hunter)