Hi Saeedeh,

You can place an operand to expose the curvature of your mirror surface, such as SCRV, or SCUR (I let you read the documentation about those Merit Function operands from the Help File). Then, you can use OPGT (operand greater than) to make it greater than zero.

Make sure you measure the curvature at X, and Y positions different than zero if your mirror is centred on the optical axis, as it makes it impossible to measure the curvature there. Also, leave the weight of the curvature operand to zero (so it is not accounted for in the optimization), and only put a weight on the OPGT operand.

Finally, place both operands above the DMFS operand. Otherwise, it'll be overwritten next time you define a new default Merit Function.

I hope this helps.

Take care,

David

Hello David,

Thanks for reply. However, I could not find operand SCRV? But only scur.

I tried scur with 0,100 in x and y coordinate but I did not get any improvement in my curvature.

And my merit function:

Hi Saeedeh,

You are right, I forgot to mention SCRV is for the sequential mode only, I think.

You have correctly put a SCUR operand, but you did not constrain it with an OPGT.

Your SCUR is Operand 4, therefore you should add an OPGT with the column Op# set to 4, with a Target of 0, and a Weight greater than 0.

Does this make sense?

Take care,

David

Hi David,

Thanks for reply. Actually not yet!  I have still this curved minus curvature. 

Hi Saeedeh,

I can see you still have the minus curvature. However, you can now see how the OPGT constraint contributes to your Merit Function. Have a look at % Contrib in your OPGT operand (I believe it is line 5), it says 0.086%. The cost of violating this constraint is almost insignificant with respect to the other constraints you have in your system.

You can try to increase the Weight to 100 for this operand, it should increase its contribution to the Merit Function, and hopefully turn the radius positive. Eventually, you also need to be aware that if you do not have enough degrees of freedom (or variables), the system might not be able to find a solution with a positive curvature.

Let me know how you are getting on.

Take care,

David

Hello David,

Thanks for reply. I increased all weigth for OPGT operand to 100 however it increased the minus curvature even more.

I have 4 wavelengths and I did for all.

Saeedeh

Hi Saeedeh,

At this point, it might be a case where the curvature simply cannot become positive, and you need to add more variable to relax the constraints of your system.

If you want to run a sanity check, just keep operands 4, and 5 (SCUR, and OPGT), and re-run the optimization. This time, I think it should change the curvature.

Let me know if this works.

David