[Webinar] Micro-Objective Design for Robotic Surgery [Q&A]
This thread will be used to collect questions before the webinar, and to answer any questions we received during the webinar. Feel free to post your questions!
Be sure to subscribe to this thread if you want to see additional discussion regarding this topic. The thread will be open to comments until Thursday, March 17th.
Webinar details:
Register here: tThe event has concluded]
Date: Thursday, March 10th
Time: 6:00am PST & 11:00am PST
Presenter: Bob Householder, Senior Product Sales Manager
Abstract:
Micro-optics for medical devices carry a higher burden than commercial or industrial applications where sterilization, compliance and materials play a greater role. In this talk, a generic set of opto-mechanical requirements will be studied to show the process of optical form and design through opto-mechanics, tolerance analysis and production considerations. This practical approach will reveal the top lens design forms and build on the initial design to construct an opto-mechanical sub-assembly for the application.
Page 1 / 1
Is it possible to limit a substitute solve to only a few select materials? Perhaps, for example, I only want to consider plastics for material substitution. Thank you.
Update: Thank you for considering my question, I did figure something out. Perhaps an expert user can suggest a better way. I copied the misc.agf material catalog to the new name “Plastics”. I deleted all but the four entries that I recognize as plastics. Then I opened up the Sabic_Thermo_Optical_Polymers.agf catalog and copied its four entries, one at a time, and pasted them into my new Plastics catalog. Now, I have a plastics catalog with 8 items. As I come across other plastics in the other catalogs, I will add them to this one. Perhaps there are some suggestions for other catalogs with plastic materials.
Now, is there a way to limit the substitute solve to one catalog?
There were some questions during the webinar which Bob will be addressing on this thread over the next couple of days. The thread will be opened for a limited time to allow for discussion and additional questions. Click the Subscribe button on the original post to be notified of new replies!
Is it possible to limit a substitute solve to only a few select materials? Perhaps, for example, I only want to consider plastics for material substitution. Thank you.
Update: Thank you for considering my question, I did figure something out. Perhaps an expert user can suggest a better way. I copied the misc.agf material catalog to the new name “Plastics”. I deleted all but the four entries that I recognize as plastics. Then I opened up the Sabic_Thermo_Optical_Polymers.agf catalog and copied its four entries, one at a time, and pasted them into my new Plastics catalog. Now, I have a plastics catalog with 8 items. As I come across other plastics in the other catalogs, I will add them to this one. Perhaps there are some suggestions for other catalogs with plastic materials.
Now, is there a way to limit the substitute solve to one catalog?
Thanks.
Hi @Stephen.Remillard. Thank you for attending the webinar and for your question. There is a material ‘solve’ called Substitute. You can enter your preferred or custom material catalog here and the optimizer will only target the materials located in this list. In your case you would enter Sabic_Thermo_Optical_Polymers.agf . Please let us know if that works for you.
-Bob
What type of manufacturing techniques are available for these small lenses? from @Anthony.Ang
Hi Anthony, if you are using glass materials, then the traditional grind / polish using a ‘single stick’ is the method to fabricate. There are other techniques such as compression molding that are useful for creating the surfaces, including freeforms.
-Bob
What should be the typical wavelength range for lateral color and axial color correction with respect to central wavelength? from @Abdullah Saif.Mondol
Hi Abdullah, I usually designed from F,d,C at least to start. Sometimes there was a requirement to ‘check’ performance at 850nm for an emission wavelength for a signal sensor. For systems that need near IR imaging then a balance must be made between visible imaging performance and near IR. Given the narrow spectral range of lasers or LEDs in that near IR range, the design can be pushed to optimize for both. The retro-focus design style lends itself to additional color correction due to the compressed image and longer length.
-Bob
Is that within only one glass catalog? / or does that substitution template give all glasses across all catalogs that fit within the specified categories? from @Anthony.Cocola
Hi Anthony, the substitution shown in the webinar was only for ‘preferred’ glasses for the catalogs listed. I did only have Schott selected. You can add more catalogues such as Ohara and the substitution will include those material tagged as ‘preferred’. If you want to downselect to just a few materials, you can create your own catalog and then use the material solved called ‘substitute’ to only use these materials.
-Bob
In your experience, what is the benefit of using Contrast Optimization vs MTFA operands? from @James.Sheil
Hi James, the MTFA or MTF (tangential and sagittal) operands in general target the overall MTF performance, ether average or each axis. This calculation is computationally intensive. Also, in the early phases of the design, the results may not converge on a design form of interest or within the bounds of manufacturability.
Contrast Optimization, in comparison, will compute the phase difference at the exit pupil up to the spatial frequency limit specified. This method is much faster to compute and may be used earlier in the design cycle. A link to the knowledge base article on this method is below. As mentioned in the webinar, a good technique is to start with something like RMS Spot Radius and then move to Contrast Optimization as the design progresses.
Could you please explain again in details how we should use th System PreDesigner? from @Eirini.Kakkava
Hello Eirini, the use of the Pre-Designer application or an excel worksheet with similar functionality is helpful to understand where your system lies on a graph of FOV vs f/#. With this general knowledge, you can select systems that operate in this region prior to simply starting with a Cooke Triplet or Acromat, for instance. After practice and then practical design cycles for a certain application, the knowledge of performance is known and then this first step is not needed.
-Bobh
What kind of light source is normally used in this small system? from @Limei.Wang
Hello Limei, for these small objective lenses, micro-LEDs are generally used. There are a wide range of LEDs available now that can be mounted to a very small mylar or even aluminum substrate with higher color rendering for surgical applications. For other uses in military or police, you can even design for near IR only enabling a ‘night vision’ mode for stealth viewing prior to room entry.
-Bob
Could the doublet lens be removed from the design in the landscape lens if we are designing a monochromatic system (NIR), i.e. does the doublet have another function besides color correction ? from @makro
Hello Marko, in short yes. The doublet is mainly for color correction. Sometimes, however, a doublet may be used for narrow band system to make it more compact. In the case of NIR, I would eliminate the doublet.
-Bob
Do you assume in these surgeries that the working distance is sufficiently large such that you can assume an object at infiinty? (i.e. 10x the distance of the focal length) from @limanatzem
Hello Natzem, starting with an infinite conjugate is certainly fine. I always enter the finite conjugate as the design progresses to have the exact system. I would not keep the system at infinity through the full design cycle.
-Bob
Could you comment on your train of thoughts when you make an initial choice for a glass? Also, can you give us a feeling for what a MTF at 46% means in terms of image quality in that context? What does this tell us about the ability to image a particular tissue for example? from @David.Nguyen
Hello David. I usually start with BK7. I ‘grew up’ on Schott glass so I have that in my mind. I then start to increase index and Abbe for individual lenses based on what control is needed.
MTF at 10% is just perceivable for that spatial frequency. 46% will give you sufficient visibility and performance to allow for manufacturing tolerances. One way to assess fitness to measure the minimum size on the tissue you want to image and then calculate what spatial frequency that equates to.
-Bob
Is there a link to the bubble chart showing different lens types vs field and f-number? / Could you explain again the advantage of using "contrast" optimization? from @Reid.Greenberg
Hi Reid, I have linked below an SPIE article that runs through this pre-designer including the bubble chart.
Contrast Optimization calculates a phase difference in the exit pupil up to the spatial limit defined in the Merit Function Wizard. Straight MTF targeting in the Merit Function is computationally intensive and does not produce good systems (generally) during the early phases of a design. Here is a link to the knowledge base article with more details.
What are the weights of each field on the initial setting? from @Ruiqi.Long
Hi Ruiqi, I started with equal weighting of 1 for each of the 3 field angles. I settled on the following weights before stopping optimization.
0deg = 1
28deg = 0.5
40deg = 2
-Bob
How long does it take to develop this lens from scratch? from @Sergey.Nikitin
Hello Sergey, I have 20+ years of lens design experience and about 7 years working on these micro-objectives. I designed this one from scratch in about a week for this webinar. There is work remaining on this design to complete: (1) tolerance study including mechanics, (2) straylight analysis, (3) design for manufacturability with lens vendors and (4) opto-mechanical design.
-Bob
Will coating of the back surface of doublet with IR filter results in strong back reflection into the doublet? from @su.liu
Hello Su. Yes, back reflection is of primary concern using this approach. This will be the first priority to investigate. It may be possible to introduce a small curvature to avoid back reflection and still be suitable for an IR filter. Great insight!
-Bob
Do you use global optimization to explore a better design? from @Shaohong.Wang
Hello Shaohong. Yes, I do use Global Optimization. You may also consider using High Yield Optimization, which helps to reduce the incident angle on surfaces to make the design less sensitive. I did not take the time to explore this step for the webinar.
Any suggestions on building up a preferred glass/plastic materials catalog to use for glass substitution optimization? (Existing databases or vendor dependent) from William Hurley
Hello William. I take the preferred glasses from each vendor as my starting point and then you can save this using the Glass Substitution as a new Material Catalogue. Please see the included screenshot.
-Bob
How to make sure of the lenses center alignment during manufacturing and assembly? from @Yi.Zhang_01
Hello Yi. When working with glass lenses and aluminum housings and spacers, you can set the diameter tolerance to be very tight with a minimum radial space of 0.03mm. By holding the diameter tight, you can loosen the centration tolerances to something more manufacturable. You can study this during your tolerance analysis. The save tolerance data feature in OpticStudio is very powerful. You can also use tolerance scripting to include figures of merit such as MTF to be included in your saved tolerance data. Please see the article via the link below on how to include any Merit Function operand in your saved tolerance data.
