EEVblog Electronics Community Forum
Products => Thermal Imaging => Topic started by: Fraser on November 05, 2022, 02:34:15 pm
-
This is a post for those readers who like to experiment with thermal camera lens designs using generic lenses bought new or harvested from surplus camera lens blocks.
Some time ago I wanted to do some experiments with the Germanium and ZnSe lens elements that I have in my stock. I also wanted to test some complete lens assemblies with unknown specifications. For these tests I wanted a cheap, compact optical bench of reasonable quality. I soon realised that such were not that common or cheap ! After buying a ‘toy’ optical bench kit from China, I dropped the idea for a while. I recently purchased a set of very nice MWIR lenses that were originally intended for use on the AGA 7xx series of cooled scanner type cameras. These lenses have a M44 threaded mount and I already knew that Miranda SLR’s once used an M44 threaded mount of the correct thread pitch :-+ Whilst looking around for some Miranda photographic accessories from which to harvest a female M44 threaded lens mount to use with the AGA lenses I stumbled upon the Miranda Focabell S Macro Bellows adapter. This accessory mounts between the cameras lens and body and permits the user to carry out Macro photography in the same way as using lens tubes, but it gives easy adjustment of lens to Camera body distance :-+ I looked at this contraption and immediately saw potential in it for thermal lens experiments. I might need to use it to determine the back focus distance of my AGA lenses as they were designed for a scanning camera and those can have longer back focus than a FPA based camera lens.
These compact little bellows adapters could be used to experiment with lens elements to determine best lens separation distance or just to see what a combination of two lenses performs like. The mounts in the bellows adapter are pretty hobbyist friendly and adapters or lens holders could easily be made. 3D printed lens holders would also be an option. The front mount that normally takes a lens is an M44 plus Miranda bayonet type combined mount. The rear mount that faces towards the camera is a female Miranda bayonet. The Focabell adapter has 1/4” 20tpi threaded tripod mounting holes on the underside of the support bar, so the whole assembly may be aligned with a camera without actually attaching to it. This thing has a lovely little track along which the front mount travels using a gear system and it all has a good solid feel to it. For me, this is the short ‘optical bench’ that I need for my experiments. It could almost have been made for the task :-+
How much do these cost ? Mine were around £10 and they are quite common on eBay. They date back to the late 1960’s and are of little practical use these days so prices are low. It may be worth buying one, or something similar for your lens experiments ?
The two models I purchased were the Miranda Focabell S and ten Focabell SII. They are both basically the same with tenSiI just being a late revision.
Fraser
-
My Miranda Focabell S and SII units :)
I mounted a pair of the AGA MWIR Germanium lenses on them to show how I will present them to my FLIR SC4000 to determine the expected long back focus distance of these lenses. I can then make an adapter for my SC4000 to permit use of the nice quality AGA lenses.
Fraser
-
As experimenters will know, I need a way to align the experimental lens setup with the cameras FPA. The fact that the Focabell S includes a tripod mount helps in this respect. I bought a batch of BENRO long lens support rigs some time ago as they looked useful and were a good price. I am glad I did as they are perfect for my plans. The BENRO lens support becomes the base that the camera and Focabell adapter mounts on. The BENRO unit offers easy height adjustment of the Focabell adapter or camera, depending upon how I configure it. I can mount the camera on the flat base plate and Focabell S on the adjustable height ARCA SWISS mount , or vice versa. I have shown the setup with the camera on the ARCA SWISS variable height mount and the Focabell S on the base plate but this limits the maximum height of the cameras FPA above the baseplate.
The BENBO lens support incorporates an ARCA SWISS type bevel along its base length and may be mounted on a tripod using a ARCA SWISS type plate holder if desired. The ARCA SWISS adjustable height pole is in two parts to provide maximum height adjustment but one section, and the adjustable foot may be removed if not needed and this reduces the assemblies height.
Fraser
-
So how could this Focabell S adapter be used with modern microbolometer type cameras with their short back focus ? Well that is indeed a challenge ! The back-focus is so short on such cameras that you cannot really mount a lens set outside the cameras casing without using a long back focus lens design. The answer might be to make an adapter tube/cone that mounts on the rear of the bellows and holds the rear lens inside the cameras mount whilst the front lens element is easily moveable using the Focabell S unit. I am likely t be more interested in supplemental lens types so the back-focus issue is less important to me. The Focabell S just mounts in front of the cameras primary lens and projects the scene energy into it. I am also thinking of doing some single lens experiments where the cameras pimary lens is removed and a single lens element is mounted in the front of the Focabell S adapter to enable precise positioning of the lens onto the microbolometer. In photographic experimentation, such an experiment usually results in a soft focus image with chromatic aberration issues.
Fraser
-
Good find.
You might want to look into one of the simple ray tracing programs that are available in the public domain. Once you have determined the basic parameters of your candidate lens elements it will help you look for issues with obscuration and reduction in contrast due to viewing support structure. It will also let you quickly try configurations that might require fabrication of custom mount. Using ray tracing for more than these simple types of objectives takes lots of computing power, expertise and good SW, but there is much low hanging fruit.
-
Yes it is early days of lens experimentation for me :) I have tried ray tracing programs to better understand some of the Inframetrics lenses that I have etc and it has been very interesting. One problem I discovered with common free ray tracing programs is their inability to simulate Germanium lens elements due to the unusually high refractive index. I got around that by halving the refractive index and doubling the distances between lens elements.
Now that I have a suitable 'test bench' for my lens experiments, I hope to enjoy some experiments with both primary and supplementary lens designs :) My primary use for the rig will be the testing of the back focus distance on the AGA lens as I have a nice set that provides 40.5 Degrees HFOV, 20.5 Degrees HFOV, 12.5 Degrees HFOV and 7.5 Degrees HFOV. They FOV spec will likely be different when married to my more modern FLIR SC4000 or Amber Radiance 1 cameras. Hopefully I will just need to make an adapter that mounts on the camera to enable use of the AGA lenses but much depends upon the back focus distance and whether I have enough room for an adapter between lens and camera mount. If not, I will come up with an alternative approach :) MWIR lenses are hard to find so I was thrilled to get these very nice AGA Germanium units. They are in 'as new' condition so perfect for my needs. I have two of the 40.5 Degree lenses so I may make a reversed lens thermal microscope with one of them.
Lenses are an area I need to learn more about and I have good reference books on IR/Thermal lens designs but practical experimentation if good fun. I may share my experiences on this forum but it is likely to be a bit of a specialist topic, of interest to only a few members. We shall see.
Fraser
-
My first job for the Miranda Focabell S adapter will be testing these AGA lenses that were designed to illuminate an AGA 700 series scanning thermal camera. The camera uses a pair of spinning Germanium prisms and I have no idea what the back focus distance of the lenses will be. From my past experience, it could be quite a long distance.