Smallsat

Week 34

Started the project by disassembling the kowa objective to make it space ready. Could not disassemble completely due to lack of a rugged lens spanner in the correct size. Cleaned and degreased the parts we were able to disassemble, including a lens that had residue on it from before. Lens was cleaned with ethanol, this left no residue. The objective was relatively clean from the factory and had no plastic parts inside. The lens might need to be drilled to prevent pressure build up between internal lenses. This poses a problem as its hard to know where inside the objective the lenses are located and how thick the lenses are. The cad only shows outer and inner lenses but not the internal ones.

Started Tokina lens disassembly but quickly got stuck as the phillips screws stripped even with very little force applied to them. Screws in question is on the inner part of the objective, type: Phillips #00

Also saw an existing assembely of the Hypso 2 payload.

I

Week 35

Started the week with a meeting on what was needed for the HYPSO 3 project: Satellite casing and a modular test bench to test different optics.
The test bench needed to accommodate different objectives with different lengths and diameters as well as different angles for testing. The test bench should be 3d printed if possible to make it as cheap as possible and to easily make changes if needed compared to the machined test setups currently used.

Got the idea to use similar system as picatinny rails as that is also made for objectives and to easily swap parts. Early sketches / ideas:

Week 36

HYPSO 3 Payload Meeting
Specs:
4 sub cameras: HSI-Tele, NIR, HSI classic (HYPSO 1 and 2 camera), RGB camera.
The camera angle relative to the light needs to be accurate to 0.1 of a degree.

More brainstorming on how to how the test bench layout should be, either single rail system with fixed angle or double rails with the possibility to adjust the angle on the fly. Both of these have their advantages and disadvantages. The single rail system is less parts, less complex, but requires a specific cassette for a given angle. The big advantage of the more complex double rail system is that you can adjust the angle without needing new parts. A disadvantage of the double rail system is that making sure the angle is accurate enough is relatively hard. However, this can be achieved by printing spacers for the given angles, still less work than making a new cassette for each angle.

Mount to rail interface was also discussed, still based on the picatinny rail system, but without the ridges in the middle. The mounts could either be one piece or two piece, the two piece had the advantage of easier disassembly from the rail itself while the one piece had to slide off the entire rail to be disassembled. The one piece was deemed better in general as we found it to be easier to print well in addition to the fact that there weren't too many times all of the mounts had to be taken off the rail. 
In addition, tolerance tests were printed by different printers to find the smoothest interface between the mounts. 

We also printed the first prototypes to see how the system looked and to decide on the single rail system or the double rail. In addition we also tested a smooth rail profile compared to the standard picatinny rail which has the edges.

The prints above were printed with the goal of finding our solution the fastest, there they do not look particularly good as the biggest possible layer height and nozzle were used. The tape in both prototypes acts as a prototype for a flexible material that acts as a sock over both cassettes when using other angles. This way you would not need a spacer between the cassettes when changing angle, just the spacer on the baseplate itself: Align the rail with the spacer on the left side and tighten down the rearward screw to achieve the angle.
Not pictured: The different tolerance tests between the rail and the mounts.

Week 37

Retried to disassemble the Tokina lens. 3 out of the 4 screws pictured in week 34 are stuck and all 3 are stripped. Got one of the stuck screws loose by folding the tip of a nitrile glove over and putting it between the screw and the drill bit. This did not work on the other 2 stuck screws.

Experimented with a dove tail instead of the picatinny rail profile as we thought that it would be easier than the picatinny / smooth rail profile. We found this profile to be harder to print as we got a sharp edge which non of the printers we tried liked.

We also made another cassette which was split vertically down the middle. This was for the single rail system and worked well. The difference between this and the previous cassettes pictured above is that this one is held together by screws and the older version was a "press fit" solution. Splitting the body of the cassette made for easier removal of the objective as well as it was less sensitive to print inaccuracies (when it comes to fitment of the objective).

Week 38

After a meeting with Joe it was decided that it was better to change the cassette and have them angle specific rather than being able to adjust the angle on the fly. Gratings are very sensitive and need to be fastened securely inside the cassette. The grating holder could be printed with SLA to increase the resolution of the print. Joe deemed the smooth rail profile to be the more practical option. He needed to be able to change the focus on the objectives while keeping the sensor in place. In addition the one piece mounting system was preferred over two pieces.

For the satellite enclosure the RGB camera needs to be off-axis with respect to the hyper spectral camera with approximately 20 degrees (This has not been calculated yet, just an educated guess).

The grating holder design would lock the grating in place with two screws. The grating holder would then be press fit into the cassette with a snug fit. Different offsets were tested to find the correct fit. These were made with fdm printers. To be tested as sla printers to determine best result.

Sketch of grating holder	Cad with features of grating holder assembly	Cross section final cad version	FDM printed grating holder

This week we also made a revision to the rail and mount profile, all the angles are the same, but we changed the dimensions with the goal of making it better to print. The picatinny profile is imperial and therefore the printers need to make corrections to adjust for the "non metric" distances. In general the profile was made slightly bigger in all directions. This was done to the mounts, the rail itself keeps the same shape but with a 0.2 mm offset in the x- and y-direction.

Week 39

Further disassembly of the Tokina lens:

• Managed to loosen outer lens ring, this was done by "cyclic" loading with increasing force to prevent the lens spanner slipping and striking the lens. The lens was glued in place at two spots.
• Outer lens slid right off after. This sub assembly has a round aluminium tube, another lens ring and a mirror (not yet disassembled).
• Rubber ring on outer lens body slid off. Revealed tape (3 pieces) covering 3 screws (phillips #00 head) acting as grub screws to adjust the focus. The slots of these grub screws were packed with silicone grease (assumed). The three screws had polymer sleeves which can easily be replaced by metal sleeves if needed. Another option is to forgo the screws all together as it needs to be glued to the correct focus anyways.
• After unscrewing the 3 screws the outer and inner cylinders came apart.
• The inner cylinder had a ring mirror attached, the lens ring was integrated in the aluminium body attached to the ring mirror. The ring mirror seems to be glued onto the metal.
• Ring mirror was also glued to the threads of the inner cylinder.
• Outer shell part (hat shaped) consist of two aluminium parts that are also glued together. These part house either two or three lenses and is difficult to disassemble.

This week we also finished the "first" prototype and gave it to Joe, it contained, a base plate, mounts and cassettes (10.4 degree) for both Kowa objectives as well as the smaller objectives currently used on HYPSO 1 and 2. In addition a mount for the Tokina outer objective was made. 

Took some photos of the assembly

Brief meeting with Joe to hand over parts and take a look at Tokina disassembly:

• Outer lens assembly and ring mirror assembly might not need to be disassembled-
• Outer shell assembly to be determined, might not need disassembly. Must(?) be drilled to prevent air pockets. Has 2 or 4 lenses potentially. Use smallest drill bit possible.
• Make new rail mount for the axial assembly of the smaller lenses already assembled.

Week 40

Made another cassette for Joe as a different angle was needed (15.7 degree). This was only made for the Kowa lens. The cassette mostly uses the same design aside from the angle, however, after feedback from Joe that the grating holder was hard to remove, we added a screw that would both help push out the grating. In addition this screw also allows us to make the grating holder even smaller and then preload the grating holder to sit snuggly inside the cassette. The new cassette is pictured below.
Notes: Test bench is just long enough, need to make a longer one for Joe, also adjust optical screw holes to be bigger.

Side view	Other side, new screw feature	Front view	Perspective

In addition, we printed the grating holder using SLA, however this print was not successful and needs to be redone:

Some work on trying tools (here generative design) for compliant mechanisms. Idea: Generative design in fusion + eigenfrequency simulation in NX/ Abaqus→ Altered generative design → Real world testing.

HYPSO common meeting:

• Preliminary design review in November
• 5-10 page draft document by the first week of November
• RGB camera angle to be decided → No angle (latest update)
• Generative design (GD) prototype by week 44
• Case: Waterjet aluminium
• Inner mounts: GD PACF
• Eigenfrequency in correct range
• RGB camera might change

Week 41