JWST

[slim_boy_fat]

Watching with interest , how large overall is this going to end up?

 

[Screwdriver]

Another fail halfway through a 2 1/2 hour print...

It is quite tricky to get a sliding fit on complex parts and after swapping for a 0.3mm nozzle, print quality changes, accuracy changes and clearances change. But I have homed in on these settings for ages, they should be spot on! There are problem areas with the model, the inner wall overlaps the outer for a start, hence the swap to a 0.3mm nozzle for these super thin tubes. I upped the temperature slightly to account for the increased pressure but it was under extruding before finally failing.

Anyhow, cut a long story short, I whipped out the brand new nozzle and stuck it under the microscope...

Yep. That is brass! Either a lump of swarf got in there or some process had failed, smearing the inside of the feed tube with a coating of brass. I hacked it out with a 1.7mm drill (sadly I don't have a 1.75) so I could at least have another go while I wait for a new set to turn up.

It would print - perfectly, then for no apparent reason just fail and disguise the failure mode in a birds nest of tangle and a hot end embedded in plastic. Hey ho. I'll be checking ALL my nozzles under the microscope from now on...

Oh @slim_boy_fat, with fwd+aft sunshield pallets extended, the model is over half a meter, tip to tail (515mm). The side boom arms extend the model from 11.5cm closed to 37.3cm when extended. Each telescopic boom goes from 4.8 to 17.6cm (!). The DTA tower takes up a chunk of the internal space since it also has to deploy.

Naturally, the primary mirror is the most eye catching feature (well it will be when it gets gold coated) and is over 17cm "diameter" when unfurled...

 

[hotponyshoes]

[QUOTE="Screwdriver, post: 1990655,
This next hinge part took ages to concoct. The JWST has angled rectangular section support arms and I wanted it to look right. Sketchup hates odd angles and the odd axes makes it super complex because the "push/pull" tool basically heads off on the axis which may or may not be along the body of the part if the same "group" has odd angles in it.
[/QUOTE]

You might well already know this but..
You can use the 'follow me' tool like you would the push/pull but that tool will pull a face along a designated line in a model or you can draw a new line in the direction you want the face to go and follow that.
There is also an option to re-orientate the axes. You can also assign individual axes to each group but I'm not sure if the 2017 version supports that?

 

[Screwdriver]

This next hinge part took ages to concoct. The JWST has angled rectangular section support arms and I wanted it to look right. Sketchup hates odd angles and the odd axes makes it super complex because the "push/pull" tool basically heads off on the axis which may or may not be along the body of the part if the same "group" has odd angles in it.

You might well already know this but..
You can use the 'follow me' tool like you would the push/pull but that tool will pull a face along a designated line in a model or you can draw a new line in the direction you want the face to go and follow that.
There is also an option to re-orientate the axes. You can also assign individual axes to each group but I'm not sure if the 2017 version supports that?

Thanks Luke, I am certainly no expert when it comes to Sketchup despite using it since forever. I just dive right in and use it like you'd pick up a pencil. Push-pull/follow-me is a great tool for moving items along a defined path but here's my problem: I am using Sketchup to figure out where that path is. It's like drawing on a blank sheet of paper except actually it's a blank 3d volume.

Once you have a surface defined at one point and it is oriented towards another point then yes, you can just drag the projection along your chosen path. In this case, defining that path within the model was something I found tricky. Here's an example:

The hinge (middle part) is at the base of a long slender articulating tripod which supports the secondary mirror. It is fixed to the backplane of the large primary mirror through a triangular hole which connects three separate parts and holds them together: a large spherical surface which transitions to a "flattened"coarse polygonal surface which itself transitions into a flat plane so that I can actually print the thing!

I made that a separate part so that I can remove it to polish the mirror(s) and it's secondary function is as a locating pin. It connects three separate complex prints together with a long screw that goes right through the back and also holds the mirror wing hinges (not shown) to the back of the primary mirror assembly. It is also high up on the list of things which will probably break!

From the hinge axis forwards, the extrusion heads off at a compound angle to meet the secondary mirror at the top of a folding tripod structure. Not only that but the "trick" part I refer to is that the cross section of that extrusion is then rotated through that axis at an angle to mimic the actual structure of the telescope.

The hinge is one of a pair which I have mirrored through the centreline of the model. Sketchup takes a pretty good stab at "parametric" modelling with the use of "components" but the downside being you can't fiddle with group axis in a component (grouped AND mirrored) without connected parts firing off into space and ending up further from the moon than the actual telescope!

I have very few known dimensions to work from. If anyone can find a decent source I'm all ears. I have some critical dimensions: primary mirror, deployable tower travel (1220mm!), overall length but no detail. Yes I have seen the "blueprint" picture and there is a magnificent Maya 3d model out there but building the ISIM and enclosed instrument modules was pure guesswork derived from hundreds of pictures. At some stage I have to bite the bullet and make a print, six to eight hours for the major assemblies so there is constant pressure only to "tweak" the model in such a way I can reuse an existing hard print. Those recalcitrant telescopic boom arms are a case in point. The "new" design has to fit on the "old" body or it will be another year before I'm done.

I was particularly pleased to finally end up with a secondary mirror arm that folds up and over the model so that it comes to rest in its flight position at (what looks like) the correct orientation. It is no exaggeration to say that one millimetre "off" in either the length, width or tiny angle in the axis and those positions can be so far "off" the whole thing looks "wrong". I can see that issue in many of the other models out there because I know just how tricky it is to nudge any of those parameters just a tiny bit...

 

[Screwdriver]

Bit of an update. Still not completely solved the sunshield problem i.e. htf to make a folding sunshield when the real thing uses as thin a material as it is possible to find even for use in a tiny model.

Getting there though:

It now folds much more convincingly to the real thing.

and there it is, gone.

 

[Seadog]

Great progress, Steve.

 

[Screwdriver]

Eureka! I have it. This isn't exactly perfect but this first prototype works a whole lot better than I had feared. There's a lot to go wrong and I really wanted to include those semicircular arcs which mimic the sunsheild on the original. They also provide some tension.

Bottom first because I need to build up the stack one at a time and retro fit to the deployed model. By the end of today I should have a much better looking version with all five stacks.

The other thing is, this view of JWST is the only thing we could possibly ever see of the actual telescope. In the final design I want to be able to mount it and display it in this orientation. The rest of it, including the giant gold primary mirror will be a dramatic "reveal".

 

[Screwdriver]

Happy with the solution to the screen issue, need to work on my execution.

Just need to fit some optional extras and we're done...

Bottom screen has additional coat of silicone hence the off colour to be sympathetic to the real thing.

 

[Screwdriver]

PS I also added the creases where they oughtta be. As others may agree, being OCD requires a fair bit of commitment. The prototype two posts up is the same material just given some treatment.

 

[slim_boy_fat]

being OCD requires a fair bit of commitment.

Doesn't it just.

 

[DanZac]

Thanks for sharing your project with us, the end results looking great and well worth the effort you've put into it. Nice to see something different take shape.

 

[Screwdriver]

Quick video to show it off before the real thing gets pounded into dust by micrometeorites...