800mm Telescope Project

[mapstar]

On 15/09/2019 at 13:58, jetstream said:

None of us needs to justify our telescopes or projects IMHO- this looks like a fantastic telescope in the making.

Agreed, life is too short and if you can do it why not. 

[Michele Scotti]

Material gathered from warehouse for the 'double H' mirror cell  bearing structure that is an integrated part of the mirror box - the long and thin one is an inexpensive levelling bar (the one for concrete work) - amazingly straight, the beauty of extrusion

     

All cut in a nearby shop - all 'paired' parts needs to be exact same length

For the same reason holes are drilled together.

 

This CAD screenshot is highlighting (maybe???) the structure for which we added some 45deg reinforcements during the execution - which were not part of the original FEA hence providing an additional safety margin to the results used so far to validate the structure.

  

   

Assembly in place using the azimuth table just after the drum sanding so hopefully the flattest thing we had laying around. The diagonals dimension is checked to have everything squared and small L shaped brackets are used to keep everything in position before the final assembly

 

Cheers, Michele

[Michele Scotti]

The structure is bonded with epoxy. Welding would introduce distortion unless done very carefully. Also welding tends to require bigger wall thickness whereas all elements here are 1.5mm thick. Some gussets are riveted in critical joint.

We used small L-shaped brackets to place the elements in a fairly precise way - the 2 main diagonals (corner to corner) were checked with a physical 'jig' to make sure the whole thing was 'squared'.

The finished part needing only the holes for the mirror cell which dictated the position of the crossing elements.

 

During the execution we added some six 45deg reinforcements - they were not part of the original FEA hence providing an additional safety margin to the results used so far to validate the structure.

Given the small thickness some big aluminium 'washers' will be adopted to spread the pressure of the fixings on the pac-men.

[Michele Scotti]

Follow-up on the azimuth table from #90

Surface preparation for bond adhesion - the strip needs to be 'normalized' as the CF layer was pretty heavy - 800gr/m^2- and left some pattern that we are concerned can cause a patchwork lack of bonding. So we poured some additional epoxy. Extreme flatness is not a concern/target in this phase.

 

No fancier way to apply some load to the 3 stainless steel arches that make the ca.1200mm diameter - we want to get the best out of this but it's not the final surface.

 

Removing some excess of epoxy - btw the one used is loaded with filler in order to control the flow and the fill under the metal strips.

 

Semi-finished part under some sun - not (just) for a nice pic but rather for the final "high-temp" bake - a compulsory phase for most of the epoxy formulations. Some epoxies reach their peak performance after several days/weeks depending of environmental conditions.

 

Next step: grinding the track

[AngryDonkey]

Love the images. This thing is a monster!!! 🤣

[Michele Scotti]

2 hours ago, AngryDonkey said:

Love the images. This thing is a monster!!! 🤣

Once we started to deal with the main elements we were surprised - in a good way- by the dimension.

Thanks for following!

[Michele Scotti]

Following up on Posted September 19  regarding the scope bottom part. Under the azimuth table, there's a sub-assembly I'd call a flattened 'tripod'.

CAD pic: Feet are not shown - along with a few other details...

 

 

What I call tripod is a 3-leg structure that houses a central journal and secures in position the 3 sets of azimuth rollers found at the end of each leg.

The central "hub" is in 3 pieces: a lower hexagon, a middle piece to slot the 3 legs in and an upper hexagon. The latter sustains the highest stress and can't be too thick.
The middle piece is shown here:

 

The central journal below in the schematics is made of a flange+bearing attached to the azimuth table and another flange with a fixed tube that matches the bearing.

 

 

We want a bearing to achieve a robust radial constraint during the azimuth rotation so that the rollers are purely taking vertical forces. Also, we need a hollow axis to manage all cables coming from the mirror box and the UTA.

After some considerations, the rear axle of a go-kart proved to tick all boxes. Its shaft is precise enough to match a bearing, it's hollow and we could make use of a matching sprocket flange. It's also dirt cheap. Here is the flange, made out of magnesium - more for the fun of machining Mg for the first time...been told it's sparky  £18 on eBay. The shaft is a section of a warped one - free.

 

 

In this picture the material for the 'legs' cut to length:

 

 

Last but not least a pic of an inspirational design - don't have a clue where it's coming from, saved on my laptop for quite a while...

   

 

Clear skies,

Michele

 

 

Edited October 3, 2019 by Michele Scotti

[Robert72]

On the aluminium box section frame, is the mirror being supported by the two long pieces? If so, what is supporting these long pieces?  Please don't tell me they are only joined to the shorter end pieces solely with epoxy?

[CraigT82]

53 minutes ago, Robert72 said:

On the aluminium box section frame, is the mirror being supported by the two long pieces? If so, what is supporting these long pieces?  Please don't tell me they are only joined to the shorter end pieces solely with epoxy?

Riveted L brackets, going off this picture 

Edited October 4, 2019 by CraigT82

[Robert72]

Ah yes. Silly me, I can see them.

[Michele Scotti]

On 04/10/2019 at 20:03, Robert72 said:

Ah yes. Silly me, I can see them.

Well it was well concealed. Here is a better picture before the bonding.

For the sake of discussion - the sections joining the lateral element of the H are 10 in total - per side there are 2 transversal elements with 3x 45deg reinforcements.  The transverse joints (being 60x30x1.5 sections) sums up to 4x261mm^2. The other 6- due to the angle are 6x369mm^2.

Total sum is 3200+mm^2. - and it's just the pure geometrical one.

Now take your  wallet and pull a credit card out - that's pretty much the surface to picture. Bonding can do wonder

The mirror is in the ballpark of 50kg. 

So gussets are there to make the assembly much easier and for peace of mind.

But do I really need them?

[Michele Scotti]

Following on the 'tripod':

Cutting hole in the of the middle section of the central hub and how the 'legs' slot in:

 

    

 

This is the machining of the flange that goes onto the azimuth table - this is the ID for a 80x50x10 bearing to match the go-kart shaft 

 

Here is the complete journal. The bearing is concealed by a piece of tape to protect from dust. The view is up-side-down compared to its actual positioning.

The machining of the Mg flange was pretty uneventful - no spark, no smoke, nothing! Phew. Just few machinings on the mill to fit on the upper section of central hub.

 

The central hub with the fix part of the journal - but missing the shaft

Lastly, another application for aluminium bonding....the three legs are put together with two 80x40 sections to generate a 160x40. The latter cross-section is somehow commercially available but pretty rare and expensive whereas the 80x40 is much easier to get hold of and the joining generates a nice reinforcing ridge in the middle of the section.

A 120x40 section-from two 60x40- would do too. There's actually little force going through this structure once the feet are placed close to the rollers.

 

Clear skies! Michele

[Robert72]

Michele, for a weight of 50Kg I am calculating your mirror is about 40mm thick?

Aluminium bonding is OK in a warm dry room, but have you tested it under load in a freezing environment?  I know from past experience that araldite gives up the ghost in the cold.

Early composite bike frames used to fall apart on Alpine summits for this reason (I believe low atmospheric pressure had a part to play also).

Well done to you and your team with the project, I enjoy following it.

[Chriske]

Dam...

Long time ago we used Araldite for an astro project.
We used Araldite to glue primary mirrors to flex'm into a parabola. Although it worked very well (until it broke) after a while, we abandoned the project.
It didn't last long,  never knew why, now I know.
Silly it wasn't mentioned on the labels back then.

[Michele Scotti]

2 hours ago, Robert72 said:

Michele, for a weight of 50Kg I am calculating your mirror is about 40mm thick?

Aluminium bonding is OK in a warm dry room, but have you tested it under load in a freezing environment?  I know from past experience that araldite gives up the ghost in the cold.

Early composite bike frames used to fall apart on Alpine summits for this reason (I believe low atmospheric pressure had a part to play also).

Well done to you and your team with the project, I enjoy following it.

 

Epoxies - a huge family of bonding agents actually - have developed quickly in the past few decades in terms of performance, durability and reliability - still they might be quite controversial.

I use a variety of epoxies that can't be found in a diy warehouse. yet commercially available.

 

Here's an example that comes to my mind to summarize what a proper application of epoxy can do. And it's a very British example, almost 25y dated.

So far there are no reported cases of failure. Only one during development as far as I know

 

 

Nowadays the applications of epoxy are everywhere and in industries that run intensive and thorough testings.

To be precise I am NOT an expert in the field.

[Robert72]

These might be worth a try....

https://www.easycomposites.co.uk/#!/resin-gel-silicone-adhesive/structural-adhesives

The VM100 VuduGlu is the one you need.  It is a Methacrylate rather than an epoxy.

Unfortunately, I have no connection whatsover with EasyComposites.

Edited October 6, 2019 by Robert72
Text addition

[Michele Scotti]

Here is a picture of the azimuth table with the central journal and the 'tripod' - this assembly is upside-down compared to its operating position.

 

The rollers are placed into a slot at the end of each leg where the azimuth stainless steel track is.  You won't see them from the previous picture as they are on the upper side of the leg, facing the azimuth table.

The four holes in the leg are going to accomodate some M5 aluminium rivnuts

I hope is clear enough and you are enjoying the build with me. Any questions? Don't hesitate!

Clear skies,

Michele

 

[Michele Scotti]

Following up on the 12 Sep post on the focuser board: 

 

The blank board was 2kg and with all cut-outs it got down to 1200g - still too heavy for our roadmap to a UTA within 15kg.

Hence the work with the router to hollow out the inner layers of the cut-outs while leaving the external side untouched. This is just one of the way I came out to 'add lightness' whilst keeping the robustness.

Final result is ca. 900g. After that a coat of clear epoxy to seal off the trimmed CF layers and protect the exposed wood.

Clear skies, Michele

[mapstar]

On 06/10/2019 at 15:05, Robert72 said:

These might be worth a try....

https://www.easycomposites.co.uk/#!/resin-gel-silicone-adhesive/structural-adhesives

The VM100 VuduGlu is the one you need.  It is a Methacrylate rather than an epoxy.

Unfortunately, I have no connection whatsover with EasyComposites.

Excellent adhesive. I've used it quite a few times 

[Michele Scotti]

Time for the spider!

The secondary is quite a large mirror -around 175/180mm on its short axis- so the spider comes with a generous central hub around 120mm in diameter.

Vanes are cut from a fairly affordable 500x400mm carbon fiber sheet 2mm thick -around £50. The inward ends are bonded to the CF hub with the help of a reinforcement specifically made with a angle slightly bigger than 90deg.

The 4 straight vanes attaches to the upper telescope assembly thought 3 coloums and the focuser board. The attachment consists in pairs of long M4 screws that poke through the columns and put tension on vanes once their nut are tightened. A small Aluminium U-shaped trough -kindly provided by Ikea- is used to clamp the vanes ends and connect it to the M4 screws. This try-out of the incomplete end might give a better idea of the clamping?

All bits were assembled on a 'jig' for a decent alighment...

This subassembly weighs in at 1,553g. In hindsight given the tension in the structure I think vanes thickness could have gone down to 1mm or 0.8mm with a potential saving of few hundred.

This is the final result:

 

Clear skies,

Michele

 

[mapstar]

Nice work Michele

I wouldn't have gone any thinner with the vanes as the weight of the 170mm minor axis secondary will create quite a rotational moment skewing the optical axis when the scope is nearer horizontal if you are not careful. 

My scope has a 130mm minor axis secondary which is 16mm thick on a 600mm dia upper tube assembly. I tried carbon fibre at 1.5mm thick but flex was a problem so settled on Steel 1.6mm vanes which are 120mm deep as they didn't flex 

I used carbon fibre to make the actual secondary holder which has worked really well. Packed with etha foam behind the secondary. 

[Michele Scotti]

8 hours ago, mapstar said:

My scope has a 130mm minor axis secondary which is 16mm thick on a 600mm dia upper tube assembly. I tried carbon fibre at 1.5mm thick but flex was a problem so settled on Steel 1.6mm vanes which are 120mm deep as they didn't flex 

I used carbon fibre to make the actual secondary holder which has worked really well. Packed with etha foam behind the secondary. 

Can I see any pictures of your scope here on SGL, specifically the spider you are talking about - I'm curious to have a look at that for similaritieswith our design. 

As of now I have an FEA that tells me that I'm ok both for modal and deflection. I might well have a look at that again.

[mapstar]

7 hours ago, Michele Scotti said:

Can I see any pictures of your scope here on SGL, specifically the spider you are talking about - I'm curious to have a look at that for similaritieswith our design. 

As of now I have an FEA that tells me that I'm ok both for modal and deflection. I might well have a look at that again.

Hi michele

I think It's here somewhere but about 4 or 5 years ago? I'll try to dig up a link but know a lot of the pics seem to have disappeared when they upgraded the forum.

It was just a thought and in no way saying it won't work as I'd been through it with trial and error myself rather than calculation and error analysis.  

I cant quite remember but my secondary weighed nearly 1kg? A secondary of your size must be in the region of 2kg+?  Its a primary in itself hung up there. 

Edited November 12, 2019 by mapstar

[Michele Scotti]

6 hours ago, mapstar said:

I cant quite remember but my secondary weighed nearly 1kg? A secondary of your size must be in the region of 2kg+?  Its a primary in itself hung up there. 

I put the density in the model and ran the analysis - I actually didn't know the wieght till you asked. However with some rough calculation your secondary should be around 850g whereas mine that has almost double the area but 12mm thickness id is ca. 1.2Kg. It's going to be 'glued' on its back rather than hung from the perimeter.

[Peter Drew]

45 minutes ago, Michele Scotti said:

I put the density in the model and ran the analysis - I actually didn't know the wieght till you asked. However with some rough calculation your secondary should be around 850g whereas mine that has almost double the area but 12mm thickness id is ca. 1.2Kg. It's going to be 'glued' on its back rather than hung from the perimeter.

The outcome of this will be of interest to me. I have two secondary mirrors for my 30" Dobsonian, one as a spare. Both are 150mm minor axis and 12mm thick. I have mostly "glued" secondary mirrors to backplates with success so did the same with these. The telescope then showed astigmatism  and knowing that the primary was ok I suspected the secondary, and the spare one when fitted for a comparison. The secondaries were independently tested and found to be good when unmounted so I subsequently remounted one in a "hung" arrangement. The astigmatism disappeared.  At least it's a simple thing to experiment with.   🙂