Scrolling text

Latest News: Doris blew the roof off! But I got it back on and added some tie downs to prevent a repeat.

Follow by Email

Monday, 20 February 2017

The Ring

There are not many parts of an observatory that are truly critical to its operation but the running ring is one of them:  It must be strong; it must be round; it must be level and it must be centred.

This then is the most important thing we have yet worked on, we need to get it right.

I spent a lot of time fretting over this part of the build.  

The only feasible material to build the ring from is plywood.  There is no way you can get the entire ring out of a single piece of plywood, plywood comes in sheets eight feet by four so the ring needs to be cut in sections then joined together to form the whole circle.   If your budget allows there are many companies that will laser cut sheet plywood in whatever forms you need, this is significantly quicker and vastly more accurate (and therefore easier to assemble) than cutting it yourself, however my budget did not allow so I had to cut them myself.

As I would have no choice but to buy the plywood I did actually try to plan out how much to buy,  I broke out Autocad and sketched out the arcs I needed on a template the size of a sheet of plywood, I figured I would get nine arcs from each sheet, and that I would need 10 arcs to make a complete circle.  I wanted 3 layers of plywood so I would be needing four sheets.

Now we meet another of my friends who can do Things.  I was talking over the problem of accurate building the ring with my friend Rob and described how many arcs I thought I would get out of a sheet and he said "I don't think so... get me some paper, I have a pencil" What followed was a lot of complicated maths ending with the declaration of "you'll get thirteen."   Well now we had to settle it didn't we?
See that big grin on his face?  Yeah, that was as he started drawing the thirteenth arc. Never bet against a maths professor (he really is)!

As you can see in the above photo we had a bit of a production line going that day and after a few hours we were able to layout the first layer of "Rob's Ring":
Not bad, its the most circular thing we have built yet!
So that was the first layer done.  Having lost my helpers I modified the drawing jig we had set up to take a router instead of a pen this made the process a lot quicker but I did have a couple of minor accidents.  The jig works by having the main plank slide up and down against a peg.  Depending on which way you are swinging the arm the torque of the motor will push the arm against the peg or pull it away, if you are swinging from the wrong side the router will dive away from your intended cut on the torque of the motor.  I found this out the hard way.  The way I set mine up I needed to start the inner arc cut from the right hand side and the outer arc cut from the left that way the torque was pushing against the stop on each cut which meant I didn't have worry about following a line anymore.

So after a few hours of routing we have the the following piles of wood:

It took a couple of weeks to glue this lot together I built it up into two semi-circles which (with an eye to any future house moves) I then joined together with nuts and bolts so that they could potentially be separated later on.

Looking through my photos archives it seems that I never took a picture of the completed ring.  That's probably because I came to feel that it was something of a millstone round my neck.   Having built a strong, round circle I now spent months worrying about how to get it level and centred.  If I can't achieve those two things the observatory simply will not function.  The whole summer went by and I hadn't committed to fixing it down on top of the walls.

Towards the end of the summer I was describing my worries about this task to another friend, Richard, (Richard is a builder but he has so far studiously avoided getting involved with this project!) he asked me one simple but oh so important question:

"Do you still have the routing jig?"

Of Course!  how could I not have seen this gloriously simple solution?   All I needed to do was wedge a bit of timber into the pipe in the centre of the observatory and fix the jig arm to this at the right height, then as I turn the jig arm round I can immediately tell when it is off centre and adjust!  Even better, if it turns out that the final ring isn't properly round I can just turn the router on and make it round in situ!  How had I not thought of this before?!   

Richard also pointed me towards frame packers as a way of levelling the ring.   So I set up a post in the pier, spun the router jig round a few times, nudged the ring this way and that and within half an hour had a centred ring!  Amazing, several months had gone by while I worried about this task and here it was, done between getting home from work and having dinner.   

The next job was to level it up, I used a laser spirit level for this, the technique went something like this:
Step 1: place a piece of wood on the top side of the ring
Step 2: shine the laser level at the piece of wood
Step 3: mark the level of the laser and the position on the ring on the piece of wood
Step 4: move the piece of wood to a different location on the ring
Step 5: repeat Step 2, 3 & 4 until you return to your original position
Step 6: identify the difference in height between the lowest and the highest point
Step 7: pack the low point up until it is the same height as the high point
Step 8: go round the ring packing any gaps which are wide enough to cause sagging
Step 9: go round with the laser level again make any adjustments as needed

With the levelling done (which took a couple of hours) I went back around with the jig to make sure everything was still centred, then I screwed the ring down on to the top of the walls.  

What a moment that was!  The sense of relief!  No only did this moment stiffen my resolve to get on and finish the project it also significantly stiffened the nature of the walls, now there was no movement in the walls or the floors, now we have a load bearing structure upon which we can place a rotating dome!

Saturday, 18 February 2017

The Walls Go Up

Up until now we have essentially built a deck, now it's time to go up!  But how far?

There are competing design criteria here, higher walls means a taller door which makes for more comfortable ingress and egress but walls that are too tall either impinge on the field of view of the telescope or require a taller pier which then means you need steps to get to the eyepiece which can be problematic in its own right. I figured the most important things is to be able to use the telescope with the least difficulty possible since I will spend a lot more time using the scope than the door!  For me this meant the scope being about 1.6m above the floor in the horizontal position so I set the wall height at 1.5m to allow for the running ring and the rolling structure of the dome.   

Here we find yet another reason not to build a decagon... You have to build more walls!

There are a huge number of ways to build walls depending on your skill level and appetite for doing carpentry.  To keep things as simple as possible I opted to build simple butt jointed frames, one for each section of the decagon plus six more for the storage pods giving a total of sixteen panels, four joints on each panel: sixty four joints to be made... Best get on with it then.

As I have mentioned before my carpentry skills are rudimentary, this and the fact that I needed all the walls in place quickly so they could be fixed together led me to use nail plates to join the timbers rather than trying to make nice joinery.   A word of caution though, whatever method you chose, make yourself a jig to work to, it is all too easy to make your frames skew when you are bashing away with a lump hammer!   Conscious that some of the spans were bigger than others I also added supporting timbers in the middle of some of the frames.

Each of the wall frames was in turn screwed down onto the floor then to its neighbour in the picture you can also the temporary bracing timbers used to steady the frames while I worked on them.  Because these frames will not be visible from the outside I wasn't overly worried about the occasional gap between frames, these will be covered by the exterior cladding later on, what is important is that they be plumb, that is they must be straight up and down otherwise you will set yourself up for big problems when it comes to supporting the dome.

I would also like to take a moment to talk about the storage pods here, I made these the same height as the rest of the walls, that was a mistake, a much better idea would be to make them a little lower so that you can add a sloping roof to them I have had to put flat roofs on mine which may be a problem in heavy rain, we shall see!

All in all the walls went up pretty quickly, they may not be the prettiest but they are up and now we can move onto the most challenging part of the construction: the running ring.

Friday, 17 February 2017

The Floor

With the foundations laid it was time to get on with the floor!    I stood inside the decagon and tried to imagine what this whole structure would be like and I found myself thinking "its an awfully small space..." So I decided to make it bigger,  I laid out the floor joists with a half metre overhang giving a new overall diameter of three metres.   Structurally this isn't a particularly great idea, it creates the opportunity for a lot of movement in the floor but it did result in the foundations becoming almost entirely invisible so now the observatory appears to float above the ground with no visible support which is a pretty cool effect.  It later turned out that - again - this was a decision that would have serious consequences later down the line:  There are quite a few commercially made products that have two metre diameter domes: oil stores, sceptic tanks, geodesic green houses to name a few.  Small composite manufacturers will happily lay up two metre diameter hemispheres for you in fibreglass... but not so with three metres.  There are very few commercially made products with three metre diameters and the few that are available are hugely expensive to the point of out costing the rest of the build by an order of magnitude.  This decision basically committed me to building my own dome from scratch and that (as we will see) was not easy.

I had a friend over to help and together we set about laying in the floor joists and cutting all those awkward joints I illustrated for you in "The Design"  and after a few hours work this is where we were:
The "Pole star" stage
Then we had a tea break.  

I had planned to use the 18mm plywood that had formed the sides of the Observatory mk 1 to make the floor. The plywood itself would have added significant strength to the structure however it turned out that most of this plywood was not in good condition and (because it had been nailed to the timbers rather than screwed) most of it was too badly damaged to reuse.  So the tea break discussion focussed on what to do next.  Again the easy way out of the problem would be to just buy new plywood but a quick call to the local timber merchant revealed that sufficient plywood would cost a couple of hundred pounds and I wasn't keen on that!  Then Ryan's eyes alighted on the pile of pallets that I had recently collected for firewood.   "Why don't we use those?" he said.  Well why not indeed, wasn't that "The Plan" after all?
The first of many such piles to be converted to building materials
Well, I don't know if you have ever tried to dismantle a pallet without breaking the wood but it is not easy!
It is very annoying when this happens!
The rest of the afternoon (and most of the next week, ah who am I kidding, most of the next year!)  was spent trying work out an efficient way to separate the deck boards from the runners without breaking them.   I'm not sure we ever found a really good method but a Gorilla bar turned out to be good enough and if you have enough pallets then you can get away with breaking a few planks along the way, it is all useful. If I can't build something from it it can always go in the wood burner to keep me warm!

After I broke up a lot of pallets Ryan came over again the following weekend and we set to laying up the floor:
Looking good!
While this is going to give it an undeniably great look it was clearly going to take a long time to do, which isn't a huge problem. What was a big problem was that the spacing between the joists was too wide to support the planks sufficiently at the outside edges and they bowed alarmingly under loading.

This is the problem with repurposing materials for jobs they were not designed for and modifying your design criteria on the fly.  The joists would have been fine with the planned-for structural plywood floor but it was clear we needed some serious reenforcing if we were going to get away with using pallet wood instead.

Fortunately some of the runners from the pallets were a good size to make the extra joists and more carpentry ensued:
Lots more wood added to the pole star
Here you can see the extent of the reenforcing needed, each original radial joist has had an extra piece of timber on both sides to allow sufficient timber to attach the boards to and each "wedge" has had an extra radial joist added.  We also added plywood cross ties underneath the joists to provide some measure of load sharing to reduce the flexing at the outside edge.

Tests showed that the new arrangement provided adequate support for the pallet deck boards and we really started to motor along by the end of that day we got to here:
Notice the extension at the far side of the central ladder, this is to become a storage pod later on.
This was pretty slow going because every plank had to be laid in place, marked up, taken up, cut to shape, replaced and fixed in position.  There are ten wedges and two ladder section, each section contains sixteen planks so that is around one hundred and ninety two planks, each end had to be marked and cut at both ends, so that's three hundred and eighty four cuts, and that's just the first layer, we did a second layer to add strength... seven hundred and sixty eight individually measured and cut piece of wood. 

about two weeks after we first started laying the floor the first layer was complete and looking good
Layer one done!
It was now that I had a floor that I could walk around on that I recognised how annoying it would be to have things stored on that floor so I added further extensions to the joists to allow the construction of two storage pods to keep all the ancillary equipment in.  I am very please I did this and I heartily recommend that if you build your own observatory you give some serious thought to where you will store things when they are not in use,  it very quickly gets tedious tripping over boxes in the dark!

Having by now gotten quite good at the process of laying the floor the second layer went down fairly quickly, here we see my son helping me glue the boards of the second layer in place:
And with that done we were ready to start building the walls!

Tuesday, 14 February 2017

The Foundations

Having gathered the materials it was time to start.  First I built the two metre diameter decagon I decided on in the design phase, then I called on my neighbour.

My neighbour is a landscape gardener (I find it very useful to have friends who can do Things, as we shall see later in the project!) and coincidentally he had rented a mini digger and an auger for his own project and agreed to help out and dig the foundations for me.  Here he is scrapping the vegetation cover off the building plot:

Once the ground was cleared we marked out the position of the supporting piles and the centre point for the pier and then dug them out with a hydraulic auger (in the background you can see the decagon propped against the hedge).

Interestingly we found that the auger easily lifted large rocks out of the holes but would jam when it encountered a pocket of smaller fist sized rocks.

Having dug the holes we then filled them in again!  This time with concrete and some fence posts which were too warped for Martin to use in his work.  And here it is:

Here we can see another consequence of choosing the decagon over the octagon.   Due to the un-compacted nature of the soil and the inherent inaccuracy of the machinery we couldn't get a post under each joint in the ring, instead we had to go with every other joint which is less than ideal as it allows a lot more flexure in the bearer ring and a lot more stress on the unsupported joints, hence the reenforcing plates you can see sandwiching each joint top and bottom.  Had I built an octagon we could have had a post under each corner and a lot less faffing about with the joints (but even now I still hadn't realised the real problems that lay ahead in using this shape).  But nevertheless we have reached a milestone, we were out of the ground!

The central pier is worth talking about, in total it was two and a half metres long, of which a metre is buried in the ground.  We cut slots in the side and placed reenforcing bars through these slots and then poured about two thirds of a cubic metre of concrete into the pit making sure it went into the tube through the slots, once up to the level of the ground we poured more concrete into the tube to fill the inside to the same level as the outside.  Safe to say this pier is not going anywhere anytime soon.   Once the concrete hardened we had some fun playing tubular bells on the pipe, but since vibration is a very bad thing in a telescope mount I then filled the pipe with sand and grit to damp down the ringing to a dull thud when tapped.   The final touch was to skim some fine cement over the top of the concrete block and summon the family.
All hail the mighty pier!

The foundations are done! Onto the next phase.

The Foraging Begins.

Having decided that I was going to use recycled materials for the build I took an inventory of what I already had and then made a list of things I would need to acquire.

I had a lot of structural timber, certainly plenty to get the early phases of construction done from the dismantling of the Observatory mk 1.  The most obvious and urgent thing that was missing was the pier to mount the telescope on so I started visiting local scrap merchants and metal works looking for old pipes or unwanted off cuts but couldn't find what I was looking for.  I was just beginning to think that I might have to buy something new right at the start of the this project when I struck gold!

I was walking to work one morning when I noticed workmen pilling up signposts at the side of the road.   They were replacing redundant signage following an extensive project to reroute traffic through a local shopping area and there were half a dozen road sign poles just stacked there!  I asked one of the workmen what was planned for the poles and he pointed me to the site foreman, he in turn was amused and intrigued by my plans and donated the largest of the poles to my cause.  He even had one of the guys help me load it onto my car which was nice since there was no way I could lift the thing by myself!

Its amazing what people will give you when you ask nicely.
So that was it: the first crisis dealt with. Four metres of ideal steel pipe for free, its hard to do better than that!  We were now ready to get started with the build!

The Design

Now, I would love to show you a whole bunch of beautifully detailed 3D cad drawings showing every dimension and jointing detail for the whole project and a rendered fly through video... I do own a copy of Autocad so I really aught to be able to do that.  In fact I spent several hours trying to produce those drawings but in the end I realised that learning how to do the drawings was probably going to take the same amount of time as the building of the thing and so I gave up on CAD and essentially started building from a picture in my imagination.

As yet no one has devised a way of printing out the human imagination so this post isn't going to feature many pictures!

Throughout this blog there is going to be a recurring theme of me essentially making stuff up and thereby making life somewhat more difficult for myself!  When I am telling you about these decisions in this blog I am now doing so with the benefit of hindsight and so may well wander off for a while exploring why this decision made life more difficult so that should you be planning to emulate my work you won't also emulate my mistakes!  I would strongly encourage any readers planning any project to research carefully what other people have done when building similar things and then try to work out the unspoken reasons for why things have been done in that way,  as a starter for 10 lets consider the basic shape of the Observatory mk 2:

It's going to be round! 

But now lets just think about that statement.   I am using recycled dimensional timer: 2"x4", these are famously not curved materials.  So although the final shape will need to have a circle for the dome to turn on clearly the supporting walls are not themselves going to be a circle.   What shape are they going to be?

Well, a square, while easy to build is going to look a bit odd, will lead to lots of wasted space and wide overhangs which will need to be supported somehow.   At this stage I was doing a lot of Googling and found that most self build observatories use an octagon as the base shape, if you don't know what an octagon is - or even if you do - here is a picture for you:
That shape - I'm sure you will agree - is significantly closer to a circle than a square is and so I applied a slightly deranged sort of logic and came up with the idea that adding more sides would make a shape that even more closely approximated a circle and this would be "A Good Thing".   Now I am sure that some of you have realised that I have taken a slightly shaky step towards the concept of infinitesimals here and are wondering just how far I descended into that rabbit hole... Well not that far really.  I quickly realised that adding more sides involved doing a lot more carpentry for rapidly diminishing returns in that each increase in the number of sides only brings us a little closer to a circle so I settled on a decagon.  Phew! We have avoided complete madness (sort of).

Lets have a look at a decagon:

There is no denying that shape is significantly more circle like than the octagon, what a good idea this is.

Here is where I am going give you - my dear reader - a really important piece of advice:

If you are going to build an observatory, build an octagon.

But why should you build an octagon?  This is a teachable moment in finding the unspoken reasons for doing a thing in a certain way.   Lets look again at those two shapes:
Now consider that you are going to have a pole in the centre of this shape where you are going to mount your telescopes, through which you cannot pass a supporting beam and that you need a floor to walk on which needs to be suspend above the ground so that you don't make the telescope vibrate when you are walking around.

Have you seen the problem yet?

Let me draw in some lines for you, first the octagon:

Four beams, four simple right angle joints and a nice big space in the middle for the pier to go in.

Now for the decagon:
Oh dear, it all started so well with the two red beams spanning all the way through missing the mid point, just like the octagon... things don't go so well when we start looking at the rest of the floor though do they?   We can't use full spans because they all cross in the middle where the telescope pier will be!  We could of course stop where the yellow lines intersect the red but now we have 6 joints to make with three different angles to cut  and no mutual load bearing, those two red lines are taking all the load with nothing to prop them up in the middle.   There are of course other arrangements we could use but they are all much more complicated to build than the octagon and all have less than ideal load bearing configurations.   Trust me on this, build an octagon.

Of course I didn't realise any of this before I laid out the foundations because, well because no other description of a self built observatory I found explained why an octagon was a better shape to use and I am not a professional builder so had no prior experience to call on...  

So a decagon it is 

Now how big should it be?

My biggest telescope is one metre long, so the maximum radius that the telescope can possibly sweep is one metre, but in reality we don't mount telescopes like that so really a two metre diameter observatory should be suffcient.

With those fundamental design decisions made I set off gleefully to mark out the foundations.

At this point I hadn't given much thought to the design or construction of the dome and I think I will cover that in a separate post because that is a significant piece of work all by itself! 

Monday, 13 February 2017

The Plan

When I first got into astronomy a friend and I built a roll off roof observatory, I got so much enjoyment out of having easy use of my telescopes that when we decided to move house in 2014 I knew I would be building a new observatory.

Observatory mk 1
In preparation for the move I dismantled Observatory mk 1 with the intention of rebuilding it at the new house but, unlike our old house, the new garden had un-obstructed views in all directions so a traditional dome observatory would make better use of the space than the mk 1.  I still had all the materials from the mk 1 so the Plan was hatched:

I would build the new observatory from recycled and reused materials wherever possible.

This decision was to have a profound impact on the build process and the final look of the Stoneage Observatory. It would also provide enormous opportunities for problem solving and acquiring new manufacturing skills and more than a few frustrations along the way.