Saturday, 4 December 2010

Twelve (29/11/10 - 4/12/10)

So this is going to be a large update, first off it was the deadline yesterday and sadly the bag was not complete. Secondly, I had to alter the acrylic ribcage design, this was to fix the problem I mentioned in my before post (about the styrene edges drooping) but in the end I had to re-build them completely. Anyways, I'll start in chronological order from how the model was at the start of the week and finish with what state it was in yesterday, Monday:

The old ribcage and how it was further supported by 1mm aluminium panels bent into shape and then glued onto the structure. This actually worked surprisingly well and made the overall piece a lot more sturdy, but even before I did this it was still quite sturdy.

Along with that I had also attached one side component to the middle component:

The inners were cladded due to the acrylic part of the middle component being covered in scratches from where I had filled it to give the appearance of a solid structure. The cladding provides a flawless finish, but even then I could still fell the "lumps" of where the acrylic splines stick up between the laminate pieces, so I ripped it off for further sanding.

Back to the ribcage and why I needed to re-do them completely. While the aluminium panels seemed liked a good idea at the time, when it came to cutting the edges off the components (so I could attach a slither of the laminate to either edge) I had to use the metal cutting bandsaw which pretty much destroyed the cladding:

In all honesty I jumped into this idea without putting enough thought behind it, maybe it was due to the impending deadline that I wasn't thinking straight, because even though I could sand and fill the styrene back to its flawless finish I hadn't probably thought of how to attach the laminate to these newly cut edges. Just gluing it wouldn't be enough as there is a lot of pressure going on the entirety of the structure especially the edge where acrylic tubing is. The only feasible way would be to drill into the acrylic splines from the top and slide a 1.5mm copper rod down it which would then slide up into a corresponding hole in the laminate.

The risk of it breaking due to this weak area was one too great seeing as this is meant to be a working product model so it should be able to withstand constant opening and closing of these parts. With this in mind and time running out I needed to come to a decision on which path to take: go with the original ribcage and hope the joint doesn't break or re-do the splines so that they're one piece which runs through the laminate as well. This may sound confusing, but thankfully I have photos of what I mean, using the same idea as I did on the middle component the new splines would be attached to the laminates with a 1.5mm copper rod running through both:

The laminate which would sit on top would have pre-drilled holes so that it could just slide right on. Going with this re-design at the last second really put me back a day or two as not only did I have to get back on the laser, but I had to re-build them, well actually one due to time constraints. With the deadline pretty much the next day and due to the snow not being able to stay late, I had to prioritise what needed to be done but in a realistic manner, I choose to get one at least the rear component assembled.

I didn't spend the last week solely focused on these parts though, most of the time I was sanding the main body (which consists of the rear and middle components assembled). If you can remember from one of my earlier updates I said I'd only attach one side component until all parts were finished and once the front and rear components were threaded into the main body attach the other side component. Well this went out of the window too as I went ahead and attached the other side component so I could sand and fill that when it was attached to the rest. But to allow me to still thread the alumuinum rod down its length I drilled a hole which went straight through the side component, so instead of attaching, sanding and filling the entire component at the end, all I have to do is fill in the drill hill.

Anyways, this is how the bag looked yesterday at the deadline:

The newly designed rear component which still consists of an acrylic ribcage but also implements the laminates to give a constant solid edge: 

This way still allows me to attach the elastic straps as I did previously, after which the back would be cladded. But before I permanently attach these straps I plan on filling and finishing the front to a good finish without worrying about accidentally spraying the straps in the process.

As you can also see I attached the draw-bolt locks as well as the handle. This proved me to be somewhat difficult (and I didn't get it done completely yesterday) as the screws would go though the laminate (which is only 5mm thick) and poke out the other side. The smaller screws on the locks don't protrude as much but for the larger ones (6mm) used on the handle I had to saw off the excess, if I had the time I would've done the same for the smaller ones. The cladding on the internal side (once glued in) will cover these holes, same goes for the holes caused by the feet.

Realistically I need two weeks more to get it properly finished, what's left to do:

 - Finish the main body along with spraying
 - Finish off the rear component, this includes attaching the straps as well as sanding, finishing and spraying
 - Cut out the front component and do the same
 - Assemble the newly design preparation table (this has already been cut out)
 - Lathe the feet and shoulder strap holders
 - Vacuum form the internal boxes

Sounds like a lot, but the most time consuming part is the finishing which leads me onto a point few others share with me. As this model is actually going to a client who will be using it as a demonstration model during meetings and that it has to be up to a perfect standard, this includes the finish and how it works. If I rushed it towards the end to meet the deadline I'd only have cut corners which I'd have to go back to to correct before handing it over to my client. I'd rather not do this but instead take the extra time to get it right the first time. It's not like my past models in which flaws could be hidden by camera angles and post touching up in photoshop.

Either way it will be left in this state until I can work on it again which will hopefully be the start of next term.

Sunday, 28 November 2010

Eleven (26/11/10 - 28/11/10)

With the deadline less than a week away time is quickly running out. At the end of last week I had glued one side of cladding to the front and rear compartments as well as thread the elastic material through it, as this can't be done once the other side is cladded. The other side would've been cladded if it wasn't for the lack of super glue, so it has been put off till tomorrow, none the less progress has been made:

This is what the front and rear compartments look right now, with the elastic interweaving it's copper rod ribcage where it couldn't go through the acrlyic splines. This problem was thought of when I was digitally building the splines so pre-cautions were taken; that the splines had a 1x25mm slot cut out so the elastic could slide through. But not all of these slots came out perfect, most warped meaning I had to thread the elastic underneath the splines. On the other hand, when it did work, it worked a treat:

The elastic ends were hand sown together to stop them coming out completely. Thankfully all of that will be hidden once the top cladding goes on, for the underside it looks like this (the gap in the styrene is where the elastic is threaded through):

And it working:

One problem I've noticed with the cladding is that the actual edges aren't as straight as I'd like, it's hard to describe without a photo, but I'll get one tomorrow. One way of fixing this is to use one of the laminates or heat bent acrlyic and cut off a slither from them and attach it to the top of the front and rear compartments. This would also help when screwing in the drawbolt locks.

Thursday, 25 November 2010

Ten (24/11/10 - 25/11/10)

Finally, giant strides. Even though weeks behind schedule I've actually got the main body of the bag in a somewhat assembled state, hopefully people will now start looking at it in a different light.

The big development since earlier in the week is the completion of the middle compartment. It took far longer than expected, mainly due to the epoxy resin being too weak for the job, which was the 5mm edge of the acylric to a milled off flat surface on the two segments of the laminated plywood (which had the slight curve), this photo should help if that was too confusing:

As you can see the plywood is also unsupported so they were incredibly fragile by themselves, for this reason I had to make two makeshift jigs so that they could be sanded and drilled. The jigs consisted of a simple 2x4 cut to the correct length so that the plywood didn't bend back on itself and double-sided to a larger MDF piece.

The reason they are in 3 segments is that 2 lasercut splines with the acrylic tube profile is put between them to make the middle tube in the hinge secure. If that once again sounded too confusing, this photo should clear it up:

Either side of the outer segments is a lasercut profile of the side components. Once again this is to keep everything level, but more importantly to stregthen the indivdual segments. Each segment has about 10 holes dripped in its edge where copper rod is slotted in, which then goes through the lasercut splines into the adjacent segment. These copper rods also act to keep the plywood up so that it meets the side component outline, as seen in this photo:

For example, this segment is sandwiched between a lasercut spline and the side component outline, so that the copper rod will thread through the segment and outline into the actual side component (which will be attached to the outline). This should hopefully make these segments stronger and less prone to sagging, which they do a lot. Also this way of attaching the side components to these outline allows me to spray the front, rear and middle components separately as the metal rod between them can not be threaded until the side components are attached.

On the front and rear side of things, both have had their internal structure done and are now in the process of cladding, which is proving to be a bit tricky, especially as I don't have a former. Anyways here is the bag assembled in skeleton form:

Tuesday, 23 November 2010

Neun (22/11/10) - 23/11/10)

Just for a quick update, the front and rear components which are now made of an acrylic 'ribcage' is coming along nicely, but by just putting it together I've noticed how fragile it really is. Enough to hold it's own weight and that, but I'm more worried about the folding preparation table inside which is quite heavy. One possibility I thought of is making it again with a blue foam interior and 0.5/1mm styrene cladding, this would make it ten fold times lighter, but it's all time permitting. On the plus side if I did do it this way I could add a spline (similar to that in the bottom photo of this post) which would connect to the acrylic tubing, giving it once again a larger surface area for gluing.

On the other parts of the bag, I've got my handle along with tuck-tiles (which I need to pick up) and the middle compartment is being put together, so hopefully that will be done along with the front and rear tomorrow afternoon. I also managed to flatten my chemiwood side components as it had wrapped when on the mill, this was done by putting it in the oven and then clamping it down to a piece of MDF, this means I can finally sand them down which has been long overdue.

Enough words, here's the ribcage as it stands this evening:

Sunday, 21 November 2010

Eight (18/11/10 - 21/11/10)

Well a lot has happened since the last update:
  1. Laminating the front and rear components has been abandoned
  2. The original former for the jig has been destroyed
  3. Moved onto the emergency back up design 
Why did I abandon the lamintating? Well for one, the edges after the curve would jut out as seen in the photo:

Even though these parts are excess and would be cut off, they had an effect on the curve which is the vital part, this effect is that they have stretched it out of the radius I need, turning it into a more eclipse segment than a circle. I can't think of a reason for this as the jig disallowed any movement for the plywood and I had enough clamps (both horizontally and vertically) to hold the formers firmly in place:

The middle photo is where I had to rip one of the side parts off to get the laminate out once it dried, this leads on to why the former was destroyed. As there is immense pressure put onto the plywood, some of the glue on it drips out of the side, thus gluing the laminate to the jig, in hindsight I should've removed the side bits completely when doing this, but having learned from heat bending acrylic, it does have a tendency to veer off to one side. For a simple fix I made these side bits removable so that I could just unscrew them and slide the laminate out, but alas this didn't prevent it breaking; having left the plywood dry over night I began removing the clamps and formers, only to find the smaller, original one stuck to the laminate. I pulled it off only to have the plywood, which was glued and nailed, peel off:

For what it's worth the newer former was fine, even though that was to be the last time I laminated.

Before I put the jig behind me, I noticed I didn't have a photo of it and the two formers:

As you can see the plywood was slotted into the trough thus leaving no room to jut out after the curve, so it perplexed me why it did. Although I didn't want all of this hard work and plywood going to waste I used a laminate to create the middle component which has a slight curve either side, although it's so slight it's hard to tell from the photo:

Moving on swiftly to the backup plan. This idea of splines running between segments of the front and rear components came to me once I started laminating. The hinge is my Achilles heel as it was always going to be the most fragile part even though it'd be the part which is most used. Having used Tensol on my earlier tests with acrlyic, finding a strong enough substitute to attach the acrylic tube to the plywood was always going to be a tricky one, in the end I was going to rely on epoxy resin. After little thought I knew that wasn't going to be enough so I thought of having splines between segments of the laminate which had the outline of the rod attached, these would not only be glued to the laminate but also threaded with 1.5 copper rods going into the laminate sgements either side:

The acryic tube would then have its ends glued to the splines and attached to the laminate down its length. Once the laminating didn't work out I had to resort to the idea of just using the splines, with 0.5mm styrene cladding to give the appearance of it being solid, when in fact it's hollow:

This does have two advantages over it being a solid; it's that much lighter and its profile is more accurate. On the downside though; it's that much more fragile and being hollow attaching the tuck-tite locks is going to be a problem.  However as the middle component is solid and segmented by the splines it doesn't suffer from these issues, which is a good thing too as the handle (when it arrives) will be attached via screws.

Wednesday, 17 November 2010

Seven (15/11/10 - 17/11/10)

First things first here's the trough fix for the jig I talked about in my earlier post:

Now onto the big news, I've given up on heat bending and moving on with laminating. This dramatic change comes as a result of too many failed attempts with acrylic; mainly that the corners warp giving an uneven profile which would be too much of a hassle to fix, in other aspects it's really hit or miss. In hindsight I should've made the jig that much longer so that I could heat bend the piece and then cut off the warped corners, but I don't have time to do that now.

With laminating I have that much more time to adjust the jig and clamps before the glue sets, one major problem I had with the acrylic. I tried my first full length piece yesterday and when I got it out this morning there was a nasty lip in it where it hadn't been flattened throughout:

To fix this problem, and one that I should've done from the start is make the jig double sided, so that both curves can be done at the same time. This is what I achieved today, making the former jig for the other half, along with milling out the side components to make them lighter:

The issue I had with laminating earlier is how to attach it to the acrylic tubing. This issue has been resolved with my idea of laser cutting outline splines which run down the curve, it's hard to describe it words but tomorrow I'll have the drawing done. These splines will be attached the plywood via copper rods which run down its length.

Sunday, 14 November 2010

Sixth (11/11/10 - 14/11/10)

Having thought my jig was complete I soon noticed one key aspect of the design was missing. When I tried a few tester pieces on it, getting the negative curve former (the part which slots on top) in all the way was a problem, as the acrylic would jut out from the one sided trough, it's hard to put it into words but sadly I forgot to take a picture of said situation. The fix to this was easy though, I cut a slice off the negative former and glued into into the jig creating the trough from which the material would sit and not jut out. With that done I got onto testing it out again with mixed results, in that the clamps which held the former under pressure had to be balanced on either side and already at the correct distance apart so they can be clamped in place quickly as time is of the upmost importance.

While trying out it with veneer and tightening the clamps I heard the former jig crack, thankfully no damage was done by obviously I needed to fill the gaps between the splines some more; this was done by simply slotting down correctly sized pieces of 12mm MDF (photos of this will appear in the next update).

On my first attempt at the final piece my calculations came up too long and the finished piece was 30mm longer than it should have been, but this is fine as I can use the curves for the middle compartment. I altered my methods in that once one side was heat bent, I'd turn it round so it met the other side of the jig, clamp it down and then heatbend the straight side down over the jig. This is instead of slotting it into the trough and hoping the heat bent curve on the other side matches the jig.

Enough talking here are photos of the heatbent pieces; the veneer and the too long acrylic:

A small slice of acrylic to test out the new method of heatbending to make sure it goes to the correct length:

The smaller slice under the CNC side:

Wednesday, 10 November 2010

Fifth haul (8/11/10-10/10/10)

Today marks the completion of the jig and a development in the material side of things. First off, the jig and its construction:

Having learned from my earlier jigs; trying to get a uniform curve is pretty hard when bending it with your hand, the best way consists of a two part jig which can be clamped together to hold the acrylic in place. On my previous one of this design the actual curve former was made of different MDF parts, sanded together to create the curve. However on this one, which has the width of 410mm trying, to maintain the exact curve along all those bits would be far too time consuming, this is why I went with using MDF splines and veneered 1.5mm plywood over them to create one smooth, uniform curve across the entire structure.

I've yet to try it out, but from my previous one of this design the curve it produces is far more accurate, and as this one is far more secure and flawless the overall finish should hopefully be near enough perfect. On the material side of things, I've got my elastic strip for both the rear and front components along with webbing for the carry handle (the shoulder strap will be brought tomorrow):

Along with those two I was told about a haberdashery website where I came across this bag fastners (MacCulloch and Wallis), which has exactly what I need. I'm going to try a few locksmiths tomorrow to see if I can find anything more precise, if not I'll be using this one.