Page 1 of 7 123 ... LastLast
Results 1 to 10 of 70

Thread: WORKLOG: 'IOU' – GAM's QubeLok Cube Case with IO connector

  1. #1

    WORKLOG: 'IOU' – GAM's QubeLok Cube Case with IO connector

    G'day modders! I just met the great Tazz today, and was reintroduced to Modders Inc. I recall checking out worklogs and featured projects here in the past, gaining insight and instruction on the some of the things I wanted to do in modding. Anyway, since I've just joined, I thought there is no better way to introduce myself than to post my current in progress worklog. Sit back, relax and (hopefully) enjoy...

    WORKLOG: 'IOU' – GAM's QubeLok Cube Case with IO connectors facing Up
    Just to get the ball rolling, for my sake as much as anyone else's...

    Not extroadinary, but definitely not ordinary
    This cube case design is an experimental variation on the usual PC Case and PC Cube case designs. It is not a true cube, as all sides are not equal, nor does it follow many, if any, standard PC Case design principals.
    • - Firstly and foremost, the IO connectors will be on top, facing up, with the motherboard positioned internally to function thus.
      - Secondly, the motherboard will be located in the left side, facing inwards to the centre of the case and will be accessible by 'folding out' the left side panel.
      - All noisy, heat generating devices and other devices i.e., hard drives, optical drives, power supply and any additional 5.25" devices will be located in the right side. Either side will not necessarily be half.
      - All fans will be mounted in the floor of the case as intakes, with no exhaust fans but simply vents on the top panels.
      - Although it won't be watercooled, there will be sufficient space provided to mount two 3x120 rads on the floor above the fans (inside); one either side, ready for a future upgrade. Pump(s) would be mounted on the right side, while a reservoir would be mounted in the left side.
      - The case will be suspended on castors with surrounding mesh between the floor of the case and the surface on which it stands, providing the only path for ambient air to be drawn in. A brush of sorts will mate the case with the ground surface as a barrier to surface particles and other matter. The intention is sit on a 'hard floor' or right hand side of a desk.
      - All interior panels will be lined with sound dampening material.
    Note, click on images for higher res. versions.

    Thoughts to paper – the drawing board

    Although I have spent a lot of time drafting and refining my ideas and concepts and have fairly detailed drawings, there are many issues still to resolve. At this point, however, my itch is too fierce and the deadline for a particular case competition is looming. So, despite a lack of time and funds, I have to bite the bullet and start fabricating, working out some of the finer details as I go along.
    Here's an older example of what I have in mind. The plans have changed somewhat since. Note that the very light grey fans and rads and small 5.25" devices are possible later upgrades/changes/room for flexibility. The short length 5.25" devices (fan controllers and such) on the left side has been scrapped for the moment—determined unnecessary. Reduce the number of hard disks and you can increase the number of full length 5.25" devices.


    At this stage, I intend to make the entire case out of 1.6mm aluminium panels on a QubeLok carcase/frame and various aluminium extrusions for internal structures, however this could change with the addition of the odd acrylic panel here and there. That said, the intention is the have a very cool, literally, and quiet PC with 'Server/Mass Storage' duties in mind.

    $20 dollars worth of aluminium offcuts from a local aluminium fabricator, consisting of:
    • - Perforated aluminium sheet: 550mm x 290mm x 1.6mm, with 3/16" (4.7mm approx.) holes
      - Aluminium sheet 800mm x 270mm x 1.6mm
      - Aluminium sheet 700mm x 270mm x 1.6mm
      - Aluminium flat bar 500mm x 80mm x 10mmEarlier in the week, I visited Capral Aluminium and also purchased for the sum of $85 something:
      - Quarter sheet (600mm x 1200mm) of 1.6mm aluminium sheet ($25)
      - 6.5m length of black, non-lipped aluminium QubeLok tubing (cut to 2 x 3.25m for transport - $27)
      - Variety of QubeLok connectors. At the time of purchase I still wasn't sure of the frame I was going with, hence the variety of connectors includes:
      • - 12 x 3-leg (P30) corners connectors ($12.72)
        - 4 x 3-leg (P31) T-sections ($4.24)
        - 4 x 2-leg (P20) corners ($3.96)
        - 4 x 4-leg (P40) intersections ($4.56)
    Above all pictured, except the quarter sheet of 1.6mm)

    QubeLok connectors:

    Perforated sheet with QubeLok connector (1" by)

    For competition purposes.

    I also dragged out my collection of various aluminium channel, angle and flat bar ready to assess any additional needs.

    Unfortunately this is all I have to whet your collective appetites at the moment. Hopefully I'll get a chance to put a preliminary frame together this weekend. The frame will be a sinch, it's all the other stuff that I'm worried about and could take me forever


  2. #2

    WORKLOG: 'IOU' – GAM's QubeLok Cube Case with IO connector

    Workshop - Day 1: "I was framed!"
    Note, click on images for higher res. versions.

    The spies are out…
    Unbeknownst to me, my wife snapped this off. For legal reasons, I can't show you his face

    Ready, setty, cut
    Before I went to out to the workshop to start hacking tubing up, I spent a couple of hours in the morning refining the external dimensions in Adobe Illustrator and preparing a cutting list. Out of two 3.25 metre lengths, there's a little over a metre to spare. Planning here ensured that I maximised off cuts and was able to get the lengths that I needed out of the material at hand. As these are relatively short lengths the danger of ending up short was minimal, but best to be sure nevertheless.

    Compound saw shortcuts, quick jigs and chop suey.

    As I had to start with long lengths, the sliding out bench jig came into play. Initially I cut the 3.5m lengths into two portions, one portion being equivalent of one run of same length cuts. As seen in the photo, I clamped a block of wood at the correct distance to ensure that each cut produced the same length.
    This type of 'stop block' is not recommended. You should have a removable piece at the end (refer later in the log) so that nothing is gripping the end of the offcut. Otherwise, the blade can catch an edge and if there is something stopping the offcut from wriggling out of the way it will get thrown back at you. Ruining your workpiece and quite easily causing grevious bodily harm!
    Using this dubious method, I cut the 8 smaller pieces.

    For the longer lengths, I quickly whacked together another temporary jig. This is simply a plank of timber that is wide enough to stay intact once the tubing has been cut through. As you can see the, there is a clamp either side of the blade, holding the jig in place. A stop block is temporarily nailed in place at the end. Following the correct stop method, I've placed a slither of timber (10mm thick approx.) between the end of the workpiece and the stop block and then set the correct distance for the next series of cuts. When the tubing is gripped in place, I remove the slither, leaving a gap between the stop block and the end of the workpiece. This way there is nothing for the offcut to catch on during the cut.

    I neglected to stop on the last of the longer cuts and cut through the jig. No harm, as each of the subsequent lengths for cutting were smaller and smaller.

    All cut up

    An hour or so later and I had all the lengths cut and was ready test that I had cut everything correctly by loosely position each section and the relative connector. Checking like this help prevent issues later. At this time, I also checked the condition of each of the faces and positioned each to ensure the best surfaces were place inside where they'd be visible. All outer surfaces will covered and hidden by exterior panels. If the tubes were to be visible externally, then the best surfaces would have been placed on the outside rather the inside.

    One piece of tubing is shorter than the length of the two tongues of the connectors inserted into it, so these had to be cut down to fit. I also chamfered the edges roughly to ease insertion.

    Checked before proceeding. The tube is in contact with the butt edges of the connectors and there is a small gap between the connector tongues. Good.

    Assembly, disassembly…
    On to the assembly. I started with each of the verticals, joining the smallest sections first and then joined the horizontals to the outside verticals and then to the centre vertical.
    Back section done… or so I thought

    I completed the front section and then once again loosely put it together. Lo and behold, I stuffed up. This is the point of the (loose) test fit. There's an extra tongue on the rear centre that shouldn't be there.

    So, disassembly time. Initially I used the QuickGrips in 'spreader' mode to loosen it up and extract the centre vertical (the one with the mistake), as I hammered them in nice and tight.

    These are the only spreaders I've got and were too big for the smaller openings, so it was on to more voilent methods of separation—vice (with additional padding) hammer and block. I was careful not to overtighten the vice as the tubing could easily collapse under too much pressure. In this shot, the tubing is in the vice. As it happens, the connectors can be extracted quite easily with several moderate taps.

    r />Loose test fit and then re-assembly… again. All good.

    Burrs and edges
    I neglected to cover this in the initial assembly shots, but the end of every section was carefully filed on the two burred edges and all four inside edges. With any saw cut, where ever the teeth of the blade exits the material a burr is left, and sometimes 'breakout' in timber and brittle material.
    Not so clear in the photo, but the first shot shows the sharp lip or burr that is created on the lower and rear edges when using a compound saw. The second shot shows the edges after some gentle filing, holding the file near flat with the end.
    Please not gratuitous stubby holder placement. This case is intended to 'win' the OCAU Mar-May 2007 Modding Competition, after all!
    Note to moderators and admins, if you would rather see your site's stubby holder here, then you'd better send one on! Message me for address details.

    As mentioned, the inside edges were also filed to reduce the 'skimming' of the plastic connectors during insertion which can leave little bits of plastic that can prevent the gaps from closing up. This filing was very quick and rough with the file close to 45 degrees--before and after.

    Final assembly
    With all burrs and edges cleaned up, I proceeded to assemble the entire frame.
    A little push here and a litte shove there, knock knock, tap tap, whack and all done!

    Voila! One frame

    Note that the flash on the camera highlights the shiny edges between the connectors significantly more than you see under normal lighting conditions.

    Unfortunately, that's the easy part done. The hard parts of engineering and fitting all the internal brackets, drives cages, motherboard tray, opening door panels, IO connector panel, exterior panels, wiring etc etc is still to come.

    I hope to get some time in the workshop again tomorrow. If so, I'll post another update.


  3. #3

    WORKLOG: 'IOU' – GAM's QubeLok Cube Case with IO connector

    Workshop - Day 2: My new mother hinges on a winge and a prayer
    Note, click on images for higher res. versions.
    Hi again all, sorry, but very little progress today. Not much of an update either, but plenty of words and pictures to put you soundly to sleep for the evening.

    Sacrificial lamb
    Today I set myself one of the more difficult tasks I'd been envisaging, as far as brain time was concerned at least--the hinged (left) side panel with the rotated, inside out, fold-out motherboard tray.
    First off all, as I had decided a scratch built motherboard tray and,er… 'rear' connector panel was too much for the time being, I set out to extract the undesirable motherboard tray and accompanying rear panel from this old Compaq case – P3 Deskpro. This case has some nice innovations that I give good credit too, but not a case that I see myself using or modding in the future—besides, it was the only 'spare' case I had from which to extract a motherboard tray and rear panel.
    Note, at this point I should suggest that if anyone lives on the Northern Beaches of Sydney and have an old case, preferably aluminium, I'd love to hear from you. I need a donor

    The old lady gets 'pantsed'

    Not much to say here, it's all a case of unscrewing… until, of course, you get to those damn integrated rivets. The drive cage was fastened to the front and back panels by 'integrated' rivets i.e., the rivet is formed through the metal of the panels and not a separate piece of metal. It just made the extraction task a little harder and longer winded than need be.

    Slice and dice

    After starting the cut with a hack saw I went to the bench mounted jigsaw but quickly realised that the protusion on the motherboard tray prevented decent workability. So, the jigsaw was removed from the bench for some freehand.
    Marked for the cut. Note the internal cut lines had to be transferred to the outside for freehand accessibility. Also note the clamps to hold the bugger steady.

    I didn't care too much for a clean, straight cut as I knew I'd be cleaning up the edges on the disc grinder." alt="" />

    Shiney but marred
    With a motherboard and integrated back panel, I was ready to start on the prototype left-side door/panel.
    Note, as this piece of aluminium was one of the 'offcuts' I picked up on the cheap, it was already marred with plenty of scratches. Nevertheless, I tapped up the ends and the area that would slide on the proud rail of the bench saw. The sharp edges of the aluminium panel also binded with the table surface making sliding the workpiece difficult. The masking tape alleviated this perfectly. Also note the piece of aluminium angle that I inserted as a runner and support between the fence and the workpiece, clamped to the bench. This ensured that the workpiece didn't drop beneath the fence, as with this width of cut on this bench saw the end is unsupported and is prone to drop below the fence and ruin the cut. If you do this yourself, don't forget to include the additional thickness of the runner into your cut-width settings.

    Lest we protect
    Once I started manipulating the frame I quickly realised that a little scratch prevent was required—masking tape around the prominent supporting edges.

    The quarter sheet (600mm x 1200mm) of 1.6mm aluminium sheet that I purchased from Capral Aluminium earlier in the week and somehow misplaced—it was still under the carpet in the boot of the wagon (83' dunny door )

    Everything hinges on something

    Now for the real task of the day… I had been playing around with hinge alternative in Illustrator for some time but still wasn't sure of my game plan. I spent another broken hour or two trying to work out the best method to achieve want I wanted and determine the required dimension to make it all happen.

    • No externally visible hinge.
    • Easily removable panel/motherboard tray
    • All gaps must to be overlapped… and subsequently lined with adhesive foam pads or strips for noise isolation.
    • Avoid unnecessary sheet fabrication—straight lines are best

    As depicted in the background illustration, I came up with with a '??' type hinge (sorry, I can't recall the terminology for this style of hinge). The pros of this hinge is that I can retain a flat, straight, bottom edge of the door, have a secondary sealing lip behind the door , clear the lower fixed panel and hold the motherboard outright (weakly mind you) without the need for 'lid-stay' or such (although a lid-stary of sort will be integrated down the track as well as support legs for when you want to push down on those RAM sticks and tight cable connections.
    Using the dimensions from the illustration, I marked and drilled the guide holes in a sample piece of aluminium angle. The piece of angle will fastened to the inside of the door panel with screws or rivets. The next shot shows the initial prototype I played with before attaching it to the frame with masking tape for furher 'proof of concept'.

    The next few shots show the prototype in action, mimicking the intended hinge mounting and operation, over the fixed lower panel (visible). The smaller piece of angle on the right most edge represents the door panel thickness attached to the hinge piece. The screw (or other methods that I've conceived) will be removable to allow the entire panel to be removed from the case easily, motherboard intact.

    Not much of an update, I'm sorry. But hey, Rome wasn't built in a day. Also, please excuse any excessive spelling mistakes and incongruous language… it WAS a very pissy afternoon :toast:.

    There may be further progress during the weke and another update, but this depends on work commitments.

  4. #4

    WORKLOG: 'IOU' – GAM's QubeLok Cube Case with IO connector

    Workshop - Day 3: A hinge a day; a modder's dismay
    Note, click on images for higher res. versions.
    Hi again all, I couldn't resist the beckoning workshop today, much to my misfortune of course. However, although it doesn't seem like much it was a reasonably productive day, although, little to show for it. Today I managed to get a working motherboard/side panel door happening and almost a motherboard tray.

    One hinge or two?

    Setting off from where I got to yesterday I was determined, at the very least, to get a pair of working hinges mounted, and if I was lucky, actually mount a door on it.
    The prototype seemed to work so I set about duplicating it. I cut a second piece of 20x2mm aluminium flat bar to length and clamped them together with a pair of vice-grips. I then drilled the two holes using the prototype hinge as a pilot. The small hole is 2.5mm and the large hole/cutout is 8mm.

    Tapping the thread for the pivot screw. The pivot screw is only 3mm (M3) at this stage, but I will either upgrade this with an M4 screw or other 'hinge pin' arrangement i.e., no thread, and nylon bush in the frame.

    Once the the second hinge piece was completed, I set about making the two door braces and connections to the hinges.
    Note the method I've used here on the drill press to drill identical holes on two (or multiple) workpeices. There are three marks on each piece; two along the length and one indicating the centre line along the length. I established the centre length using the mounted drill bit as a guide and then set up a temporary fence on the bench drill table using a piece of anlge line and two clamps. A piece of masking tape was then stuck to the fence and across the drill point centre. I then drilled a hole through the masking tape, providing the precise entry point of the drill bit. Using the butt of a square, I then traced a line through the drill hole and up the fence. I then aligned the marks on my workpiece with the pencile line. The combination of the fence and the alignment of centre to marks give me precise drill holes without going to town measuring and marking up each piece to be drilled.
    Note, you may need to view the higher res. version to really see what is going on here.

    With two brace pieces and two hinge pieces I was ready to fasten them together. The hinge piece was aligned with the brace and then clamped with vice-grips. I then drilled through the new holes I had made in the brace through to the hinge piece. The holes in the braces were then redrilled with a 3mm drill bit, tapped and countersunk. Screw them together and voila!

    Because the screws I have are about 8mm long (the smallest I think in M3) I ground the protuding threads down to meet the surface of the hinge pieces.

    Mounting to side panel to the hinges
    Well, after stuffing up the pivot screw holes in the frame by about 2mm and subsequently having to file the erroneous hole down to be aligned the hole right on the right, I was feeling a little let down and more cautious. I temporarily fitted the hinges to the frame and fastened the braces to the frame with masking tape. This was so that I could place the side panel on the frame and drill through the panel and just into the braces, providing aligned guide holes for drilling through the brace. Because the hinge pieces created a 3mm gap between the brace and the frame, using masking tape again, I fastened a 3mm spacer piece between either brace and frame to ensure they were plumb with the frame before drilling. Once the 2.5mm guide holes were done I removed the sidepanel and drilled them through. After redrilling the sidepanel holes to 3mm and countersinking, and then threading the braces, I assembled them together with M3 countersunk socket screws.

    Finally, a hinged sidepanel!
    Note that 'lid-stays' or other bracing system to support the weight of the motherboard and its components will be incorporated at some point.

    Now for the 'railed' Motherboard tray

    Again, I can't remember the correct terminology, but using the table saw, the cross cut square and the fence as an initial guide for the first cut, slid the workpiece back and forth across the blade, moving the workpeice incrementally away from the saw on each stroke to remove the material off side of each end. These sections are cutouts to be to sit over the top of the rails beneath.

    Note the use of common thickness material to define the gap that I needed for the top panel when it is attached and another piece of channel to form a straight edge to align the rail to the sidepanel for clamping and subsequent drilling.

    The same drilling process of pilot holes and then redrilling, tapping and countersinking was used for these screws too. Also, as I was working by a little trial and error here, I had reconsidered the required distance between the horizontal rails and subsequently need to modify the reinforcing/spacing sections of the vertical rails. Another vertical rail will be made for the centre run of motherboard mounts. I actually ran out of 'noise' time and couldn't cut an additional length for fitting. Time to give up for the day.

    Testing the top panel fitting with a 1.5mm aluminium sheet offcut. And then the satisfaction shot of the day… one hinging side panel!

    I seriously don't expect another update until next weekend… but you never know


  5. #5

    WORKLOG: 'IOU' – GAM's QubeLok Cube Case with IO connector

    Workshop - Day 4: 6-32! This is for your mother? Sorry, we don't have that size?
    Note, click on images for higher res. versions.
    Hi again fellow modders. Once again, I couldn't resist the beckoning workshop today… I'm gonna' be sorry soon.

    EDIT: Important note, this post previously stated '3-32' thread/tap size erroneously!!! Should be 6-32. I must have had 3 from M3 on the brain that day. Apologies all. References in this post have been corrected.

    Another somewhat shitty, disfunctional day. I was all set to get the motherboard tray, MB back panel and right side panel finished today, but alas, if you don't have the tools, you can't do the job. I have a 'good' set of metric taps from M3-M6 (iirc), and two very good quality M3 and M4 taps. I also have a cheapo' mixed set of taps and dies. The set includes a 6-32 UNC tap and die… or at least it should. There are two taps missing from the set; guess which? Well, the 6-32 is missing, forget the other as who cares, the 6-32 UNC is what I need to tap the rails for the motherboard standoffs. Until I can get a starting point with the motherboard, at least, fixed by one mount, I wasn't game to try and come up with dimensions and transfer cutout shapes to panels etc. It just isn't worth the stuff ups and subsequent fixes, remakes, etc. So, no motherboard tray, end of story. Dissapointed I am, yesss.
    I did try to purchase another, two hardware stores and two quality trade 'tool' suppliers later and I am waiting on an order for a 6-32 UNC tap. 'Should be here in two days' Ugh!

    Anyway, despite this, I did make some progress.

    Motherboard back panel, er… top panel

    I cut the required piece for the motherboard back panel, drilled, tapped and fitted it to the side panel.

    A little teaser; I clamped an offcut to the centre vertical to illustrate the visible centre panel section. Note that this connector area will be covered with some sort of 'flip panel' as well, so when complete the box will be uniform and 'un-stepped'.

    I cut the third vertical rail for the motherboard tray and drilled one end of each to accommodate the screws that will hold them in place on the horizontal rails. Note, once I had marked up and drilled the first piece, I made another temporary jig using a piece of alu. channel and a block of wood aligned to the first piece with the drill bit inserted. With this method, there is no need to mark up the other two pieces as the drill will hit the same spot as the first one, each time. Not visible, but the channel is clamped to the bench drill table at either end.

    On the right hand side of the picture you can see that I have screwed in the first one at one end. I was going to tap my first 6-32 UNC hole for the motherboard standoff when I realised that I didn't have the tap anymore. From here I would have placed the motherboard, worked out its exact location and traced a profile for the backpanel cutout. But no…
    25a.jpg" alt="" />
    A donor to the rescue, or not…
    On my journey to purchase a replacement tap, I called in on an hardware/peripheral repairer/supplier to see if they had any old machines to get rid of. This was fortunate as they were more than happy for me to scrounge in their basement trailer for whatever I wanted. Unfortunately the trailer was emptied a day ago and the oly thing they had was this old P2 box. I thought I was on a winner as the IO panel and slot cutouts were appropriate, but the recess of the back panel connector area was very deep, more so than the Compaq. From what I can gather from the three relatively current boxes I have here, the recess around the 15mm mark and not 20mm. I don't want to position my motherboard for drilling standoffs and the like using something does not conform to the current standards. Besides, I really hate those backpanels where the connectors are so deep you can't a finger in there to release to the ethernet cable or unscrew your VGA card adaptor.
    Note, it is a Zenith proprietary box with a very unusual opening style. It took me a few minutes to work out how to open the bugger up before I realised it was inappropriate.

    Right side panel

    Well, disheartened but still motivated, I started on the right side panel. It didn't take me long to cut the two pieces required; one 75mm high fixed bottom panel and one flip-down access panel. I also managed to cut up the bracing, including mitred corners, before it got too late for the noisy stuff. I did consider doing the drilling and screwing as this is relatively quiet, but hunger got the better of me. (I think tonight is a good night to settle on the lounge for Star Trek or some such dreamy programme.

    Well, that's it for today. Catch you next time.


  6. #6

    WORKLOG: 'IOU' – GAM's QubeLok Cube Case with IO connector

    Workshop - Day 5: Half baked, half waste
    Note, click on images for higher res. versions.
    Another day, another frustration. You've read it before and you'll read it again… once again, I couldn't resist the beckoning workshop today… I'm already sorry. :|

    Today started with grand plans of completing the right side door/panel, the left side door/panel--sans motherboard tray--and anything else that I could get could done. Alas, this wasn't the case.
    A new hinge that wasn't to be…
    I spent some time figuring out this simpler and more easily removed hinge arrangement for the right side door which accesses the hard drives and optical drives area. No sooner had I printed up the dimensioned drawing ready for fabrication that I realised this wasn't suitable for the design of the case and door. Why? Because the door overlaps the frame work and this method relies on the lower flap of the door rotating inwards. I would have had to cut the corner sections (>25.4mm square) out to allow the bulk of the centre section to pass inwards. This wasn't on. No doubt I will have to use the same or similar method as the left side (motherboard tray) version.
    I really want this door to be easily removed without undoing any screws or such. I will give it some more thought soon enough.

    Note, the patterned blocks represent insulating foam strips (3mm thick that will compress to 1-1.5mm) that I have in mind for lining most or all panel-to-frame matings, or at least the right and left side panels/doors.
    Intricate cuts on a table saw

    So… on to the next task…
    This piece of sheet aluminium sits vertically, above the folding left side panel, between the top of the case and where the horizontal motherboard IO panel meets the middle of the case.
    I didn't want to add another piece of angle to the framework as the whole point of the framework is to provide the outside structure and some internal fixing points. So, I extended this piece all the way to the front of the case so I could fix the forward end to the centre vertical tube. Hence, I needed to cut out a section that size of the tube (1" square). I did this on the table saw, at least to begin with.
    I started by setting the fence to 25.5mm and cutting gently into the end of the section, stopping well short of the required distance as the underside cut is longer than the top side cut. Next, I used a drop cut method whereby the workpiece is positioned forward of the blade and VERY carefully lowered down over the blade at the required distance. WARNING: this method is quite dangerous and at will ruin the workpiece if not done with care. Typically, the spinning saw blade wants to throw the workpiece back at you, however you're usually behind it holding it firm and pushing it towards the blade. If you in front and letting the workpiece go backward over the spinning blade, the blade has immensely better chances of catching the material and flinging it for miles… or your mating standing by . Be warned.

    Having sawn two lines close to my final destination, I finished it off with a small, unbowed, hacksaw. The internal corner was cleaned up carefully with a file.

    Visible left side panels
    After marking, drilling and threading, I temporarily fastened the panel in place using a few button head screws, which will be replaced with counter sunk screws once I happy with everything (and I've replaced or sharpened my countersinking bit.

    Next, I cut out the smaller panels for the front-left 5.25" bay top cover. As I hadn't actually designed and drawn these yet I stumbled a little bit working out how I actually wanted these panels to fit together. As it happens, I made some errors and are still not sure that I'm happy with the current arrangement. Therefore, I haven't fastened these yet other than with some masking tape. Initially I stuck some pieces of double-sided tape to the edges of the panels or pieces of angle, which will help form the corners and provide some strength, but when I started to put them together I realised that I had it all wrong… Is this piece above that one, or is that one inside this one, etc etc.
    Before doing any of this, I laid the pieces out and marked them with notes on orientation before going any further.

    You can see where I've adhered one side of the double-sided tape ready for assembly.

    Note, I'm still deliberating whether to screw these together or glue them together. Although I am inclined to glue them, given the nature of this design feature, however I feel I should probably keep with the general design of the rest of the case and simply screw them together with countersunk cap screws as I've down with everything else so far. Any opinion? Please let me know. From the following photos you get the idea of what it would like withough screws (despite them being covered in masking tape).
    Also, that piece of angle that is visible on the far right, behind and top of the door panel, must either be cut off or some other changes made to allow the door to open and close without obstruction. Another possibility is to mount the whole 5.25" bay cover section to the door and have it open with the door. Thoughts people? I can't decide at the moment.

    Incomplete right side door
    Earlier in the day, before I gave up on it due to the useless hinge effort, I had started on the stiffening braces/framework for the right side door. I didn't complete this because I knew that changes would be required depending on the hinge arrangement I adopt.

    The sum of today's lacklustre effort
    Note the back panel sitting there for illustrative purposes. Anyone in Sydney that has a spare one of these in aluminium or thin steel that want to donate. I refuse to use this one as it is very deep and made from very thick, heavy steel. The depth is the main concern though as I don't want to position and finalise the motherboard mounting holes unless I have a standard, current, backpanel. This one is 17mm from outside edge to outside of recess. I am looking for something the more common 15mm or less, constructed from material that is <1mm thick (lighter).

    Thanks for viewing, and if you have any suggestions for the issues/uncertainties that I've raised, please feel free to comment. For that matter, if you’ve any questions on any aspect of the design that I haven't covered off adequately, don't hesitate to ask.


  7. #7

    WORKLOG: 'IOU' – GAM's QubeLok Cube Case with IO connector

    Workshop - Day 6: When in doubt, jig out
    Note, click on images for higher res. versions.
    Well today was not a 'mod' day as such as I was too busy with other duties and responsibilities, unfortunately. I did, however, manage to get a couple of hours in in this evening.

    Firstly, I was anticipating the arrival of my '3-32 UNC tap', required to thread the motherboard standoff mounts on the motherboard tray rails, but this didn't happen. Three people called in sick at the supplier, one of them was the guy I ordered from, but naturally, today's supplies were not processed by the company. Bummer for me. Nevertheless, I headed out to the workshop later this afternoon to see what I could accomplish in the short time. Looking at everything and still having the same doubts and uncertainties, I was quite inunspired. When in doubt, build a jig.

    Repetitive panel-fixing hole jig

    Due to my lack of inspiration (and available time) I thought I'd focus on something very simple, lest I stuff something up. The decision was; fix the lower 'fixed' side panels--the panels beneath the opening side panels. Realising that I had a number of these holes to drill for the project (not just these panels), and the fact that I wanted to ensure the same spacing from the perimeter of the case, I decided to make a repetitive drilling jig that I could reuse for all the panel fixing holes. Instead of just catering to one common distance of 8mm in from the outside edge, I decided to incorporate a few other common distances; 6mm, 8mm, 10mm, 12mm and 15mm.
    First, I marked and centre-punched the various distances from the inside face of a piece of aluminium angle which will be the fence that I clamp to the drill press table. I spaced these out so that each pilot hole wouldn't overlap the other. Unfortunately I didn't think this out enough and are inclined to make another one (another day) as the holes are neither centre on the fence nor I can even use the outer ones as the fence is sufficiently off centre for the little clamps I use to make contact with the fence and the bench at the same time.

    Anyway, note the centre-punched locations for drilling. I drilled these out using a 2.5mm drill bit which is the size I'm using to tap the 3mm threads--standard screw diameter for all exterior panels on this case.

    Once I had my prespaced holes, I located the fence for an 8mm inside depth by plunging the unpowered 2.5mm bit into the '8mm' pilot hole on the jig. The jig was then clamped in place appropriately. Note, be careful and gentle as the fence can move very easily while you're trying to get the clamps on it and fastened down.
    Note also, I've realised that I neglected to picture the vertical guide lines that I 'scribed' into the aluminium, using the butt of my square aligned with each of the centre punch locations, similarly to the 'masking tape' guide I've documented earlier. I also had to attach a piece of a alu. channel to the back of the jig to provide something for the clamps to clamp onto!
    I use the vertical scribed guide lines on the fence to align the drill centre lines of my workpiece and then the drill bit itself, as it approaches the workpiece, to be sure I'm going to drill the right spot.
    Once the fence is set, all holes are drilled very quickly, accurately and consistently in very little time. The advantange being that I am only marking one set of guide lines and I don't need to do any centre-punching. This is also ensuring that I have a consistent distance-to-edge for all panels, be it 8mm, 8.25mm, 8.13mm etc--they're consistent!

    Fixing the lower side panels
    Using a couple of spare pieces of panel material under the frame to give me the right overlap depth for the vertical panel and some clamps to hold it in place, I drilled through the pilot holes into the frame using the same 2.5mm drill bit. Once all the pilot holes were drilled in the frame, I removed the clamps and panel and theaded the holes using the 3mm tap. Note that all thread have been tapped using the tap in my small cordless drill at relatively slow speeds. Note, ALL THREADS have been done this way and every screw that you've seen so far is tapped into the frame or adjacent panel--not nuts! Once the threads were done, the panel holes were redrilled using a 3mm bit to ensure the screw clears the panel and fixes it firmly to the frame behind.

    Both left and right lower side panels were done and fixed using button head hex screws temporarily until I can get another countersinking bit that is suitable for these small diameters. Once I have the required countersinking bit, I will countersink all exterior fixing holes and repalce the screws with countersunk versions as I have used on the motherboard tray panel.
    Also note in these shots, the little strip of aluminium that I've brushed, literally in a few seconds, using 240grit wet and dry paper as an experiment in the 'brushed' finished that I'm contemplating for this case.

    Homemade reamer

    Sick of cleaning up the breakout on the reverse side of each drill hole, I decided to convert an old drill bit into a reamer for these small holes by very simply rolling some masking tape around the bit for comfort. It works perfectly by twirling the bit in your fingers over the hole, see… Also note the countersinking bit that doesn't have quite enough 'cutting edge' for these small holes.

    That's all for now. Later, modders!


  8. #8

    WORKLOG: 'IOU' – GAM's QubeLok Cube Case with IO connector

    Workshop - Day 7: Spendin' is better than usin'
    Note, click on images for higher res. versions.
    Same story as yesterday. Busy in the morning, a few hours in the arfternoon for modding. The good thing about today is that I finally got a payment in… which means I could buy some more materials and the odd tool that I need.

    AU$40 worth of assorted aluminium offcuts care of my local aluminium fabricator. Amongst the sheet stock are 1.2-2mm sheets (mostly 1.2mm and 1.6mm) and two types of perforated sheet (Clean, unpainted MeshX type stuff, and the larger holed variety that I scored when I went there last week). Amongst the extrusions are various angle, channel, flat bar and unequal angle. For this mod, I'll only be using about 10% of what you see, if that, as my original purchases has/will cover off most of the remaining material requirements. The rest will be added to the collection of stock that I'm building up for future mods and cases. The extrusions include a 6m length of 12mm x 12mm x 1.6mm angle, which I'm using a lot in this mod.


    Two tool suppliers later and I have these goodies to add to the workshop. Most importantly, 3-32 UNC tap for threading the holes for motherboard standoffs and a replacement 210mm (8 ¼&#34 saw blade for cutting aluminium on my compound saw. Also, there are some replacement hacksaw blades (18, 24 & 32 tooth varieties), 'Speed Grip' conversion for my disc sander with various grit change over sandpapers, new 45º 4-10mm countersinking bit and a new ¼" straight/end cut router bit for routing acrylic.

    Details of the new saw blade and the inevitable result of cutting too much aluminium (the final straw was some 6.5mm tubing) using a strictly 'wood' saw blade on my compound saw. It may be hard to tell from the picture, but I've marked the three 'missing' tungsten tips with arrows--the entire 'tips' are gone, not just chipped. There are a couple of other chipped-tooth casualties there too.

    Test driving the new 6-32 UNC tap. Pictured, is a standard harddrive/PCI card thumbscrew threaded into a piece of scrap aluminium.

    On to the modding…

    Auxilliary 5.25" top cover… continued.

    Removing and cleaning up the double-sided tape and adhesive that I decided to abandon. Note the 'Big Kev's Goo Remover'! This is top stuff. Drip a little on a rag or paper towel and rub for a while and away it goes. Unfortunately, I forgot to take the 'end result' picture. Trust me, every bit of adhesive and every other bit of foreign matter was removed after a minute or two of continious rubbing.

    This relatively minor part of the case is turning out to be a real PITA. It's easy to go through the motions in your head, but once you get to the 'doing' side of things there are numerous set backs, surprises and difficulties. Anyway, some pictures…

    Some points on the previous pics (I can't help myself); as I hadn't properly planned or thought through the panel fixings and fixing screw locations, I messed around a lot considering where the screws and braces should go and where I 'wanted' them to go. In the end I am not that happy with this arrangement, but I feel I have to progress, knowing that it will be much easier to recreate certain pieces later if I decide that just can't live with them.

    Test driving the new counter sinking bit
    I couldn't help myself, so I decided to countersink two of the several fixing screws in this cover and use the countersunk 'flat hex screws' planned for all exterior panel screws.

    End result of today's few hours in the workshop

    Note that some of the edges have intentionally been left overlapping to allow for sanding/filing down flush, later, when everything is constructed and I have final panels and edges in place, to work to.

    Not much to show for it I know, but believe me this was a real nuisance to put together.

    At the end of the day, 'shopping' was the best bit.

    Oh, on another note, I am having difficulties sourcing a back panel for the motherboard. The retailer/repairer that I had mentioned before gave me the brush off today. Not real happy about that (I don't think I'll be buying my next printer from them afterall ). I'll be trying a couple of other PC suppliers/repairers in the area early next week, but if that fails I will probably resort to buying a whole case (cheapo' generic) and ruining it just for the back panel. So, if you're in Sydney, on the Northern Beaches and have a spare 'donor' case lying around, please let me know.


  9. #9

    WORKLOG: 'IOU' – GAM's QubeLok Cube Case with IO connector

    Workshop - Day 8: A Donor Today Lets The Modder Play
    Note, click on images for higher res. versions.
    Well, I managed to find a donor back panel today from my new contact, Jodie, at UYC ComputersUYC Computers-check them out. Nice guys too.

    So, with this donated 'standard ATX' back panel, I can now proceed with the side panel/motherboard tray. Yippee! Unfortunately, after much aggrevation, missing items, measuring, checking, remeasuring, rechecking, again and again, I still ended up with problems and quite deflated. Considering my options while I fell off sleep, I have an action plan for overcoming these problems which involves remaking a few pieces of the side panel that I stuffed up yesterday.

    Anyway, as I have to get out into the workshop, I'll keep the verbosity of those brief. (I can hear the applaud now )
    Motherboard Tray, Alignment, Marking, Test Fitting – Compaq DeskPro

    Before I secured the new donor backpanel, I played around with the existing 1.2mm steel version to obtain the various dimensions for clearances, spacing and the like.

    As I wasn't prepared to commit to motherboard standoff holes yet, I made some 6.5mm temporary spacers to elevate the motherboard to the correct height from the rails.

    'Square' clamped in place to hold the panels square to each other.

    Not convinced about using the 'Compaq' back panel, I frigged around for a while extracting the 'Zenith' P2 version (everything had to be removed from inside the case to get the right side panel off and expose the rivets!!!). As it happens, this one is useless as the IO shield cutout is both a different size and slightly different location… but of course, you all knew this

    Enter… new donor case/back panel!

    Motherboard Tray, Alignment, Marking, Test Fitting – Take 2 (Donor case)
    Despite having the ATX Form Factor diagrams and dimensions, I wasn't convinced and needed to reassure myself of the exact location of mother standoff holes relative to my opening side and top panels. Also, I noted (as can barely be seen in the cross section pics) where I need to remove material from my horizontal rail for the PCI card back plate. You see where it is contacting and elevating the motherboard beyond the temporary standoffs.

    Well, after much test fitting, measuring, marking and the like, I commited to some holes in the horizontal and vertical rails for the motherboard standoffs. While I had these removed, I cleaned up the side panel holes and re-did some of the countersinking… with less than perfect results. (I see a new side panel being made here. )

    I fixed the top of the vertical rails in place and used a square to align the rails square to the bottom horizontal edge of the side panel before drilling and screwing down the bottom ends.

    To make space for the PCI card back plates, I used a drop cut method on the table saw, see photos…

    Note the masking tape in indicating my start and stop points for extending the cut as I didn't want to increase the depth of the blade as this would penetrate the opposing face of the channel. The second photo shows the final cut.

    After some filing and sanding… one neat cutout.
    /288x/CIMG5890.25a.jpg" alt="" />
    Back to the motherboard tray rails.

    Plaintiff dog, "Ellie", letting me know it's dinner time. You know she's really concerned when she just stands right next to me.

    After careful marking, I committed to the standoff holes, threaded them and inserted the standoffs I had. Unfortunately I hadn't noticed that I only had 3 standard ones and the other three were extended versions. This was real pain as I had to relocate them for each vertical rail to ensure I drilled the correct location for the bottom rail-fixing points.

    Not obvious in the photos, but I had vertical guide lines on the bottom horizontal indicating where the verticals should be. As I was fixing them with the MB in place I noticed that these were not lining up!!! It hadn't occurred to me that somethnig else was wrong! (may have something to do with it being the end of the day and the several beers under the belt )

    The reason for the F'up! When I re affixed the top horizontal rail and top panel, I hadn't ensured the location. The result, as it happens, was that the top panel was jutting out because the rail wasn't square and not located in its original position! Damn. Screwed everything I did yesterday afternoon.

    Today, I fix what I f'ed!


  10. #10

    WORKLOG: 'IOU' – GAM's QubeLok Cube Case with IO connector

    Workshop - Day 9: If I could choose my mother again, I'd get one off the shelf
    Note, click on images for higher res. versions.
    Today was relatively productive despite errors, time wasting and remaking…

    Yesterday's problems, today's duties… correcting the MB tray's horizontal and vertical rails
    As you can see, I managed to correct the misalignment problem, but not without sacrifices. The initial misalignment actually came about because of my re-countersinking endeavours which actually opened up the facing holes off centre. Unfortunately I can't think of a way to correct this problem without remaking the panel, which will be undertaken down the track. For moment though, I've replaced the countersunk screws with button head versions and washers that sit on the surface of the panel. Ugly but works.

    Here you can see the other oversight I made the day before. The left and centre vertical rails were 12mm x 12mm angle aluminium and used to be fixed in the holes that you can which are not being used. The 12x12 rails were fixed in those holes and overlapped the PCI back plate cutout. So today, I remade the vertical rails using unequal angle (24mm x 12mm iirc). The fixing holes are now located 12mm to the outside of the original holes.
    The next picture shows the new rails fixed in their respective locations. The last pic shows the cutouts in rails after some filing and general cleaning up.

    Prefitting the motherboard to determine the motherboard standoff locations on the rails. This task was a real pain. The combination of the rail fixing locations top and bottom and then the motherboard standoff locations proved to be quite problematic and fraught with easy errors. Next time, I'll either be much surer of my dimensions beforehand, or more likely, I'll use an existing motherboard tray/back panel combination as is typically done and suggested by colleagues.

    Back panel IO cutout… er, Top panel

    After establishing all the motherboard standoffs and fixing the motherboard using the three (yes, only three) standoffs that I had, I marked out the IO back panel on the cases top panel, in preparation for cutting out the required sections.

    I fastidiously marked out and then drilled each of the corners of the section to 10mm using a unibit--highly recommended for any sheet work, particularly acrylic!
    Corners holes done. Nice clean entry hole using the unibit and one of the reasons why I recommend them so strongly.

    Internal straight cuts on a table saw
    Rather than using the bench mounted jigsaw for this job (it wasn't set for one reason against it), I decided to cut all the internal straight edges using the table saw and the 'drop' or plunge method. This is tricky business which can work beautifully if you're on the ball. Unfortunately did make two mistakes… see if you can identify them and tell my why it happened?
    The first cut was a test cut a couple of millimetres inside the final cut. No problems, the method is going to work just fine, as expected, and I have a 'fence calibration' adjustment established. The adjusted fence callibration is required because I must use a piece of angle against the fence to prevent the thin aluminium sheet from slipping beneath the fence of the Triton saw. If you own one, you know that their sits about a millimetre above the surface of the table.

    All straight cuts done! Can you see the errors? Know why?

    The final bits holding the section in place were cut through using a hacksaw… of sorts.

    Beer check. Actually, I didn't drink all these today, but I did clear them out yesterday at some point. The reason for the only injury so far… cut to the finger trying to catch a slipping bottle… the bottles in my arms connected and smashed. My finger the caught the broken bottle.

    Back panel (top panel) IO cutout
    Back to more serious matters… after some careful filing and a little sanding, I have a neat IO panel cutout. Both panels have been clamped and braced to ensure my hole drilling marks are accurate. Using a fine pencil, I outlined the drill holes by following the inside of the existing IO panel fixing holes.

    Once that was done, both panels were disassembled from each other and the motherboard tray entirely. After punching the hole centres, I drilled them out a 3mm drill bit on the drill press.

    One top IO panel… Voila!

    Before putting it all back together again, I decided to do a quick test and sample of the 'brushed' finished I am thinking I am going with. Both panels are currenlty fixed together using black button head capped screws and black washers and nuts on the reverse side. I am considering rivets, but I really want to keep with the 'black and aluminium' finish.


    Closed position in case

    Note, sorry I forgot to take the shot of the motherboard tray/side panel its 'open', accessible position… tomorrow.

    Final beer count, plus the one I was half way through

    If anyone was unsure what I was doing with the motherboard and IO panel, hopefully it is all clear now--IO panel Up!


Page 1 of 7 123 ... LastLast

Thread Information

Users Browsing this Thread

There are currently 1 users browsing this thread. (0 members and 1 guests)

Similar Threads

  1. Rosewill Legacy V4 Aluminum Mini-ITX Cube Computer Case
    By Americanfreak in forum Modders-Inc Hardware Reviews
    Replies: 0
    Last Post: 03-12-2014, 05:23 AM
  2. Vic - XcaliburFX Worklog ... BlewByU Hot Rod Case Mod
    By XcaliburFX in forum Mod Off - Staff VS Staff -2010 Sponsored by Thermaltake
    Replies: 74
    Last Post: 10-12-2010, 07:42 PM
  3. Mazzola's Ultimate Cube Case
    By mdm08 in forum Modding Work logs
    Replies: 6
    Last Post: 06-29-2008, 10:50 PM
  4. New lian-li cube case
    By brightshininglights in forum Talk Freely
    Replies: 0
    Last Post: 10-22-2007, 08:58 PM
  5. Cube Case - 56k death
    By faceless105 in forum Case Mod Gallery
    Replies: 7
    Last Post: 07-13-2006, 12:04 PM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts