Dalek Spaceship - Remembrance of the Daleks

[Davros Skaro]

I worked in a place once & we had to change lights on a control panel when they blew, so they sent me some connectors that when I cut the old light off, I attached the connector, (no soldering) then the new light plugged into it. I looked on ebay & found these that might help you to change the fans over, (again no soldering). Have a look & see what you think, might save you some hassles.

https://www.ebay.com.au/itm/2P-Spring-Connector-Wire-Clamp-Terminal-Block-No-Welding-For-LED-Strip-E9I7/323287597924?epid=2070500287&hash=item4b45714364:g:2RkAAOSwwAdbFT6k

Hope this helps,

Chris.

[d33j r093r5]

... alright, let's see if I can remember everything that's happened   ...

... printed a few more of those roundel panels before going to bed. Set some more up to print, and also retried the same 1D Bottom Front Module on Printer 2; I expected it to be fine, and that the failure was just unlucky coincidence...

...

...

...

... woke up this morning to this:





       

... ok, that's essentially what it started doing the previous day when I stopped it, only I was there to see it happen and stopped it before it got to this stage. This is what I was talking about; somehow, the printer is losing its X positioning at a particular point in the print, but it keeps going nevertheless. I told you I heard a loud bang before it started doing it. This would indicate that the printer hit something, although there's nothing I could see... anyway, just so there isn't any doubt...











... these prints have stuffed up in exactly the same spot. Now that's too much of a coincidence for it to just be an accident... so, either there's something very wrong with my printer when it gets to a particular position, or there's something very wrong with the G-Code file that I'm uploading to it... decided to try an experiment. I started printing the Top Rear of Module 1D on that printer. It's the same as all the pieces essentially printed up until now. If it's going to fail somewhere around that spot, I'm going to be watching it to see what is actually happening at that moment. If not, then I've made a successful print! Now, I just have to wait about 7 hours...

I also removed the last batch of Roundel Panels from Printer 1...





... and started the first lot of Door Panels...





... which, as you can see came out pretty well... except NOW, the extruder on Printer 1 jammed... *!SIGH!* ... and I did my level best to fix it. It didn't seem like a significant jam. Unfortunately (there's that word again) despite everything, the interior of the heater was completely clagged! I mean there's just a hot melted mess on the inside that I can't get out. Not without some serious disassembly an a few hours of trying to clear it...     ... not in the mood for that, so I took the easy way out; I pilfered the nozzle, made a new feed tube, and stuck it on the left side. I mean, it's  dual extrusion printer (they both are, technically speaking), and as far as I could tell the left side was in good nick, possibly hadn't yet been used. Besides, the fan on that side works properly!    I think that was probably a good decision overall, because it started extruding no problem, and produced this...





... ok, it's probably difficult to tell those last 2 images apart; here's a better view of that latest batch...








... please note the apparent lack of ANY artefacts whatsoever! It would seem that everything on the right side was currently failing and a change was necessary. Thinking back a few years, I swear I already had this experience before, a couple of times actually. It would seem that my learning curves are not so steep...  


... getting back to Printer 2, as the 6.5 hour mark approached, and it crept closer to 90 odd percent, I was getting worried that this would ultimately be another failed print. 90% passed, then 92%, then 93%, etc... it was well and truly passed the height at which the previous couple of prints failed. I say height, but in truth the print head height doesn't change; the bed lowers by 0.25mm every layer. So I started to think ok, it doesn't look like it's the printer, could there really be such a serious error in the code? That really doesn't seem likely. I started to look around the interior of the printer to see what it might have hit, when I suddenly realised just how far forward the print head was as it was approaching the end. In fact, in terms of geometry, the greatest distance it travels is forward as that front angle gets further and further away from the floor of the print...

... bear in mind that the forward and rear sections of all of the Module 1 units are different lengths. The geometry was deliberately made like that to allow for the holes in the floor without crossing the break-line of the sections. The Front sections of Module D are longer than the Rear sections, and generally speaking, Printer 2 had been printing all Rear sections up until now. In fact, the Front sections have an extra 6mm on the Rear sections. So I went and I measured the failed part of the failed print (where it suddenly looks like a bad transporter accident). Sure enough, the length of the bad bit of the print was 6mm. So, I know what's happened (I think!)...





... this is the finished good part from this evening. If I push the print head back to where it would have been on that final run on the front...





... you can see it's the furthest forward it would be for this print. And for this print it was fine because, overall, it's shorter than the other one...








... here, it's much easier to see how far forward it is on the printer, but still ok. Nevertheless, it's still verrrrrrry close to the front of the printer. Almost no room left. Now...








... this is the first failed print. It's a longer piece and, even though it's not finished, you can already see how much further forward it's sitting. Actually, I had to place this in there by hand as a guesstimate of it's position, so it might be a tad exaggerated. Nevertheless, it would still have been at least 3mm further forward than the successful print. At this stage, more then likely, it collided with one of the hinge panels at the front of the printer as it traversed laterally in the X axis, effectively changing its relative position compared to its absolute position. Hence why we ended up with an offset. And, because it had now effectively found a way around colliding with the hinge again (ie colliding with it the first time until it "cleared" it), it continued on it's merry way and finished the print! There wouldn't have been any such collision with the other prints on that printer because none of them were that far forward. You can actually see just how far it would stick over the edge if I put in that finished failed print...











... I understand that people pay good money for hairstyles like that.     ... actually, that it finished the print with all that structure intact hanging over mid-air is fairly impressive to me. I think I'm going to keep this "failure" for posterity...

... anyway, here's the next big question. Why, if half the prints were all that length, did Printer 1 not run into the same problem when it was printing them? Well, in spite of the fact that they're the same printer, there are a few differences. For one, they're running different firmware (Printer 1 is more up-to-date). As a result, they do respond slightly differently to the same G and M commands in the code. And the print heads on them are slightly different build versions. None of these are really significant, but enough for there to be a few little things here and there. For one thing (and this is the important bit), there X and Y offsets from the origin to nozzle positions on either machine are slightly different. So, even though in the plater / print software the positions are the same, the different printers offset them slightly differently when they go to print. Verrrry significantly (and I checked this at the start of the next prints), the difference between those offsets (in the Y axis) is between 5 and 6mm. So Printer 2 starts about 6mm further forward than Printer 1, gets to about 92% of the way through a print on the longer module sections, hits a particular point at the front and gets it stuck as it's travelling in X (it doesn't realised it's stopped), and changes direction when it reaches that point in the code and continues on its merry way, now offset by however long it was stopped for, and however much distance that covered (about 26mm)...

... as with most problems with tech, it had nothing whatsoever to do with the tech itself. Not a printer problem, not a file problem; in IT we call this a PEBCAT (Problem Exists Between Chair And Terminal)...       ... hopefully lesson learned. Although, I'm pretty sure that THIS has also come up at some point in the past, and I'm still making the same mistake. My learning curve is starting to look like a very flat line...  

... anyway, that last successful print...





... perfect! Nothing to clean up at all! I'm really relieved, this printer was printing so nicely, I was quite worried that something serious had gone wrong with it, and that was going to be the end of it... at least for a while anyway...





... looks happy there. Getting to the most recent developments, given that Printer 1 is now behaving itself in terms of print quality, and the Printer 2 problem has been solved, and since I still have 2 module sections left to print, I've assigned one piece to each printer. Printer 1 I'm hopeful that I'm not going to see any blobbing or corner curling happening as the change of print-head appears to have fixed that, and Printer 2 shouldn't have any difficulty with it's remaining piece, as I've shifted the start position for the piece right back on the printer bed. Both those prints should be ready for inspection in the morning... And it's nice to be back on track...

... as to the current state of the model...













D.

[d33j r093r5]

... apologies Chris, didn't see your last comment until just now. Those plugs actually would be an ideal solution, and the design of them means that one could take care of the 2 fans on one printer, yet still be modular enough for me to swap em out one at a time if necessary. Awesome! Thanks very much for that.

Cheers,

D.

[Davros Skaro]

What a dilemma with the printer hitting the front of the case, that really put it out of wack! Interesting how it printed in mid-air too!  Glad you solved it. Those panels are looking good too, nice job! 

... apologies Chris, didn't see your last comment until just now. Those plugs actually would be an ideal solution, and the design of them means that one could take care of the 2 fans on one printer, yet still be modular enough for me to swap em out one at a time if necessary. Awesome! Thanks very much for that.

Cheers,

D.

Your welcome, glad I could help. 

Chris.

[d33j r093r5]

Ok, turns out I HAVEN'T actually solved it; woke up this morning just in time to witness the saaaaame thing happen at 90% completion; same spot too. There seems to be an unusual accumulation of problems right at that moment. What's really strange is that the Bottom Front section of Module 1A is absolutely IDENTICAL to same section of Module 1D, and Printer 2 printed that just fine (I went back and checked). I had an option this morning; either get Printer 2 to print that piece of Module 1A again and see what happens, or just print it on Printer 1. Given that Printer 1 seems to be behaving itself, thought I'd give that a go. This piece DID fail on Printer 1 before, but that was the result of a tangle; that's unlikely to happen again. I was tempted to continue printing panels on Printer 2, but it would have meant removing the tape, putting on a glass platform, re-levelling the bed, etc, etc. I wasn't ready to do that given the rest of the printing that needs to be done. Instead, I've started printing the Rocket Supports; nice, small, easy, quick. Am considering printing the Rocket Flues out after that as well. That should be enough time for the final Module piece to print on Printer 1. Can then finish of the remaining panels...

As to Printer 1 now printing nicely, seems that was somewhat premature as well. While not nearly as much as before, there were still quite a lot of artefacts on the Top Front section of Module 1D (will post photo later). Aside from that though, no problems...

Cheers,

D.

[galacticprobe]

What's really strange is that the Bottom Front section of Module 1A is absolutely IDENTICAL to same section of Module 1D, and Printer 2 printed that just fine (I went back and checked).

Well, before I saw what all was going to be involved with getting Printer 2 back in operation, I was going to ask if you'd considered printing another Module 1D bottom front section since that printed correctly. I don't know much about 3D printing and the files associated with it, but I do know electronics, and thinking along those lines makes me think there could be a glitch in Module 1A's bottom front section's files if they were drawn up separately from Module 1D's bottom front section's. So if the parts are identical, and Module 1D's bottom front section printed fine, printing another Module 1D bottom front could be something worth trying if all else fails.

I hope some of this is helpful.

Dino.

[d33j r093r5]

Exactly! But, the other way round ; 1A printed correctly, 1D keeps failing. In any case, I'm currently printing that final 1D section on Printer 1. I don't envisage any problems. I'll redress the 1D problem later if it is actually a print file error. I'll know it soon enough if the same print fails in the same way on Printer 1... that's when I'll go to Plan B, which is print Module A as Module D on Printer 2. Everyone still keeping up...?       

Cheers,

D.

[galacticprobe]

Exactly! But, the other way round

Blah! Leave it to me to get things the wrong way round... even when I'm reading the bloody thing to make sure I get it straight! (Well, at least you knew what I was talking about... even if I didn't! )

Dino.

[Davros Skaro]

What if you changed the orientation of it 90deg. to what it's printing, if it is getting too close to the front then it may have a bit more clearance?

Not sure if this makes sense,what I'm trying to say , instead of the open section being to the back of the printer, it's to the side. not sure if it prints this way, but just a thought.

Chris.

[d33j r093r5]

all good Dino, I'm having trouble keeping it straight in my head as well. It is confusing as hell; I need to go back and re-read what I've written to make sure I'm not mixing things up and confusing things for myself, let alone everyone else!

I actually started setting up a print with exactly that orientation Chris, but then realised that it's actually wider overall than it is long! It would stick out even further at 90^o. So, I scrapped that in favour of the other printer, which is now at about 70% of the way through the final piece. After that, all those sections are done, so...

... as an aside, Printer 2 has finished all the Rocket Supports, and is about 20% of the way through the first 2 Rocket Flues (done in black )... I will post an update about it later... like, probably tomorrow night at this stage (well, night in my part of the world anyway). I have an exam tomorrow afternoon, and I'm kind of focused on that for the next 24 hours or so... I'll see y'all on the other side of it...

D.

[d33j r093r5]

Hi all. Update time. A fairly longish one too...

As I mentioned in a comment, I hadn't solved my failed print problem with Printer 2. Actually, I solved it about 25 minuted ago, watching a new print start on Printer 2   I was sort of right. Watching the new print start (which was for panels for the hull) the gantry traversed all the way to the right hand side of the bed to print on the edge, hit the limit switch () stopped it's traverse in the X axis until the code caught up, and then began traversing in Y, now offset from its original position. Fortunately this was right at the beginning (like the first line of just the warm-up skirt), so it made no difference whatsoever to the print. So I could have fixed the issue, not by moving the module piece backwards on the bed, but further left! Lesson learned. Hopefully...

Just for the hell of it, here's the failed prints from Printer 2...








... you can't beat consistency    

... anyway, ALL the main module sections are now finished and attached to the model. There was a little warping still from Printer 1. I get a sneaking suspicion that it might actually be the filament itself. It does seem to be oozing quite a lot, and a bit runnier than usual. It IS a slightly different batch of the same filament, a little bit older, but usually that makes it more brittle, not runnier... I imagine that using a slightly lower temperature setting might fix it. The reason I suspect the filament and not the equipment is that the other filaments aren't doing it. Anyway, I digress...





... still lots of artefacts and spider-webs...





... and a bit of warping... As I mentioned earlier as well, I began / completed printing the Rocket Supports on Printer 2...

















... which came out perfectly. Both times. Which is nice... and then moved onto the Rocket Flues (which I have named as Rocket Boosters on the PC. See, I can't even keep the names straight in my own head! )...








... they were still printing when I went to bed on Thursday. Woke up to them completed on Friday morning...














... these pieces are really aesthetically pleasing to me. I don't know why, but these parts in particular I find really satisfying to look at, and feel very proud of. There's just such a nice solid look to them, and a well finished feel. That the printer printed them perfectly as well helps...! Anyway, I needed to see how the finished module looked with those on, so... even though I should have been studying, I assembled them...








     

... should maybe watch that ego of mine a bit there...

... I did take other photos during the piece finishing / assembling stage, but I'm saving the finished model pics for last... I did finish those Rocket Flues and Boosters, and a few more panels as well (Printer 1 just finished another batch of Door Panels about 5 minutes ago, and Printer 2 is currently working on the first of the... hmmm... I don't know what to call them really. On the computer I just called them Dalek Panels, because they appear to have the outline of Dalek on them if you were looking at from the bottom up...


http://tardisbuilders.com/index.php?action=dlattach;k=-170;topic=8789.0;attach=185475;image


... these ones. Lets just stick with Dalek Panels, it'll be easier).

... some other things which probably bear mentioning. You may have notices that each of the modules have little "studs" sticking out on the faces, and that the panels all have holes in the corners. I did this as a way of aligning the panels to the correct positions when needed, and also as a potential way of fixing them together. That might still work, even though they're a bit small, and I needed to re-drill the holes with a tiny, manual, hand held thing I got specifically for working with models / figures etc. Not much I can do about the studs though, they're little more than bumps in the surface, but they do serve the purpose of locating the panels. My latest thought is to affix the panels in place with a small piece of almost flattened blu-tack, and then use a a small blob of glue in the corners where the holes / studs are... I see no reason why that won't work...





... more Door Panels, and the second batch of Rocket Flues...





... ta-dahhhhhh!!! Happy with this...    


... as I mentioned, I got Printer 2 up and running with those "Dalek" Panels. For that, however, the bed needed to have the glass layer...








... I'd had this for a few months now, but hadn't attached it to Printer 2 yet. There wasn't any need, and it was doing such nice work. But, unless I only want to be printing with Printer 1 from here on out, it needs to go on. So...





... get rid of the blue tape, and clean the surface...





... attach the glass plate and give it a good wipe down, and a clean with with Acetone...





... and re-level the bed! Proof of that particular pudding will be in the print...





... Printer 2 printing on the glass bed! Nice when things go according to plan. In the lead-up to my exams things did seem to be going a bit haywire. I guess I shouldn't be trying to focus on extra-curricular activities when the pressure is on; that's probably why so many mistakes got made and so many mishaps happened...!

... anyway, current state of the model...




















... as I mentioned, more Door Panels just came off Printer 1, and Printer 2 is due to finish the first of the Dalek Panels shortly. I will post some pics up when they're done...

Cheers,

D.

[Rassilons Rod]

It looks amazing!!!

[Davros Skaro]

That looks amazing, well done! Love it & those Rocket Boosters, they turned out great. Instead of the Blu-tak, you could try the thin double-sided tape that cheap shops have, I've used it many a time & holds really well for a permanent fixing.

All the best with your exams too.

Chris.

[d33j r093r5]

thanks rassilonsrod, thanks Chris, much appreciated as always Sorry I've been a bit slack in posting the last few days, been busy with a lot of things...

I had my exam on Friday afternoon. I think it went well. Just have to wait a couple of weeks now to see whether I actually do know what I'm talking about, or whether I'm just deluding myself...

I've actually never used double-sided tape for anything before, although I understand it's very useful. Haven't been out to pick any up though, so in the meantime I'm just sticking with the blu-tack. I may not actually end up gluing those panels down; I'm not sure just yet. I may swap how they're held in place though from the blu-tack to the double-sided tape; I think it will actually sit and hold a lot better. Thanks again for the suggestion Chris.

Anyway, I have a lot to update, so I'd better get on with it...

Cheers,

D.

[d33j r093r5]

Quite a lot has happened since I last posted. Apologies for the long delay between posts as well; been a hectic week. Anyway, going back to last Saturday...








... this was the quality of the "Dalek" Panels Printer 2 gave me. Over the moon by this; these panels, while not the most intricate DO have some fine detail and curved surfaces. They came out splendidly! Of course, kinda figured they would, Printer 2 has been keeping up its end for a while now. Also, compare the Dalek Panels to the Door Panels done on Printer 1 (bear in mind, these have already been cleaned up). They're a lot smoother and cleaner, and required no clean up. I don't know if you can really see from this photo (it's hard to tell unless you're looking for it, but using your eyes without the camera filter it's easier to see), but there is a difference in tone in the filament! I had assumed that this colour was consistent across the board given the general quality of the filament (generally very good) and the same product code. They were purchased at different times however, so it's possible. It might also explain the difference in quality between the prints /  why the filament on Printer 1 tends to ooze a lot more, and hence leave those artefacts. Anyway...





... here are 2 separate runs of those prints, Dalek Panels printed on 2, Door Panels printed on 1. Except, for the second batch of Door Panels, I reduced the extruder temperature by 5^oC (9^oF), from 220^oC to 215^oC (428^oF to 419^oF). You can see the difference this has made even from this distance. Here's a closer view. 220^oC:








... and 215^oC:








... the older batch seems to have a slightly lower transition temperature range, which is why it oozes at 220. Printer software have ooze compensation built into them if you need to use it, largely by retracting filament during rapid movements between prints, but it's not going to do much if you've got liquid filament in the hot-end! There was still a little drizzle, so for the next run I set it 210 to see if it improved it any more. There was no noticeable difference between that and when it's set to 215. If anything, the prints were slightly worse quality. So, for the remainder of the prints on Printer 1 using this particular filament, the print temperature was set to 215. Saved me a lot of clean up work. Only wish I'd thought about it sooner...  

... I was adding these panels to the ship more-or-less as I went. However, in order to have big reveals too soon, I'm holding off on major photos (mostly!) until the end of the post...    


I was with finished Door Panels by late Saturday / early Sunday, so I then started printing the next lot. I really have no idea what to call them; when I did the original model, I just called them Level Panels, so we'll stick with that. I forget what my logic was behind the name... anyway, Printer 2 was still working on Dalek Panels, so the Level Panels were assigned to Printer 1...











... temperature still set to 215, so clean-up work was at a minimum. The pictures above are pre-clean-up to give you an idea...

... I should point out at this point that, between prints, there's been a slightly lengthy and annoying process of re-loading filament into the extruder. For some reason, at the end of prints and after the axes re-home themselves, the stepper-motors continue to run for a bit, and the filament retracts. This is actually one of the reasons of r many of the jams. It means that everytime I want to print, I need to first remove the filament and then re-load it. The printers actually never used to behave like this, it was a fairly recent development. And both printers are doing it, even though the firmware is slightly different. So, the problem must be in the post-process / completion code that I'm giving it...

You're probably not that interested in G-Code (or, maybe you are *shrugs*), but here's the completion code as it stood:

M73 P100 ; end build progress
G162 X Y F2500; home X and Y axes
G0 Z150 ; Send Z axis to bottom of machine
M18; disable all stepper motors
M109 S0 T1; set bed temperature to 0
M104 S0 T1; set extruder temperature to 0
M70 P5; Display "We <3 Making Things!" message
M72 P1; Play Ta-Da song

... and here it is after I altered it:

M73 P100 ; end build progress
G0 Z150 ; Send Z axis to bottom of machine
M109 S0 T1; set bed temperature to 0
M104 S0 T1; set extruder temperature to 0
G162 X Y F2500; home X and Y axes
M18; disable all stepper motors
M70 P5; Display "We <3 Making Things!" message
M72 P1; Play Ta-Da song

... it still does all the same things. Here's the difference; in the first instance it's sending X & Y home first. This is setting the global speed (Feedrate) to 2500mm per minute (F2500). So when it runs the G0 command directly afterwards, which is a command to to send whatever axis is specified to an absolute coordinate (in this case Z to point 150, which is slightly above the bottom of the machine), it's running it at 2500. Z axis is a scew mechanism, so 2500 is REALLY slow for it. And because nothing else has been deactivated yet, the X/Y axes, heated parts, extruders are all still running. Until the Z axis reaches the bottom. Really slowly! By moving the lines to where I did, Z axis sends itself to the bottom of the machine first at a rapid traverse rate (like F30000 instead!), THEN the bed and hot end get turned off, then the X/Y homes itself at 2500, and then the steppers and extruders all turn off. Now, the machine is ready for a new print without having to unclog or unjam or reload all over again! Glad I solved that one, it was a pain in the proverbial

Actually, I don;t know how it ended up like that in the first place. I must have had to re-type it out at some point in the past due to making a mistake, and had obviously either typed it wrong, or moved lines about without realising it. Anyway, it's now doing what it was doing originally, and working again!      


... ok, minor reveal time   :











   








... an essentially completed Module 3...





... Imperial Dalek Bob is not impressed. "It's all disassembled again? This might just be me being nit-picky, but this seems like a step backwards..."

All the Dalek Panels were completed by this stage, which means the only panels left to do were the remainder of the Level Panels (which was most of them). So I assigned Printer 2 to help speed up getting them done...





... this is actually a really good comparison photo of the quality the 2 printers are producing. You can pick which one is which, right...?





... and the subsequent batch...








... a slightly mollified Imperial Dalek Bob...

... not a huuuuuge amount happened on Monday in terms of printing. I had a lot of things to take care of, yes, but also the last of the Level Panels were completed...





... and so other things needed to be setup to print. Printer 1 finished its final batch first, so it was assigned the first batch of the Bars that adorn the doors underneath the panels








... and Printer 2 was to be assigned the printing of the ACTUAL doors!     However, it didn't quite work out that way...  

Having thus far been a very cooperative and hard-working printer, Printer 2 obviously decided it was time for some time off. I needed to reload the white filament in order to print the doors (it would look weird if the module with opening doors was white, bit the doors for it were grey). And, wouldn't you know it, the printer started to make a "clunk, clunk" noise like the feeder was jammed. The filament was flowing, and the clunks only came once per revolution or so, but it meant that something in there was preventing the filament from flowing smoothly........

...

...

...

...

...

...

...

...

... I took that head apart 4 or 5 times and still couldn't seem to fix it, even though the assembly appeared clear of obstructions. If anything, it got worse! The filament stopped flowing altogether and the clunking was just constant! I had a lot of other things to take care of on Monday, so I just left it...

... by Tuesday I had a nice pile of bars...





... and no real way to attach them precisely! This might be the biggest flaw in my design. I can't print these bars directly onto the surfaces as they will just turn out a huge mess. And the way I set them up to line-up and afix hasn't really printed very well to be effective. I MAY be left with a large pile of Door Bars and no way to attach them. I haven't addressed this problem yet; I'm still thinking about it. I have since starting this project realise how I can completely re-design the ship and how the different parts can be printed and fit together, which will solve that problem (and a few others besides), but that isn't going to work here... anyway...








... the disassembled bits of the head assembly from Printer 2, which I returned to on Tuesday afternoon. I finally dissembled everything, including the hot-end and nozzle assembly to see if I could find the blockage. I didn't find it, but it was obviously there. I replaced the plastic feeder tube and cleared out all the extraneous filament, reassembled it all and started to feed the white filament through...





... which worked perfectly, so there was obviously something. I had started to worry that maybe the filament diameter was off (too large), or just REALLY bad filament, as the printer had worked fine up til then with the grey, and only started to misbehave with the white. I did check it, and the filament was in spec. Besides, I'd used it already without any problems. Anyway, I always feel a bit down and disheartened, and like I'm climbing an insurmountable mountain when problems like that arise, and SOOOOO much better when I fix them. I can now start printing the doors...








... which APPEARED to start well...





... eeewwwww, that's not good. I mean, it might still be fine as a print, but that's a messy couple of first layers! You know what, I know what this is; I completely disassembled and reassembled the head, and didn't re-level the bed! Of course I'm going to get this, the nozzle tip is now in a slightly different position. D'oh! Well, anyway, after the fist layer or so, the prints are all at the same level, so it might not make a difference, and the prints might be ok...? Spoke too soon of course, a few layers later, both prints had started to curl and warp off the bed. That's not going to work at all for the doors. I needed to cancel, level out the bed and start again...





... which turned out beautifully...








... and they line up well too...








... oooooohhhh, ok, this is getting exciting...!!!  


... the plan was always to glue these down. The doors hinges will actually see a bit of torque, as I'm expecting those lever bars to have a bit of twist in them. That's ok, even with a small amount of torsion, they should still perform as they've been designed to do. The problem was always going to be how that twist would affect the positioning of the doors and / or how the hinges would behave in that scenario. The hinges are metal, so they probably won't really be affected, but then all the force will be on whatever is holding them in place. So, if I AM gluing them, it needs to be a pretty strong glue. There are a few different types recommended for PLA, but only a few of them are also multi-purpose glues for use with other materials as well. In the end I settled on this...





... it's actually just a super-glue, nothing particularly special about it. I like Loctite products, they usually do what it says on the bottle, and this had a really nice, fine control application to it, which was kind of necessary. Also, it claimed that I wouldn't glue my fingers together with it. Always a bonus...    

At this point, there was really only one thing left to print: the lever bars that would control the door mechanism...














... which have printed beautifully. You'll notice that I had to print them with support material given that dog-leg turn in the design. Printing with support can be an iffy process. Ideally, you want a dual-head printer, that extrudes support material from the second head, and that material should either be PVA (Poly-Vinyl Alcohol) or HIPS (High Impact Polystyrene), both of which dissolve in water, which is why they're ideal. Unfortunately, dual-head printing is not an option for me, so support material is inevitably the same material and colour as the final piece being printed. That already makes it difficult, as what you end up with is print and support essentially as all one part!

With the print software that comes with my printers (Replicator G), support is difficult, nigh impossible to remove completely. It attempts to make the support "different", but it still all holds together like it's one piece, and you can never remove it all. However, I generate my G-code in Slic3r, which is infinitely more versatile. I can specify any parameters I need with it. It's more complicated to use in some ways, but the interface is actually relatively intuitive. Once you've got your head around the ins and outs of 3D printing, it all makes sense. Anyway, it has a very clever support generator built into it, that takes into account that you are likely to be using the same material for your support. Without going into too much detail, it allows for a minimal footprint, and a detachable interface layer. On my last project, the War Doctor Console Room, I used it in a similar application when I was printing pipes for the floor. It worked really well in that instance, so I was feeling pretty confident about it on this occasion as well...

... to cut a long story short, it DID work well on those 2 smaller barsl; the support pretty much just snapped off all in one piece! On the larger one however...    The interface was a little too close to the wall, and bonded too well with the sides and top. It was also a little to tall compared with it's length, so i really couldn't get a good grip on it and apply enough force to just twist it off. it came off in bits and pieces, and wasn't particularly clean...

... in addition to this, it turns out that I hadn't made the holes or slot for the pivot pins large enough. I think I already knew that, as I had to enlarge them on the Front Bottom Section of Module 2 as well! Completely forgot about altering them. On the 2 smaller lever bars I made short work of that with the drill   , but there really wasn't any way for me to do that with the slot on the longer lever bar...


...


...


...


...


... didn't stop me from trying though did it...?





*sigh*

... it's not so much the majority of the length of the slot; the material is flexible enough to accommodate that. The problem is when it gets to the ends, and it needs to mesh nicely into the end diameters. Then, it's going to just stick before it gets there, possibly even breaking the lever bar. Anyway, I already stuffed up this one, so I printed it again with the larger diameters. I adjusted the support material so it would have a much more thinly spaced interface layer, which SORT of worked. The print wasn't as clean though. It didn't matter, the part worked as it was supposed to...   ... now, onto the door assembly...

















... so far so good...








... this was another of those moments that made me really happy, similar to the one I had when the Rocket Flues printed. This is seeing an idea come to life. You may or may not be able to notice on that last image that there IS a slight twist in the assembly, like I suspected there would be; the doors, when viewed from above, don't sit parallel, and it's fairly significant. You can straighten it out easily enough by twisting them though, and the levers flex accordingly without too much strain. This bodes well for the hinges and the glue. Pretty sure both will be strong enough...


       


... Bob is in a huff! "Disassemble AGAIN??!?!!?"... but the door mechanism assembly looks good though...








YAY! The doors close! You can see better here that that twist is putting the top door off centre. That's fine at this stage, the hinges haven't been glued in place...


   


... the assembly in the closed postion looks good from this angle too! And, it works!      








... Dalek Bob is underwhelmed "Wow. It has Doors. Great; where's the rest of it? You know, I think my days of not taking you seriously are coming to a middle..."


... so, now we're onto something I was dreading; actually gluing bits together. I've always been a bit of a slip-shod gluer, inevitably ending up in a sticky mess, with very haphazard results. I'm hoping that my preparatory work on the model pieces, and the glue applicator can split the difference...

I initially thought that, because the hinge was at a different height than the top of the module / door, that they would need to sit on 2 surfaces of the same height with a gap in between to allow for the hinge in order to glue it. I thought about what I could use for a while, until I remembered I had 2 copies of a number of books in my library that would work...











... 6th edition of Software Engineering to the rescue. These were the flattest and most equal in height of my duplicate hard-cover books, all of which are to do with programming. Long story, which I won't bore you with. And, to cut another long story short, I needn't have bothered. It was actually easier, quicker and more accurate to just hold them in place by hand, and glue each piece separately rather than all at the same time. My prep work was good enough...

... and the finished result...














... cool. There is a gap between the edges of the doors and the hull when it's closed. It's like the edges of the hull have been splayed out. That may have been a result of the printing process, and the fact that there was a gap there, maybe also a result of the weight of the other modules sitting on top without support. In any case, something to address in the next iteration. Otherwise, I'm not unhappy with this. I figured it wasn't going to be perfect. I just hope that when it's finished that it's good enough. Anyway, followed suit with the lower hull as well...

















... and the hinges are in the right places too! The doors open fantastically as well, nice smooth motion, no dramas. If anything, the hinges are a little stiff (probably because they're new and haven't been used that much), so they stay open / closed on their own. If I wanted to, I could just leave them like that, and it would probably be fine, but I image that after opening and closing them for a bit it would get loose. So, glad I incorporated the door mechanism.


I'm actually coming up on 80 images for this post. 100's the limit. So, I'm going to continue this in another update. Let me just leave you with these...


       


   


   




D.