3D Printing

The recent heatwave has brought a whole new bunch of problems. The extruder motor, it sits right above the "Hot Bit" of the extruder, became overheated, what happened was that the filament was softened by the high temperature, instead of sliding neatly down the guide tube, it wrapped itself around the drive gears, and tucked itself into any space it could find. I had to dismantle the extruder in order to clear the jam. This had a "knock on" effect as rebuilding the extruder meant that the Z axis had to be recalibrated. This took a couple of hours, fortunately the printer carried on working after the rebuild, and the current print of a Piper Cub J3 with floats was resumed. I managed to keep the printer working, by putting a fan salvaged from an old power supply, close to the extruder. This kept things working, and no jams happened. I checked the Prusa website for extruder motor cooling mods, and found one! The mod is a cowl that fits over the motor, takes a 40mm fan, and costs less than £20 including postage. Should get the bits before the weekend.

Brian, you could have used an edf to blow over the extruder. :lol: EDFs have other further uses like assisting with getting a charcoal BBQ going quicker. If you had been a Scale Day you would've see a live demonstration.

Recently I had the 3DPrintLabs "Joker" up for sale. There wasn't much interest shown and today I removed all the hardware. Good thing that the model didn't sell. While holding the fuselage quite gently as I took out the elevator and rudder servos, the front of the fuselage started to crack. Further squeezing along the fuselage produced more cracks. The material used for the 3D print was a "bargain" from China. The wings and tail, which were printed from a different filament produced in the EU, were as strong as the day they were printed. Fortunately that was the only model I printed "on the cheap". Now its all in the rubbish bin, including the original spool of filament. Could have caused a very unpleasant incident if it had broken up in the air. I've printed a couple of models during my "layoff" from flying, and now before I begin to assemble them, I'm going to print some critical parts and test them for strength.

My bits to modify my printer arrived earlier in the week, I'm pleased to report that they were in a smallish padded envelope, not a bit like Colin's monster box. The downloaded (and free) STL files were sliced and copied to the SD card, the printer did its stuff and within a couple of hours I had a shiny new fan shroud ready to fit. It was a couple of days before I could slot in the hours needed to open up the printer's control box and wire in the leads for the new fan. Its a bit disconcerting when the control box door swings open and all the wires make a bid for freedom. Fortunately nothing fell off and nothing broke or got hopelessly entangled in something else. Fortunately the lead for the new fan was the correct length, and with small modification to the connector, plugged in to the main board without any fuss at all. The most awkward job was taming the bunch of assorted wires that connect the extruder head to the control board. I cheated a bit and used some masking tape as a temporary hold until the flexible sleeve could be refitted. The whole job including recalibrating the printer took around five hours. No chocolate biscuits came to any harm during this time, only because I forgot that the Boss had popped them in the fridge during the recent hot spell.
I've done a couple of "push/pull/squeeze/twist tests on a couple of newly printed parts. So far nothing has succumbed to the abuse. I wonder if its an age related fault that affected the Joker's fuselage, that and being in unfriendly environment. In a shed with no insulation, temperatures do tend to be at each end of the scale, and everything is more or less permanently damp. Although the flying surfaces were all OK. I think it must be the poor quality of the filament that caused the failures.

I'm getting a fair amount of exercise these days. My computer is upstairs in the spare room, the printer is in the shed at the bottom of the garden, consequently, there's a good deal of trotting back and forth as well as a considerable amount of stair climbing. I suppose its "all good for something or other". Today I've done a lot of stair climbing as I tried to get a program for a wing section to print out. Finally I gave up and sent an email to the designer. I was surprised and delighted to get a response within three hours. Unfortunately the suggested fix didn't do the biz, so I did a spot of intense looking, commonly known as " Research", but only because it looks "posh". The problem I had was caused by a slicer programme. Fortunately, changing to a different slicer programme fixed the problem I was having. Funny thing this 3D printing biz, its all very " suck it and see ". At least it isn't boring!

When I started 3D printing back in 2019, my first major project was an Eclipson BlackWing 160. This model printed out cleanly and the parts fitted quite nicely. This was done in October 2019. I didn't pick up the model again until April 2020, unfortunately none of the parts fitted and were quite warped by being stored in a damp cold shed over the winter. Recently I looked at the BlackWing again, and wondered if it could be printed with lightweight PLA. I emailed the designer and asked the question, I didn't expect an answer, but I got one twenty minutes later. With reservations it was possible. Now here I am, after 40+ hours of experimental printing, watching the first (apparently) successful part of the BlackWing being printed. After so many previous prints had failed due to poor finish or excessive "zits" on the outer surface, I emailed the designer again. The reply was almost instant, a setting which I didn't know about needed to be adjusted. The default setting value was 0.067, the new value, which seems to be working, is 0.0. Strange how tiny increments can have such an effect on the appearance of the finished item. Its a bit like having a model that is difficult to fly, turned into a pussycat by just one click of one trim switch. I'm keeping my fingers crossed that the present set-up will work for the rest of the parts. One unfortunate aspect of using LW-PLA is that it continues to expand during moves from one part of the print to another, this makes "retracting" the filament between non-printing moves, ineffective. The wing section being printed has a fairly clean interior at the leading edge, but looks like a relic from a shipwreck at the trailing edge where there are lots of stops and starts due to the assortment of internal stiffeners.

The past couple of weeks I've had temperatures as high as 45.7°C in my shed. I should have known better than to keep "pressing on" with some printed parts. The added cooling to the extruder motor kept things going, and putting an old 12V fan close to the printer kept things reasonable. Unfortunately not quite enough. Today I closely examined the parts made over the last week or so, and scrapped the lot. I'll start up again when it isn't quite so hot. When it comes to compensating for the ambient temperature, there are no values in the tables provided for the printer/model, seems to be a suck it and see kind of arrangement. The rejected parts were either warped or showing patches of poor finish. The sellers of LW-PLA assure the buyer that when printed, the material is "easy to sand". Perhaps they have never tried to sand a wing skin that is merely 0.45mm thick.

Although there is an upside to extreme heat. I left a canopy in it for 15 mins in the back garden at 1pm on Thursday and then quickly placed it on the cockpit shell and it now fits perfectly. :lol:

I've found that the major snag in building a model from 3D printed lighweight PLA is that the assembly/gluing is pretty tricky. The parts are fragile and need to be held very carefully. Trial fitting the parts is OK, but when the glue is applied, the assembly has to be quick and accurate, and it isn't always a 100% fit after gluing. I printed the free LabPrint Piper 3J Cub some time ago, since abandoned as a lost cause due to the cumulative inaccuracies in assembling the fuselage. I wondered if there was an alternative to "SuperGlue". I'm currently remaking the Cub with a new set of printed parts. I'm using DeLuxe Materials Canopy Glue. I use a small bottle fitted with a fine tube, the joints are coated on both sides and the parts gently assembled. The glue allows plenty of time for adjustment and a bit of "jigging" with an assortment of balsa blocks and lead weights holds everything in place. So far the fuselage has gone together with no twists or warps, the joints aren't absolutely perfect, but they are a lot better than those made with Cyano. The canopy glue is very economical in use, and is quick and easy to clean up with a damp tissue. Might manage to finish this Cub after all.

The free downloaded Cub from 3DLabPrint has now gone to a good home and I have found another model to play with. I saw a video of a 50mm EDF Hawk , it seemed to perform very well (everyone else's 3D printed models always do) I bought the files to print the model, unfortunately the only working files are in G-Code, the supplied STL files are corrupted and cannot be used. I've tried to repair these files using Meshmixer with no success, I sent the files to an outfit that claimed they could fix the problems and I'd be able to print from them, that didn't work either. I contacted the designer, the result was an interesting dialogue which had no bearing on the problem whatsoever. I have printed a set of parts to test the supplied G-Codes, the result is definitely "Hawk shaped" but the finish is awful and I can't do anything to improve the quality. Its a bit late for me to attempt to learn a new set of computer skills  which I wouldn't need if the files I bought actually worked. I assume the designer is guarding against illegal copying of the design, if the STL files worked I could have tweaked the quality of the prints with my slicer programme, not only that I could have scaled the files to enable different sizes of the Hawk to be printed. Thankfully the program wasn't too expensive at around £20, but it rankles that I haven't actually got anything I can use.