Thursday, 8 December 2011

Happy Holidays!

This will be my last post for a few weeks, as we are about to embark on a family holiday until early January. (It's going to kill me, being away from "My Precious" for 4 and a bit weeks!)

Anyway, don't despair that you don't hear from me for a while - I haven't disappeared off the face of the Earth, just taking a break with my partner and kids.

Until I talk to you all again - here's a seasonal print, hot off my new heated bed:

See you all in the New Year - happy holidays, and happy RepRapping!

A Hotbed of Intrigue

One thing I have been planning to do right from the start of my RepRap build was to add a Heated Print Bed – partly to allow me to print with ABS, but partly because by all accounts, you can get better quality printing even with PLA on a heated bed.

I bought my Mk. I PCB Heated Bed on eBay , and it arrived a couple of days ago. It seems to be very well made, and a quick check with the multimeter confirmed the total resistance was within spec. I scoured the forums and Wiki looking for ideas on how to mount it. In the end, I bought a small sheet of MDF board and cork from a local craft shop – they are intended to be used to make a table coaster / protector, but I have other plans! I also bought a small mirror with MDF backing board while I was at it, figuring I could cut myself a piece of mirror to use as a heated printing surface.

I wired up the heated bed and a 10 kOhm thermistor to my RAMPS, modified the Sprinter settings to have the correct thermistor properties, and started up Pronterface so I could set a bed temperature of 60 degrees. Success! The bed heats up, and when the thermistor detects 60 degrees, the heater circuit shuts down, starting up again when the thermistor temperature reading drops – brilliant! Time to get serious about mounting the heated bed and try some printing.

I didn’t want to start disassembling and reassembling my Y-Axis prematurely, so I have taped my thermistor onto the heat trace element side of the PCB, and then placed the PCB board onto the cork board heat-trace side down. I chose this orientation so that if I have a “head crash” I should not damage the heat trace elements. I have cut a bit of a rebate into the cork board for the thermistor wiring and the main power supply wiring (because they are also on the underside of the PCB) so it will sit nice and flat on the cork.

I placed the MDF board onto my acrylic print bed (still covered in Scotch Blue Tape), then put cork sheet on top, followed by the PCB, and finally the glass. I clamped the whole lot together with bulldog clips, and fired up the RepRap again. Initially, things were heating up quite nicely, but then – disaster! Crack! The mirror has split in two. In hindsight, clamping the glass down was probably a mistake; I suspect that If I had let it “float" on the PCB heart, all would have been well, but by clamping it down, something had to give, and it was my piece of glass.

Oh well – Plan B!

After cleaning up the broken glass, I disassembled everything, and then covered the top face of the PCB with my 20 mm Kapton tape, and put it all back together again. This time, when I fired up the RepRap, all seems to be fine. The heat bed and the extruder nozzle can both heat up independently, and seem to be able to hold temperature within a fairly narrow temperature range. It looks like it will take just a couple of minutes to reach PLA printing temperature of 60 degrees; rather longer for ABS @ 110 degrees – maybe 5 minutes or so.

Scanning over the PCB with a non-contact IR thermometer indicates there is some temperature variation across the surface of the PCB, with the edges being about 5 degrees cooler than the middle, due to the greater heat leakage. Pleasingly, the IR thermometer indicates the top face centre of the PCB is within about 1 or 2 degrees of the reading given by the thermistor on the underside of the PCB – it looks like my heat insulation is doing its job!

Nothing for it but to try some prints. I thought I would try a couple of the calibration test pieces first, to see how the quality compares with my previous “best work”.

It went brilliantly – the PLA filament seems to stick beautifully to the heated Kapton, but also releases beautifully when the heatbed cools down. Quality seems to be very good as well – although I have only done a couple of test prints so far. At the moment, I am still printing on a 2-layer raft, but the underside of the raft is super-smooth. I think I will try to get my bed really level, and then try printing straight onto heated  Kapton without any raft – that is a job for the future – as is trying to print with ABS, which I am now dying to try.

So – in summary:

You don’t NEED a heated bed to print PLA, but I’m told you DO need one to work with ABS. However, seeing how good the quality is when printing PLA onto heated Kapton, my advice to any newbies would be to definitely include a heated bed in your future upgrade plans – or else, build it into your original plans. It really does seem to help in getting quality prints!

Thursday, 1 December 2011

RepRap Material Test Facility - Part II

Hmmmmm ...

... this may need a little more thought.

I made some sample test pieces and a test bed, and the first thing that became apparent is that the original design is all a bit unstable for the sorts of loads that are required to load the test pieces to destruction - basically, the whole assembly will probably topple over long before you can get the full load applied, if you use my original part designs.

So I modified the designs to make it all a bit more stable, by adding some broader  feet to the test bed, and making the supports on the test pieces a bit wider. I also added the option of a removable Load Platen, to allow the test load to be applied more easily to the load pad on the top flange of the test piece, if desired:

and here are the parts I made:

and here's the complete assembly (using the lighter, weaker I-Beam test piece, rather than the solid rectangular beam) sitting on my kitchen scale (so I can do some quick load testing):

and here it is with my thumb applying a 5 kg load:

You can see that the test piece is bending, but it has only deflected about 3 mm with 5 kg load, and it is nowhere near structural failure. At a guess, it is going to be able to sustain more than 10 kg of point load.

So, what's the problem? Well, 10 kg is going to be quite a pile of weights, and I am not sure how to apply that much weight to such a small load pad in a stable fashion , and still have room to get my digital calipers in to accurately measure the deflection for the various load increments. I'm going to have to work out a simpler loading mechanism which can apply 10 or 20 kg of load, while still being able to measure both applied load and the resulting deflections.

Does anyone have any suggestions?