DIY Moon Phase Dial

[Stub Mandrel]

You could try chopping the front of the LED off then sanding it flat - I've done this and it gets rid of the lensing effect that narrows the beam and helps diffuse the light.

[Gina]

Thank you - I can try that

Very rough night here too.  Much better this morning   Even had the sun out for a while but clouding up again now.  Forecast is still dreadful

The latest print from Thingiverse came out smaller than I thought - about 62mm diameter, and the detail is barely visible.  It might be better if printed bigger.  If I printed it at 1.5x scale it would be 93mm diameter which would fit inside my 100mm black hemisphere.  Printing time would be over 5 hours though even as just a shell with no fill   That's with the little UP printer - should be less with my Titan with twice the nozzle size.  Of course, with the real thing the detail is only really good with side lighting rather than at full moon but for the moon dial this is not practical.

[Gina]

Here are two views of a model (diagram) of a fixed white moon globe and rotating shield.  I've shown the parts coloured for clarity - the shield is shown in green and the drive tube in red, these will be printed in black.  The drive gear is shown in blue and has been printed in transparent ABS which looks white.  The LED will be placed inside the white globe and wires to it will go through the white tubular "stalk".

[Gina]

Set up to print the white moon globe on my Titan (Large) printer.  Using 1.5x scaling giving a 93mm diameter globe without fill (shell only) it's scheduled to take 1hr 14m which I'm happy with.  I have loaded the printer with white ABS filament and currently bed heating.

[Gina]

Hotend not hot enough - filament not extruding, had to increase temperature setting and re-slice.  With such a complicated object this takes a long time as does uploading the g-code to the SD card which is proceeding ATM.  All this is taking longer than the print!!    If this run doesn't work I'll have to set up with a simple object first - I don't have infinite patience   Often slicer settings need changing for different filament.

[Gina]

Got the little UP printer working again this morning and now printing the shield in black ABS.  I shall be setting up the Titan with test runs for the white ABS filament.

[Gina]

Moon globe shield and drive tube printed and glued together.  Now setting up the Titan for printing the white moon globe.  I'm trying natural nylon filament with a 1.2mm nozzle, after the 0.8mm nozzle got gummed up with ABS.  That's now soaking in acetone.

[Gina]

Test cube printed well enough so now trying the moon globe    Printing with natural nylon at 260°C all layers and bed at 120°C all layers.  Printing time is estimated at 1hr 14m.  You can add 15 or 20 minutes to that for uploading the g-code and another 10-15 minutes or so for heating.  Printing direct to glass.  Blue painters tape is supposed to be better but I haven't got any.

[Gina]

It's printing but it's come loose from the bed and wobbling around somewhat so whether it will finish the globe without falling over remains to be seen

A bit later...  It parted company with the bed just before half-way.  Trying again.  I have Frog Tape which should be similar to blue painters tape except that it's green

[happy-kat]

How accurate do you find the printed measurements please Gina?

If you design a 30mm diameter circle is that the printed size you get or in practice is there a +/- 1mm tolerance or is it more?

[Gina]

How accurate do you find the printed measurements please Gina?

If you design a 30mm diameter circle is that the printed size you get or in practice is there a +/- 1mm tolerance or is it more?

The UP Plus 2 printer produces prints which are consistently 0.1mm bigger than modelled.  I allow for this in my prints.  To get this accuracy required careful calibration by printing a calibration test print and carefully measuring the resulting print.  For this I used digital calipers.  The error is then fed into the software to make the correction.  With my home built printers I get slightly less accuracy - something like 2 or 3 times.  Again it depends on accurate calibration with the error calculated and alterations made to the Marlin firmware.

With care I can print parts which fit together pretty well.  Usually some cleaning up is required as 3D printing tends to produce random little blobs of filament on the surfaces of prints.  A craft knife and/or small file soon fixes this.

[Gina]

More photos.  The print having been dragged out of position and filament deposited all over the place and printing on Frog tape.

[ChrisLX200]

Wow, that's hot (260 and 120)! I've never tried printing with Nylon but I did make a slight mistake and tried printing PLA at 250 because the reels got mixed up. It doesn't work.... ;-)

ChrisH

[Gina]

I use 270°C for ABS filament reducing to 260°C and 110°C reducing to 100°C for the bed.  I've never had much success with PLA but intend to have another go with it sometime.  The Frog tape didn't work very well but it might be better at a lower temperature.  The suppliers reccommend blue printers tape on glass at 60°C so I'll try a lower temperature with the Frog tape tomorrow.  I thought I had some 3mm white ABS but seems I haven't - must get some.  I've got 3mm white PLA though so maybe it's time to try that again.  Neither nylon or PLA can be glued very well which is one reason I like ABS.  Solvent welding with acetone works very well - with added disolved ABS scrap as a gap filler glue.  ABS can be painted too whereas paint does not stick well to PLA or nylon.

[Gina]

Tried various bed temperatures with the Frog tape and no joy.  At lower temperatures the nylon doesn't stick to the tape and at higher temperatures the tape doesn't stick to the glass.  With the post upside down over Christmas I think I'll try the white PLA again and see if I can get that to work.

[ChrisLX200]

I recently bought a can of this stuff: http://www.amazon.co.uk/gp/product/B013I8APBE?psc=1&redirect=true&ref_=oh_aui_detailpage_o02_s00 because I was having a lot of difficulty getting ABS to stick to glass, but the laquer is supposed to work for all. I'm trying it at the moment for a print that caused me problems in the past, but to be honest I improved my method of using a glue stick and that works pretty well too now. I'm not used the slurry method.

ChrisH

[Gina]

Thanks Chris - that's interesting - probably have a go at that   I would like to use PLA sometimes and maybe nylon as it has quite different properties.  Doesn't smell either

I find ABS sticks well to glass at 110°C then comes completely loose when it cools down to around 40°C.

[Gina]

I'm doing a test print of the moon globe using the nearest ABS I have to white - which is fluorescent yellow.  If this is successful I can print a white one when I get white ABS filament.

A bit later...  It needs better support.  Whereas the test cube works fine the moon globe print has a lot of overhang and this is not adequately supported.  I've got the support angle set to 30° which would normally be fine for a sphere but this isn't exactly a sphere though I would still have expected more support to be generated in Slic3r.  I seem to be having so much trouble with this moon globe from Thingiverse that I'm wondering it this is such a good idea after all.  I like the idea of a textured globe particularly if it actually looks like the moon but...

[Gina]

I think I'll just print a plain globe for now and get the clock working rather than any more messing about.  I think the shield idea is still a good one as it saves slip rings.  I could paint markings of craters etc. on the surface of the globe later - after all I am supposed to be an artist

[Gina]

Done a bit of research and found a suggestion of using diluted PVA glue on glass for PLA filament and heating bed to 55°C.  Tried it and found lack of adhesion so increaded bed temperature to 60°C and that seems to work.  Printed a test cube with filament at 230°C which seemed to work - sort of - so now trying to print a white hemisphere.

Later... Print not satisfactory

[Gina]

Decided to print a 92mm hemisphere with supporting tube on the UP printer using transparent ABS.  That will fit inside the black hemisphere and face the front.

[Gina]

Printed and cleaned up then assembled with shield and ali tube support.

[Gina]

With the shield being bigger than the white "moon" the terminator is not quite right but I might be able to improve this by printing a thinner shield or chamfering the edge.  It may not look too bad when the moon is internally illuminated, the shield coated matt black and the background covered with flock paper.

Here's the moon phase images with black level altered to show more contrast.

[saac]

Gina that is looking really good and, if I may say, a much neater solution. Good luck with the remaining design and build.

Jim

[Gina]

Thanks Jim