Cheap 3d printer kits – Experiences?

I think this one is going to fail too. The infill is pretty crappy so i might try reslicing with a different pattern

20180203_174321 by WCA![/url], on Flickr

I have stopped the print and am calling it a day for today.

You an see that the infill never really worked and for some reason there appeared to be no ‘top’ to the top of the piece. I am wondering if this is something I am doing wrong in the Fusion360 model so will investigate that tomorrow.

You can see the beginning of the square for the crutch clip is beginning to form but because of the lack of a top to the surface it has nothing to build on so there are big gaps

I suspect that this failure is something in the model, as I said, so will look their before wasting more time. You can see the solid, smooth bottom of the original piece, same on the failed one, which is laid down properly before the infill starts.

20180203_180228 by WCA![/url], on Flickr

Yowzer, what material is that?

The infill looks like it isn’t bonding and is just crumbling away as the nozzle passes back over it.

I have only used PLA and I use a really old glue stick (I found in an old pencil case) for bed adhesion.

If you are using PETG, do you need to keep the ambient conditions warmer to reduce warping?

My procedure today and last night that seems to be working is:

Turn printer on and manually set bed and nozzle temp.

Clean glass bed with window cleaner and kitchen roll.

Give it 5 mins and check bed level.

Load filament and manually squished out a few seconds worth.

Run glue stick over approximate print area.

Re-home and start print.

Given how thin your part is and it’s purpose, I would probably use a higher infill percentage. Does it really have to be that thin?

I haven’t had a part lift yet but given how small the Z-axis movements are, I think that as soon as the part starts to lift, you are in bother.

Do you review the print steps once sliced? I find this quite useful to look for potential issues.

The only thing I can think of for PETG is that fan cooling wants to be much lower. I had mine at 0, but some use it 20% or so.

The first layer looks so good and most slicers will have cooling completely off for the first layer.

Petg. No cooling. Cura automatic infill pattern. The main part is 2mmthick

i have modified the model to put a face on the top where it appears to be missing and then projecting it 1mm back into the model. Hopefully that makes sense and we shall see what tomorrow brings.

Got the printing working today so put together a short video. https://youtu.be/ehTUSENoGRw

It it shows the printer working i it not the failed prints. I fiddled with some Cura settings and it seems to have worked. All the settings I used are shown in the video if that helps anyone

An actual use for it WCA 🙂

I got some new filament in and started printing the big bed leveling knobs for cr-10 that everyone loves. First one on top, notice anything wrong? Looks like the leaning tower of Pisa…

A quick google on leaning brought up mention of belt slippage. Had a look at the X axis motor and found the grub screws were loose.. hmm it’s got to be this I thought. After tightening them I printed again and it was perfect (bottom one) phew!

Unlike my benchy though, this was a bugger to get off the bed.

Stumpy, these are well worth printing. Make adjusting bed so much easier. – https://www.thingiverse.com/thing:2408748

Agree about the big adjuster wheels. I actually successfully adjusted the bed for one of my prints while it was doing the first layer and I realised a corner was high.

i have another useful design and print planned for the near future. Possibly another video if you think it is worth me bothering

Keep it coming WCA, we are all learning from each other! Why are you using petg over pla?
I was going to print the other three wheels tonight but feeling a bit nauseous for some reason and can’t be bothered. Will do them tomorrow hopefully.

Petg is harder wearing like ABS but easier to print with. It is also food safe so if small children decide to chew the Baby Groot I printed they won’t die as quickly. I want to print some stuff that will be used outside and PLA degrades quite quickly.

it seemed the best compromise for ease of printing vs durability

papamountain – yeah, bigger adjustment knobs are on the to-do list….

These ones;

https://www.thingiverse.com/thing:2644549

I printed the squash ball vibration feet at the weekend but will be fitting then to an 18mm MDF board, rather than directly to the printer. Just need to chop up the board, paint it up and get some squash balls…

Dunno why but the bed level seems to have stabilised and it took very little effort to get half decent prints running.

I think I just need to fiddle about with the retraction settings now.

WCA – I’m thinking a setting somewhere in the slicing is incorrect. Are you using a profile for your printer? It looks like it is under extruding on the infill, which could be the printer not being able to print that fast, most slicers will speed up the infill compared to the outline. Try slowing the infill speed down.

EDIT: Just seen your video, I think those print speeds are about the fastest it can print, trying bringing the infill speed down.

Top/Bottom Layers – This should be set to about 5-8 depending on the layer height. What you are seeing in your print is pillowing, but this is also be because the infill isn’t being laid correctly
Shell Thickness – Increasing this will help retain the detail in the print, especially if there are overhangs and things. I usually always print with a minimum of 2, but you do have a nice simple print.
Skirt – Increase this to 2 or 3. This will mean you can level the bed while it prints this, giving a you a bit more time to get it level. You’ll be able to judge if the lines are too thin or thick after a few times.

Buy yourself a roll of PLA, the last thing you want to do is spend hours chasing the settings and find out it was a bad roll of filament.. Trust me, I’ve been there!

“i have modified the model to put a face on the top where it appears to be missing and then projecting it 1mm back into the model. Hopefully that makes sense and we shall see what tomorrow brings.”
When designing models, always make the part solid, your slicer will then replace the solid part with infill/top bottom layers based on your settings. You can always put the model on dropbox and I can run it through my slicer to see if the model is the problem.

I will slow the infill as I agree it seems unable to keep up with itself.

On the other bits of the print I get the occasional clicking from the feed wheel which I think is because it is trying to push through too much filament sometimes. I thought this might ause over extrusion but haven’t worked out if it is causing any problems at the moment so I am just leaving it alone.

Why PLA instead of PETG? I know it is easier to set up, I had a very small sample roll with the printer but PETG seems a better material for what I want to do. It is not as dificult to print with as ABS and much more rugged than PLA. I would like to get my printing working with this unless there is a good reason not to.

The filament was supplied by http://www.real-filament.com. I have a second roll in black that I will be using for my next project so we will see if that behaves differently.

When I designed the model in Fusion360 I thought I was designing everything as a solid so everything would have a surface. I only guessed at the issue being caused by the ‘extrude’ function causing an issue. In Fusion360 all sides, edges and planes of the model look solid. I think the issue with the infill may be the real culprit and I was guessing wrong.

The thing I am modelling now is giving me a headache. Just trying to work out the correct approach to start the design when everything is either curved , angled or not parallel. I might post a picture if what I am trying to do a bit later if I don’t make much progress but right now I am learning a lot from doing things wrong, redoing then, scrapping that and trying it differently etc

Clicking on the feeder reel usually means its skipping, therefore will under extrude. Causes for this could be, hot end not hot enough or partial blockage, or the feeder wheel needs cleaning out.

PLA would be easier as the optimum settings are a wider range. You don’t need the settings bang on with PLA to get a half decent print. I agree PETG is a better material but harder to dial the settings in, especially when you don’t have the experience yet.

This has inspired me to dust of my printer, update my OctoPi and print a few things for my sim rig! 😉

My best piece of advice for designing 3D models is there is no right or wrong way, just might be a quicker way. 😉

V1, 2 and 3 of my design starting furthest away

20180205_160513 by WCA![/url], on Flickr

V1 proved the concept was sound but the clamp wasn’t raised far enough from the main part. The poor surface was caused by bad infil settings I believe

V2 was better but the clamp was slightly too high this timer and the wings of the clamp were a bit flimsy. Also that back edge of the clamp dug into the wrist a bit.

V3 has stronger clamp plus two holes to wire it in place if needed. It also has the bas plate extending backwards beyond the clamp bracket to remove the pressure point

And once fitted the crutch is much more comfortable without having to wrap it in scarves, socks etc…

20180205_160604 by WCA![/url], on Flickr

The video explaining the update is now available on my channel : https://youtu.be/eWfsXhpFy98

Milkie – I kicked off the second print before seeing your comment about the clicking. I suspect 235 for the hot end might be a little low. The clicking isn’t always there but is often enough to know it isn’t right. I will knock the temperature up on my next print.

My next print will be this and will be printed in my black PETG filament :

20180205_162817 by WCA![/url], on Flickr

20180205_162827 by WCA![/url], on Flickr

It is the roof rail end cap from my Range Rover. I am really ‘enjoying’ trying to model it at the moment but am learning a fair bit as I go along.

None of the sides are parallel, none of the surfaces are flat  and none of the faces are parallel which is a challenge.

I am starting from the bottom where there is the nearest to flat edge. 2 rectangles (not parallel) down each side the width of the thickest bit of the plastic and a thinner rectangle at the back joining them. Extruded that up to the height if the wedge/ramp you can see on the inside each side. Used lines to mark the wedge/ramp and remove the unnecessary material. Added the little tag at the bottom.

I hope that this correctly matches the bit of the fitting that slides onto the roof rail, basically the bottom bit of the fitting. I have then just extruded up to create the rest of the box without caring about size or style as I just want to check the ‘slide fit’ first. In fact, I will probably remove most of the box I have created before printing it, just to save time.

20180205_165156 by WCA![/url], on Flickr  ***

Once I have got the bottom fitting correct I will fous on the front edge with it’s recess to meet the other part of the roof rail, angled differently on each side of course. Once that is sorted and looks okay I will mirror it and print two. That way I can ignore the original one and be sure that the two I fit will match.

***I am having to do all this either laid on my belly or my back with my ankle elevated so the picture angles are all to co. Please trust me that the model looks more like the proper fitting than it does in the photo.

V3 looking good WCA 🙂
3D design seems to be bloody annoying lol. I spent hours the other night designing a door handle for the freezer. Scrapped it in the end… More practice required.

Stumpy, I’ve now decided against the squash ball feet. Seems to make the whole rig shake a lot more from what I’ve read. Make sure you get the least bouncy squash balls you can get – triple yellow spot.. Let us know what you think.

Milkie, so what does the OctoPi bring to the table? I keep seeing it mentioned but haven’t looked into it yet.

Also, I could do with some kind of cover to throw over the whole thing to keep dust etc off it when not using it. Why isn’t there one..

What tools are people using for designing their models? What file formats are needed for online printing facilities?

What sort of profiles are needed for strength? For example I want to print a 100mm cube but have it hollow so that I can bolt stuff into it. Ideally it would also have latticed sides so you could see into it. What width “edges” would I need to make it pretty strong, as in strong enough to be able to pass it around?

What tools are people using for designing their models? What file formats are needed for online printing facilities?

Some will take 3D files from particular software, but most of the time STL is the common way to send them.

Fusion 360 is free for hobby or startup use. It can save STLs no problem.

What sort of profiles are needed for strength? For example I want to print a 100mm cube but have it hollow so that I can bolt stuff into it. Ideally it would also have latticed sides so you could see into it. What width “edges” would I need to make it pretty strong, as in strong enough to be able to pass it around?

It’s obviously going to vary with material and ‘infill’ percentage, but PLA is reasonably strong. 5mm walls on your 100mm cube should be ok, unless you’re hanging weight off it. The 3D printer will want to fill those 5mm walls with ‘infill’. You can choose 100% (solid), or they can fill it with a lattice grid of any percentage below that.

For stuff I need to be strong, I choose 50%, for visual stuff, I tend to choose 10-15% depending on complexity of the top surface.

BTW: If you print a hollow cube, the top surface is doing what is called ‘bridging’ and can be a problem on long spans – it’s basically laying plastic down in thin air and it wants to droop. So some kind of support structure would work best here, if your design can accommodate it. Some printers/materials can bridge very well though, so it depends on where you get it printed.

SammyC – there are loads of free design solutions out there. Plenty already mentioned in this thread: Tinkercad, Fusion360, Onshape, Sketchup.

For online printing facilities, I would suspect that they require stl files.

I can’t really envisage what you are trying to achieve with the cube you mention, but it sounds achievable with 3D printing. If it’s just a display part that is needed to show a concept, the ‘edges’ could be pretty thin. It depends if you want the whole thing to be a lattice structure, or just part of it. You would also have to consider the areas where things will be ‘bolted in’. How will you bolt them (e.g. a hole with a nut & bolt passing through, or perhaps a threaded hole in the plastic).

“Milkie, so what does the OctoPi bring to the table? I keep seeing it mentioned but haven’t looked into it yet.”

OctoPi is for controlling your printer, slicing, live viewing the print, creating a timelapse. I use it for uploading my gcode files to the printer SD card, pre heating and remote viewing with a Pi Camera. I also have a touchscreen that you can control the printer with, but I prefer using my iPhone. It has Cura built in with the capability of slicing. I think I have it setup so I get a notification on my phone when a print has finished.

TimeLapse with this would be great, but I need to find a location for the camera that doesn’t interfere with the head, doors, rods, belts and has shorter focal length than the Pi Cam. I need about a 20cm focal length.
EDIT: Just found our you can rotate to Pi Cam lens to reduce focal length.

papamountain
<div class=”bbp-author-role”>
<div>Stumpy, I’ve now decided against the squash ball feet. Seems to make the whole rig shake a lot more from what I’ve read. Make sure you get the least bouncy squash balls you can get – triple yellow spot.. Let us know what you think.</div>
</div>

Hmmm. Only just seen this.
Well, I’ve bought some 18mm mdf sheet to cut to size & mount the feet on the bottom of, rather than directly to the bottom of the printer. Hopefully the added mass will stop it wobbling around. I did wonder if the squash balls would have enough stiffness to combat the translation off the axes (mainly the Y-Axis, I guess).
No great loss if it doesn’t work. I should probably have just bought a 4-pack of Sorbothane feet in the first place.

Definitely need to cut the noise down. The printer stands on a cheap old wooden computer cabinet , which sends the vibrations straight onto the varnished floorboards. These then send the noise throughout the rest of the house!

I’m curious to try the Nema stepper dampers that seen like a popular ‘upgrade’ on the X & Y axis. There seems to be an equal number of people who think they work with no detriment to quality versus those who think they definitely reduce quality.

What software to design?

I used Fusion360 and learned everything I needed to design and print my crutch modification (ooo-err) by watching 3 YouTube tutorials. Here is part 1 : https://www.youtube.com/watch?v=A5bc9c3S12g&t=2s

I spent the first few months with a computer that wasn’t powerful enough to run any software and just downloaded designs from Thingiverse which gave me a chance to get a bit more used to printing things.

There are definitely two very different skills required. 1 is to design stuff and 2 is to print stuff. They are related as a good design is easier to print and a good printer will work around some of the issues of a poor design but don’t think of them as the same thing.

I can highly recommend Washing Machine Anti Vibration Pads:

https://www.floormats.co.uk/washing-machine-mats?gclid=CjwKCAiAweXTBRAhEiwAmb3Xu-rcrTQFjQ4xXnZlZYevC98Z6h0lgTSVHfvY8del0MugWL-XIg8IyhoChloQAvD_BwE

I have a couple of 20mm thick ones on a sim rig and it cuts the vibrations down by quite a lot and there is a lot of vibrations with the bass shakers. I haven’t bought one for my printer though. I’m also using an old TV cabinet for my printer with a couple of paving slabs on the bottom shelf which has helped reduce any wobble the printer creates.

WCA – Your print is looking a lot better! Have you solved the infill issue now?

Thanks all, very useful info.

The cube needs to be taken apart for fitting stuff into it so I think I’ll design it as a set of flat panels that are then screwed together. I was thinking of adding holes for self tapping screws, I’m assuming that because it will be plastic that is fine as long as the screws have a deep enough thread to grip?

Slowing the speed solved the infill issue – than ks.

Now struggling with Fusion360 and 3D modelling the roof rail end shown earlier. I designed the base and printed it to check it slid in correctly. So far so good. I had added the top of the piece which doesn’t line up with any of the edges of the bottom.

I now need to build the walls to connect to bottom to the top but am struggling as there are no common planes and Projection doesn’t seem to do quite what I want but I suspect it is the right thing.

The back edge (left as you look at the picture) leans forward about 2cm to reach the back of the top piece.

The left edge (top as you look at the picture) leans in about 1cm but also projects forward from the front of the base be about  1cm.

The right edge (bottom as you look at the picture) leans in about 2.5cm and again, about 1cm forward of the base.

Each of the edges is 2.5mm thick.

Also wondering if Lofts are relevant. No idea what they are but the picture looks similar to what I want.

I shall persevere, sod all else to do.

How to Join by WCA![/url], on Flickr

It was Lofts I wanted but I had forgotten to select the New Body option.

An entire day spent on this and I have decided to scrap it and start again in the morning as there are so many sketches and deletions that the whole thing is a mess.

Hopefully I can apply what I have learned and do it much more quickly and cleanly in the morning

I changed from yellow PETG to black PETG. Both rolls of filament fromt he same supplier but dramatically different results.

20180207_115521 by WCA![/url], on Flickr

In the photo I am holding the top of the hot-end with the nozzle facing up and left. I guess I have some cleaning to do.

I’m just using my M3D (original) for the first time in maybe 6 months.

Used it almost daily printing anything and everything but then ran out of things to fix / modify / make.

A bit leaner. Now to re-assemble, re-calibrate and re-try

20180207_121945 by WCA![/url], on Flickr

Podge – If you are bored and want a go at the Range Rover Roof Rail End in the earlier photo’s I can share the dimensions…

Thanks, not sure I have the build volume at around 100mm cube.

Currently fixing / improving the boy’s Brio set and tipper truck.

I think you’ve dissuaded me of any ideas I had about buying a 3D printer WCA. Looks like a WCPITA to be frank.

tthew – Yep. A great experiment for £150 but unless you spend more and spend more time learning how to configure it and do everything properly it can be frustrating. If you want to design stuff and then get it printed so it is useable you are probably better off getting it professionally printed.

The main reason I have spent so much time with mine for the last few days is that I am stuck in bed with my leg elevated so there is sod all else to do.

Trying to learn Fusion360 for the modelling has been fun but the RR Roof rail is stretching my talents. Having said that, the crutch modification is great and I am really happy with that. I guess the trick is to design things with parallel flat surfaces and simple shapes. Complex stuff can be bought more easily.

Refreshingly honest! Nothing wrong with learning new skills either, I’m having a bash at programming using Python at the moment.

As an aside, I reckon a Raspberry Pi would make an ideal homebrew 3D printer controller, but that’s WAY beyond my skill set!

WCA- shapes like that are fairly complex to reverse enigneer. I do some of that through work- we have a package called Geomagic which is very good for CAD from scan, but it’s frighteningly expensive and we have 3D scanning kit to back it up.

One thing you could try is using photogrammetry. I’ve not done much of it for detailed work but some people are doing pretty astonishing things on fairly impressive budgets. I mentioned it on another thread but Agisoft Photoscan is a good one to start with but does cost money after the demo. Colmap got a good review here- https://pfalkingham.wordpress.com/blog/ . It doesn’t appear to be as ‘end-to’end’ as agisoft though.

For an idea of what you can do given a lot of time and a fair amount of cash this is a very interesting page- http://www.pi3dscan.com/. I’d like to make a mini version for ‘things’ at some point when I find the time! Something like MeshLab should be able to handle the final mesh cleaning and closing for 3D printing (although I confess I have no time on it). Alternatively I think fusion 360 has some mesh editing tools.

That would give you the outer shape of the roof bar end in mesh, then you could shell it before print.

(I’ve been lurking on this thread for a while, found it very interesting. I’ve thought as few times about buying a 3D printer for home as we have industrial grade stuff at work. However, I have a friend who is a few years further down the line so thought I’d be better off following the ‘scan’ process so we have complimentary skills).

I tried an Autodesk scanning app, it captured what I wanted but the quality was poor and in a format I couldn’t edit.

Was that 123D catch?

Yeah possibly, I think they binned / merged it shortly after I used it

It wasn’t the best, as an app at least. Good for novelty. I think the engine behind it was good and you got more control if you used the desktop version. Also got the option to use a better camera. You can get some fantastic results if you know what you’re doing (my experience with it has not been in the ‘fine detail’ field).

Thinking about it I believe fusion360 has some photogrammetry capability in there somewhere.