3D Printing for Home Improvements

[davidktw]

I will use this thread as my main mode of sharing my personal 3D printing journey, hopefully to entice new comers to the 3D printing world. It has been years that 3D printing spring off, but back then one will need to assemble their own 3D printer and also I would consider it a hobby with certain level of engineering and tinkering learning curve. For awhile I have been procrastinating on whether to get one, and it is until Snapmaker U1 made it to the news in kickstarter that I finally decided to pull the trigger.

I'm not advertising it in anyway, but the fact it made it to the news is because if it's tool changing capability and price point that makes it a no brainer choice unless you are willing to fork out US$3,899.00 for the proven Perusal XL5 (a 5 tools changer) as oppose to the Snapmaker U1 pricing at only US$799 for a 4 tools changer. Bambu Lab don't even have an offering until awhile later with their Vortek Upgrade kit priced at US$760, which still need to procure the H2D or H2S, while their H2C retails at US$2199 meanwhile. Hence you can see why Snapmaker U1 is such an attractive offer back then, and even now.

It took me weeks after I got the parcel to unbox. and assemble it, and try out some silly dragon decorative piece with their PLA filaments. But I quickly jump into the parametric modelling with Onshape, for the following reasons:

1. Much like Steve Jobs with his typography classes back in his uni days which brings forth nice looking vector fonts to Macintosh, the one lesson that connects the dot backwards for me is ME1101 in NUS which teaches Solidworks. I am not in the mechanical engineering course, I took ME1101 having my interest in 3D modelling (back in my secondary days with 3DS and then 3DSMAX). Looking back, it has to be one of the best cross faculty module I took
2. Because what I want to achieve is accurate modelling that can fit into the real world usage, and not some fuzzy art forms, a engineering CAD make sense
3. Onshape is a full cloud CAD product and free model works for an individual whom doesn't mind open source their works.
4. Onshape doesn't require you to have a highly capable machine. You can use it with any recent (or not to old) iPads, iPhones, laptops of any OS with a modern browser. All the compute happens in the cloud, meaning your machine capability is secondary. For me, I uses it on my iPad so that I can easily model sitting infront of what I'm dealing with.

As for plastic materials that I have tried so far, they are PLA, PETG, PCTG, TPU. I have yet to try on other like Nylon, ASA, ABS for they do produce rather toxic fumes during the process. I have also not invested in materials fused with carbon nanotubes or glass fibres yet mainly I have not come to any critical stiffness requirements and these materials can be hazardous to the environment you are living in if you do not have a dedicated workshop away from your living quarters.

So you are here to see what I have done so far ?
My main objectives for 3D printing is not to always create something totally by 3D printing. I use it as a tool to supplement other traditional manufacturing processes or to create adapters integrating readily available parts together to accomplish a final outcome. Works that I'm sharing are not in chronological order, I share based on how interesting they might be.

 

[davidktw]

I will start off with the bike pole bottom skirting that I have created to prevent someone from injuring themselves when kicking onto a retraction spring mechanism by encapsulating with a bell bottom skirting. It also helps my robot vacuum cleaner to navigate better around it

From the 2 side profiles you seen above, that's it how I use it to create the final product. I uses it to create the following 3D model

It is created using a lofting method to morph between 2 side profiles, top most is a circular profile which embraces the pole tightly. Bottom is a eclipse profile so that the protrude mechanism can be covered within. It is then split into 2 vertical halves, which I then added dovetail joints to 4 areas so that I just need to position on halve to the pole, and the other halve will slide down from above on the other side of the pole. Hence after it is completed, you will only see 2 seams, but no binding mechanism externally. Just a very smooth bell bottom part

This is printed using PETG and took me more than 6hrs to print.

 

[davidktw]

Here is the most recent design that took me 9hrs to print. Despite Snapmaker u1 is already quite fast when changing between materials(tool heads) as oppose to Bambu Lab's AMS, it still does introduce excess time when printing layers consisting of multiple materials.

What you are looking at the above image is a 3D printed adapter that integrate my toilet fan to an aircon duct. The aircon duct will lead to a panel at the sliding window which direct the exhaust air out of my apartment. For some silly reason, my apartment common toilet only have 1 window, which does out lead out of the HDB flat. It leads into the kitchen/service yard area. I have fitted a exhaust fan on the window so that it can clear off the smell and also keep the toilet dry. I automate the entire drying process using smart home concepts. Just a voice command from me will switch off the toilet light, and activate 2 fans, one to blow into the toilet and another to extract out of the toilet in a timed manner.

When I first have this idea, I haven't even got a 3D printer. So snapmaker kickstarter came at the right time, I pledged for it, waited for months and it finally arrived last year November. Then it took me months for me tinkering it, getting to understand the manufacturing process and working with different materials, to finally I got it done few days ago. Few hours to come up with the design and then hours of waiting for it to be printed.

This design also make use of the fact that PLA and PETG does stick, but doesn't bond. Hence I created my own support materials in the CAD modelling process, so that I can control much better how the support are placed. Those green PLA parts are basically sacrificial parts which helps with ensure the underside of any final parts they are floating in the air are supported by PLA from underneath. Normally one can use the same material by leaving 0.2-0.3mm gaps between them so that they don't bond well, but it also results in badly printed undersides of the actual parts because those soft molten plastic materials will be hanging overhead of the same plastic material, it may or may not bond well depends on the type and quality of the filaments you are using and also your extrusion and cooling settings. Still so far I discover none of them will come out well, the surface will tend to be wavy and have that very obvious strands of lines going across. Thus by using support materials (depending in your actual part material), could be PVA, PLA, Support PLA, or so forth as long as they don't bond well together, you can leave no gaps between the parts and print. Once they cooled down, u can just pry them apart with some force.

So my design take into consideration that I need something flexible to lock the adapter to the fan, hence you see the 2 bended tabs at the top, and also tabs inside the circular part for the aircon duct to screw into it. This adapter greatly improves the look and also the efficiency of airflow and ease of cleaning. No adhesive or no double sided tapes are required.

 

[yhw22]

Nice for you to start a post to share your learning curve. I'm still learning it. I own a Bambu Lab A1 printer. So far I have printed using PLA and PETG filament. I don't have an enclosed environment, I just make use of a bedroom to use it for 3D printing. Humidity level was always at least 60-70% but it will go down while printing to 50%+ (min).

I tried printing PETG material, maybe it was due to the printing environment, which causes the product printed to be brittle too.

I used TinkerCAD to design my interest, which I had learned through the free NLB MakeIT 3D Printing course. So far, I had done so for festival ornaments.

 

[yukita]

Please keep sharing. this is very interesting.

 

[davidktw]

Allow me to share something less innovative.
Containers, I’m sure along the way, one would be creating containers using 3D printers to store stuffs. For me, one of the early containers is this headboard container which I created to store my fan and curtain controllers for ease of accessing them when I’m in my dreamy state. For obvious reasons it is full of curves.

My early 3D printing skill tell me there is no good way to position this without resorting to support structures and for some odd reasons, my initial experience using PLA as interface layers wouldn’t helpful because the brand of PLA and PETG just don’t stick well to each other, causing failures in printing. I only managed to solve it recently.

That is why when you look into the depth of the hole, the surface looks so unsightly and I’m prepared for it because the container is printing with the hole down, meaning the support structure touches the hole bottom has it is facing downwards.

To solve that problem, I created a negative component that is exactly the same as the hole, then I printed it upwards with 0% infill, and only 1 wall thick sides. This is also known as a fake vase mode. I cannot do the spiral vase mode because the bottom requires support due to the curves of the inner hole being not steep enough. Hence you see the black 1 wall thick part, before I slot into the hole and then uses super glue to glue it to the entire piece.

 

[yhw22]

Most people stated that PLA and PETG don't stick which is why a lot of people use both vice versa as support for easy removal.

 

[rosschang]

Very nice of you ts to share your journey.

I'm very interested with 3D printing but I can never justify to spend the $$$$$ because I just can't find any day to day use.

Do post more stuff.

PS: I can't see your image links... I think is block by corp network... but ho well.

 

[davidktw]

Most people stated that PLA and PETG don't stick which is why a lot of people use both vice versa as support for easy removal.

they stick, don’t bond. if they don’t even stick, you will have problem printing one on top of another as you are drawing lines, it will slip off before you even can finish forming a layer… especially the wipe tower

perhaps more technically we should say they don’t adhere well, but they should adhere to a certain level.

 

[yhw22]

Very nice of you ts to share your journey.

I'm very interested with 3D printing but I can never justify to spend the $$$$$ because I just can't find any day to day use.

Do post more stuff.

PS: I can't see your image links... I think is block by corp network... but ho well.

If $$$ is a concern, I can sugggest you go attend free course by NLB MakeIT. It will teach you on how you design to print using a Bambu Lab P1S machine.

After you had attended the course, you can book a slot of 2.5hr per session and everything including the filament is provided to you FOC. You just bring yourself, the rest is provided by NLB even laptop and sdcard.

 

[yhw22]

they stick, don’t bond. if they don’t even stick, you will have problem printing one on top of another as you are drawing lines, it will slip off before you even can finish forming a layer… especially the wipe tower

perhaps more technically we should say they don’t adhere well, but they should adhere to a certain level.

Yes, both PLA/PETG dont adhere properly so it makes it easy to remove the support. I just learn that not the whole support is using the either PLA/PETG, it is the support/raft interface that using it. If not, it will be wasting lot of material, especially when multiple filaments are used, it will produce poops per layer.

 

[jinsatkilife]

Can share why not use vibe coding blender using openclaw, Mac mini agent to make designs?

 

[davidktw]

Yes, both PLA/PETG dont adhere properly so it makes it easy to remove the support. I just learn that not the whole support is using the either PLA/PETG, it is the support/raft interface that using it. If not, it will be wasting lot of material, especially when multiple filaments are used, it will produce poops per layer.

From my personal experience, the choice is probably not the actual monetary cost, but your time (actually it is also a form of cost). The PETG filament that I'm using now cost more than the PLA filament. So if I print the entire support using PLA, it is actually more cost effective.

But I don't do that because it will be wasting a lot of my time as each layer will consist of 2 materials which means there will be time incurred for the tool changing operation as oppose to only happening during the layers where the interfacing material is printed.

The typical process of how the tool changing works is as follows

1. Put back the current toolhead
2. Pick up the new toolhead
3. Wait for the hot end to reach the extrusion temp
4. Extrude(wipe) into the wipe towel
5. Proceed to extrude into your model

Step 3 can be very fast or slow depending on how long the toolhead has been parked. Tool changer will maintain the temperature but I understand it is not at the extrusion temperature. So it will take time for it to be heated up when you need it. From experience, it is quite fast say 1sec max.
You can also adjust how much you want to extrude at the wipe towel to gain back some time, but generally it can be between 1-3s of time wasted.

Let me share the different scenarios with you.

The above only uses 1 single PETG(black) material for the entire model inclusive of the support, it will take 5h43m to print.

Then I change the support to also PETG(green). So the model now requires 2 materials to print, yet both are the same type of material, just different colour. It increased to 7h36m to print if I will to choose the entire support to be printed using the PETG(green). It requires 444 filament changes.

What if I only choose the support interface uses PETG(green) ? The time immediately drop back to 6h16m to print.

You might now ask why did I choose the support using PETG instead of PLA ? Because I know by default PLA filament settings are set at faster printing speed than most PETG unless you are using some PETG High Flow formula filaments. All I am demonstrating at this point of time is how much time is wasted during tool changing.

Hence I'm showing you the next scenario where the entire support is printed using PLA(green). You realise it still takes 7h31m to print. Sure the PLA faster printing speed save you 5mins, but the tool changing process of 444 times dominates.

Also the more you tool change, the higher chances that your printing might fail. Especially when the wipe tower shift. The wipe tower may be fundamentally useless, but during the printing process, you need it to be there or else the wiping process will be wiping into thin air and something it will just simply screw up your actual model because the printers are not so intelligence to handle a misaligned wipe tower.

 

[yhw22]

From my personal experience, the choice is probably not the actual monetary cost, but your time (actually it is also a form of cost). The PETG filament that I'm using now cost more than the PLA filament. So if I print the entire support using PLA, it is actually more cost effective.

But I don't do that because it will be wasting a lot of my time as each layer will consist of 2 materials which means there will be time incurred for the tool changing operation as oppose to only happening during the layers where the interfacing material is printed.

The typical process of how the tool changing works is as follows

1. Put back the current toolhead
2. Pick up the new toolhead
3. Wait for the hot end to reach the extrusion temp
4. Extrude(wipe) into the wipe towel
5. Proceed to extrude into your model

Step 3 can be very fast or slow depending on how long the toolhead has been parked. Tool changer will maintain the temperature but I understand it is not at the extrusion temperature. So it will take time for it to be heated up when you need it. From experience, it is quite fast say 1sec max.
You can also adjust how much you want to extrude at the wipe towel to gain back some time, but generally it can be between 1-3s of time wasted.

Let me share the different scenarios with you.

The above only uses 1 single PETG(black) material for the entire model inclusive of the support, it will take 5h43m to print.

Then I change the support to also PETG(green). So the model now requires 2 materials to print, yet both are the same type of material, just different colour. It increased to 7h36m to print if I will to choose the entire support to be printed using the PETG(green). It requires 444 filament changes.

What if I only choose the support interface uses PETG(green) ? The time immediately drop back to 6h16m to print.

You might now ask why did I choose the support using PETG instead of PLA ? Because I know by default PLA filament settings are set at faster printing speed than most PETG unless you are using some PETG High Flow formula filaments. All I am demonstrating at this point of time is how much time is wasted during tool changing.

Hence I'm showing you the next scenario where the entire support is printed using PLA(green). You realise it still takes 7h31m to print. Sure the PLA faster printing speed save you 5mins, but the tool changing process of 444 times dominates.

Also the more you tool change, the higher chances that your printing might fail. Especially when the wipe tower shift. The wipe tower may be fundamentally useless, but during the printing process, you need it to be there or else the wiping process will be wiping into thin air and something it will just simply screw up your actual model because the printers are not so intelligence to handle a misaligned wipe tower.

You may want to tryout what the expert had advised, by changing the "Support/raft interface" to another material only. "Support/raft base" as default. So that it will have so many changing of extruder. Save time and less wastage.

 

[davidktw]

You may want to tryout what the expert had advised, by changing the "Support/raft interface" to another material only. "Support/raft base" as default. So that it will have so many changing of extruder. Save time and less wastage.

I have been using that all the time, it doesn’t save much material lah. your support still need to come up from the base plate right? so the question is u print using what?

When you are dealing with a many hours printing…. unless your material is those engineering material that are costly… i highly doubt you want to add another 10hrs just to save a few dollars. Besides it is never cost effective to do 3d printing. We either do it out of novelty, a hobby or fast prototyping for mass production using other means

 

[davidktw]

Can share why not use vibe coding blender using openclaw, Mac mini agent to make designs?

does it works for you? I don’t see why i need to do that if I want complete control over how my models comes out, as opposed to hope that how my prompt will be interpreted properly by the AI which then will be properly modeled using the right topology during modelling

I uses onshape because i like how parametric modelling works. If you have experience with it, you will understand how that you can go back into your modelling history, change some parameters, come back to the latest state and see the updates happen.

 

[davidktw]

All home improvement and no fun wouldn't make sense right ?

Hence I 3D printed a luggage tag awhile ago. Then I bought some UV epoxy resin, pour into the recessed area, totally submerging the raised design and then cure it with a 30W UV lamp for roughly 8mins.

Honestly, not very well done. First I didn't remove away the bubbles, which I could using a torch flame, or maybe even a lighter. Next it also wrapped the tag during curing due to epoxy resin contract during the curing process. Might try multiple thin layers another time.

 

[davidktw]

Awhile ago, I was trying to hang one of my wall clock, however I decided not to drill any fixtures into the wall to hang it.
Also for my use case, I would like the wall clock to be 45 degree showing from a corner of my living room. So I decided to leverage on a nearby blind casing. At the side there are 2 screws which I can use to leverage on by simply 3D printing parts to extend from it. Here is the design

It comes as 3 parts, one part is fixture to leverage on the already mounted blind and its casing. The other is a Y clamp to grip on to my wall clock.
There is a third component that link the 2 parts together. It's bought off the shelf

Basically I used it to provide the linkage which allow me to freely tune the facing of the wall clock to whichever inclination and direction I need.

Because it uses the standard 1/4" thread, I 3D printed the threaded holes onto the parts, and they exist as separate parts because I wanted to print the threads as detail as possible using 0.08mm layering, yet the rest of the parts I just printed with 0.28mm layering for speed.
I have already taken effort to split the parts when it is larger than what the 3D printer can accommodate and also to ensure they are placed in optimal direction during printing to minimise on supports. Assembling are just simple slot in and super glue them. There are mechanical features such as the dovetail joints used during assembly to ensure load bearing and fitting.

Here are some images to share the outcome of this DIY Home Improvement project.

 

[yukita]

I will start off with the bike pole bottom skirting that I have created to prevent someone from injuring themselves when kicking onto a retraction spring mechanism by encapsulating with a bell bottom skirting. It also helps my robot vacuum cleaner to navigate better around it

From the 2 side profiles you seen above, that's it how I use it to create the final product. I uses it to create the following 3D model

It is created using a lofting method to morph between 2 side profiles, top most is a circular profile which embraces the pole tightly. Bottom is a eclipse profile so that the protrude mechanism can be covered within. It is then split into 2 vertical halves, which I then added dovetail joints to 4 areas so that I just need to position on halve to the pole, and the other halve will slide down from above on the other side of the pole. Hence after it is completed, you will only see 2 seams, but no binding mechanism externally. Just a very smooth bell bottom part

This is printed using PETG and took me more than 6hrs to print.

I wonder how much does it cost roughly to print each item? lets just say this one?

 

[davidktw]

I wonder how much does it cost roughly to print each item? lets just say this one?

This particular print takes around 200g of filament, which you will knows after you slice it. I bought this Bambu Lab PETG at SGD17.30/kg, hence it will be $3.46. If you consider the negligible electrical bills too, I will say it comes down to less than $4. I design this part in just 1-2hrs as I exactly knows what I want. If you also want to consider your own time and effort, it can be potentially in tenths or hundreds of dollar depending on how much you want to calculate your value.

Generally unless you are going for really large multiple KGs worth of filaments, and if you just stick with either PLAs or PETG, it can be very cost effective. On the other hand, if I will to do this using PCTG, that's going to cost me $62.70/kg, hence the same part will be $12.54 or I will say less than $13 in totality.

The part may look big, but because it is mainly hollow inside and even if it is a solid cube, it still can be quite hollow and rather strong when you are using say <25% of sparse infill, so it is quite inexpensive to print with 3D printing. But if this part is mass produced using mould injection using better plastic for stronger part, it can be really dirt cheap at cost. Retail might be different.

Hence personally I don't really look at it from material cost POV. That's quite an incorrect way for my use case. My main question is "Can I buy this item off the shelf, or how much will it cost me to outsource to someone to DIY manufacture for me ?" Normally the answer is rather straight forward. I'm probably the only person that need it in the world.