3D printing discussion

[De Ranged]

Gidday 

With all the time off I've been looking at 3D printers.... this is something that both me and the miss's have been a bit interested in 

I've reached a piont on a build where I've got to build a set of headers for an LS and given how bigger job this is I was going to buy a Header Molding Kit from ICEngine works, awsome set of tools.... but there is a slight hitch with them.... they are imperial and my steel supplier stocks metric pipe (they can supply imperial at a marked increase of cost and have to buy whole sticks, that means I end up with excess stock in the rack as "odd size" material), the next issue is I've currently got a 2L 4 pot, 3L V6, 4.2L Staight 6 and the LS and all of them need headers that are going to be tight, now I could get away with using a small dia set for all of them but it I know this is going to lead to some "band aid" engineering due to clearances not being right lol.... and given these sets arn't cheap lol not even close to it! I'm not buying more than one..... then I had the thought well how much is a 3D printer lol surprisingly its going to be cheaper so the plan is to model my own the right size and I learn something new..... my misses wants to get me into recycling plastics and with a bit of work this gives me an outlet for the recycled stuff 

So being in lock down I've spent a couple of days watching Youtube and surfing forums etc I'm about to pull the pin on buying a printer, an Ender 3 Pro... 

It's cheap..... yes there are alot cheaper options but I don't want to be "problem" solving a kitset 

It has a resumble bed size 220 x 220 x 250mm 

With some tuning work (and some mods) will produce some very nice work.... off the bat its good enough for what I want to do, I can make it better as I learn more about it   

Can print in PLA, PETG and ABS but this one is a bit of a funny one, it needs some mods to do this properly (nothing challenging to build).... PETG is the one that I want for the recycled plastic 

This is a good one..... it is modifiable and has a good community support network 

I thought I'd put this up for advice.... see if any of you have experiance in 3D printing and any advice, before I pull the pin

 

[Blanco]

This is interesting,... I can't help with the core request for experience because I don't have any yet, but I had no idea that printers were down to this level of price already, ....that in itself is food for lockdown thought.

So many questions follow,.... what are the parts modelled in? can such a printer read different softwares,... can existing parts be scanned etc. etc. 

Don't let me derail the topic, but what a great prospect.

[elbekko]

I bought a DaVinci Jr last year on sale, and it has actually made me a couple of useful parts. It's not easy tuning for dimensional accuracy though, don't expect a part you print to come out within 0.1mm of what you modelled unless you put in a lot of work tuning the printer and software.
On mine PLA works fine, but PETG just clogs the nozzle.

Also, print in a well ventilated room. I've felt not so well after printing PLA, but that might just be because it's biodegradable and they included some sort of mushroom product (which I'm allergic to).

[landroversforever]

12 minutes ago, elbekko said:

I bought a DaVinci Jr last year on sale, and it has actually made me a couple of useful parts. It's not easy tuning for dimensional accuracy though, don't expect a part you print to come out within 0.1mm of what you modelled unless you put in a lot of work tuning the printer and software.

IMO that entirely depends on the printer. Can't remember the model of the ones my old department use, but the parts that come off them are very accurate straight out the box. I want to say they're Witbox2 models, but not sure. 

[FridgeFreezer]

From what little I've done, setup is everything - getting the thing square, the bed level, and having a rigid printer - a lot of designs seem to assume that unlike a CNC machine, there's no load on the head so it doesn't need to be rigid... wrong! You can see wobbles and inaccuracies in the print layers.

Also there's a lot of fine-tuning in the settings that can make a huge difference, but take a lot of farting about to dial in perfectly - and can change when you swap filaments.

[Turbocharger]

I'd suggest subbing out the fiddling if you're looking for products and not a project. A friend runs a 'farm' of printers and he turns CAD files into objects for other people, then posts them out. His focus is on the production and dimensional accuracy, getting the layers to bind etc, and the risk is his when it chews a reel of filament or spoils a print at 95% progress.

His work comes through 3D Hubs, so you can just chuck your CAD file at them. You'll need a fairly large printer for exhaust bits I'd expect, so probably cheaper to rent than buy. Need to be honest with yourself - are you looking for a hobby (the process of printing it) or the output (the finished bits)?

[elbekko]

5 minutes ago, Turbocharger said:

You'll need a fairly large printer for exhaust bits I'd expect

I think he wants to print those small pieces that clip together and allow you to make an exhaust that you can then replicate in metal. Those individual bits should be doable on a 3D printer.

[landroversforever]

4 minutes ago, elbekko said:

I think he wants to print those small pieces that clip together and allow you to make an exhaust that you can then replicate in metal. Those individual bits should be doable on a 3D printer.

They are! I’ve got a pile of bits stashed away that I printed at work.

[lo-fi]

I bought the cheapest of cheap Prusa I3 clone a few years back, it's been brilliant. Don't get the absolute best quality prints, but perfectly good enough considering the price! 

The Ender looks great, I'd say go for it. I would have if it had been available at the time. 

Few notes:

The machine is dumb as dumb. The slicing engine to generate the tooling paths you use makes a huge difference. My preference, for what it's worth, is Ultimaker Cura. 

Use good quality filament. Over here, I swear by rigid.ink.

I upgraded mine with a new brain. Turned a mediocre printer into an great machine. Quieter too using clever stepper motor drivers. 

To be honest, once you've got a handle on how it works I reckon you'll knock up a welded metal frame to mount the hardware in. These things are marketed at "makers" who are usually not engineers or metalworkers, and certainly don't have access to a welder... Rigidity and having everything orthogonal, as Fridge mentioned, are key to quality. 

[De Ranged]

7 hours ago, FridgeFreezer said:

From what little I've done, setup is everything - getting the thing square, the bed level, and having a rigid printer - a lot of designs seem to assume that unlike a CNC machine, there's no load on the head so it doesn't need to be rigid... wrong! You can see wobbles and inaccuracies in the print layers.

Also there's a lot of fine-tuning in the settings that can make a huge difference, but take a lot of farting about to dial in perfectly - and can change when you swap filaments.

That is the what I got off youtube.... there seems to be alot of youtube channel support on how to setup and optimize the Ender 3, it was part of my decision to go with this model 

 

6 hours ago, Turbocharger said:

I'd suggest subbing out the fiddling if you're looking for products and not a project. A friend runs a 'farm' of printers and he turns CAD files into objects for other people, then posts them out. His focus is on the production and dimensional accuracy, getting the layers to bind etc, and the risk is his when it chews a reel of filament or spoils a print at 95% progress.

His work comes through 3D Hubs, so you can just chuck your CAD file at them. You'll need a fairly large printer for exhaust bits I'd expect, so probably cheaper to rent than buy. Need to be honest with yourself - are you looking for a hobby (the process of printing it) or the output (the finished bits)?

I have looked into this for a product I was going to market, but I haven't been able to find anyone in NZ, I imagine they are there, but I just haven't had any luck finding them.... what are 3D hubs? 

This isn't about a product. Its about learning (I want to get into CNC), convenience (the bulk of what I'm going to want is custom one off bits that I'm going to want as I build something) I want it as soon as possible.... I imagine it will be simialr to profile cutting where they buy in bulk and there process is optimised to the piont they can produce the parts cheaper than I can buy the materials. I imagine in the future I will come up with stuff that will benifit from farmed out work, but not for what I'm planing at the moment. My partner also wants to get into recycling and reusing plastics. HDPE mainly as there is currently no market for the coloured version here and it is all dumped... 

6 hours ago, elbekko said:

I think he wants to print those small pieces that clip together and allow you to make an exhaust that you can then replicate in metal. Those individual bits should be doable on a 3D printer.

Lol If you've ever made headers, this system from ICEngines is awsome.... each piece is 1" and they clip together like lego so you can tune your lengths of each runner by making each runner the same number of blocks, the blocks come in different radius's as well as straight (to match readily available pre-made bends) and they even have alignment marks to make cutting and mitering simpler 

 

6 hours ago, lo-fi said:

The machine is dumb as dumb. The slicing engine to generate the tooling paths you use makes a huge difference. My preference, for what it's worth, is Ultimaker Cura. 

Use good quality filament. Over here, I swear by rigid.ink.

I upgraded mine with a new brain. Turned a mediocre printer into an great machine. Quieter too using clever stepper motor drivers. 

To be honest, once you've got a handle on how it works I reckon you'll knock up a welded metal frame to mount the hardware in. These things are marketed at "makers" who are usually not engineers or metalworkers, and certainly don't have access to a welder... Rigidity and having everything orthogonal, as Fridge mentioned, are key to quality. 

Yea, I've noticed this off Youtube and also noticed that different softwares do different things..... from the different reviews and watching what others are using I figured I'd download both the Cura and Pursa (sp?) ones and spend some time with them before the printer arrives 

That is another thing I've noticed on Youtube the changes in filament and have noticed that some of the prints even changing colour (within the same manufacturers range) has had an effect on qaulity.... to that end I've done some research on the local market and found a supplier who is recomended for filiment..... I'm also going to run temp towers for each of the products as part of my learning curve lol 

Lol its a give in that I'm going to modify and play with it as I learn.... I've already ordered some common "hop ups" for it lol 

I was planing on buy this after our "lockdown" is over and had found a good deal from a NZ site (same one I'm getting my inital filiment from, but because they aren't essential they couldn't send till after the lockdown) they had a special on that was slightly better price than even buying it from China.... went onto there website to check out some details and its no longer available lol...... there is a new model that is going to be available at the end of May and $100 more lol so I pulled the pin on the Chinese one from Aliexpress.... $300NZD and another $278 in freight and Grab Snatch n Take.... now I just have to wait n see how long it takes to get here (given the world situation lol) they are estimating between 9-17 days hmmmm

I will be ordering a Glass bed (with a special texture for ABS), some PLA, ABS filament and the rest of the common mods later tonight 

I don't plan on playing with the ABS straight away as it seems to be a bit of a tempermental product lol and I want to build a few mods for the unit and an enclosure before I pull out the ABS 

I have found a number of different Youtube channels that have some very good information both for the Ender 3 model and 3D printing in general I can post up some links to these if people would like 

Now I have to wait for toys to arrive lol  

[De Ranged]

Lol I hate AliExpress...... I spent hrs last night on there to find the best deal for this printer 
I go on there today to order some of the accessories I was going to get from a local outfit (just updated there website to say out of stock lol) being lazy I hit the same search as last night..... and the very first option that pops up on the list is my printer free delivery $100 less than I paid lol..... bugger!

[Cynic-al]

My first one was a reprap. That worked in a fashion but was definitely the project version. I had to heat the bed to get the material to stick and fiddle with every job. There was no support structure so you had to design the part for the limitations of the printer. The software for slicing was really flaky and after a java update on the pc it just died altogether and I lost interest.

Based on that bad experience I got an objet 24. It's different in terms of it's a polyjet not an fdm and it prints with heated heads in a heated cabinet. Accuracy is 0.1mm, layer thickness is 28 micron and you just load your stl in and it does everything for you. It can only do 1 material which is expensive and a little limiting. 

I would be interested to hear how you get on with your machine, I have no experience. There is an exhibition in Frankfurt once a year full of them, very impressive, but a bit of a trek!

My advice would be to use this time to pick and learn your cad software first. That can be done for free. You need to be able to design the part on that before you can print it. You can always send it to someone like shapeways for a price if you want to see what it should look like before you buy a printer. 

[Turbocharger]

16 hours ago, De Ranged said:

what are 3D hubs? 

Exactly as you've surmised - it's mail-order fabrication. Email a CAD file, receive your bits. The cycle time will certainly be faster with your own printer (esp depending on where they're printed and posted from!), but there will be more messing about time, certainly to start with.

https://www.3dhubs.com/

[De Ranged]

13 hours ago, Cynic-al said:

My advice would be to use this time to pick and learn your cad software first. That can be done for free. You need to be able to design the part on that before you can print it. You can always send it to someone like shapeways for a price if you want to see what it should look like before you buy a printer. 

The CAD software isn't too much of a stretch.... I have to use it to a lesser degree for work, the company I work for (hopefully I've still got a job at the end of the virus lol) is international and comunicating technical design issues is like beating your head against a wall so I've taken to Modeling things in Solidworks.... I won't say I'm proficient, but I am good enough to be able to work out how to achieve the "odd" things with a bit of research and sometimes alot of swearing lol

The biggest hurdle I'm going to have to deal with is this is a medium I've never really play'd in before and its broad lol there are heaps of different plastic types with different properties and even different performance with type due to heat etc so I've got to learn materials... preperations etc. Then I'm going to have to learn how thick a wall thickness needs to be for a given part in a given material to achieve the end result I want 

For example these header jig pieces I want to make 

Notice that they clip together, now consider I want long headers (more torgue) so I'm going to end up with long chains of these clipped together and I don't want them "opening up" at the joints as this will change the bend radius's so when I build them out of steel, things won't line up..... so I want that clip together joint strong and tight. I also need these to be as light as possible so there own weight doesn't become a problem. I need the clips able to be clipped in and out lots of times so a good flexabilty and not brittle 

So how thick do I make each part of this model lol do I use infill, what material do I use, what temperature 

This is what I'm researching at the moment and for anyone interested Stefan from "CNC Kitchen" has some very good videos on Youtube on just this  

 

8 hours ago, Turbocharger said:

Exactly as you've surmised - it's mail-order fabrication. Email a CAD file, receive your bits. The cycle time will certainly be faster with your own printer (esp depending on where they're printed and posted from!), but there will be more messing about time, certainly to start with.

https://www.3dhubs.com/

Just run a search on 3dhubs in NZ lol interesting reading..... it once was a community a world wide community where you could go in as either a builder (with different tools 3d printers, castings what ever) or as a client and request tenders.... now it is just a number of companies in the Europe/UK 

On a plus note it has given me alot more search hits than say cnc or 3d printing.... cheers for that 

[landroversforever]

4 minutes ago, De Ranged said:

For example these header jig pieces I want to make 

I can't remember where I found them, but there are designs for them readily available to just print them out  

[De Ranged]

26 minutes ago, landroversforever said:

I can't remember where I found them, but there are designs for them readily available to just print them out  

This might make a few people laugh, but it doesn't feel right to just grab the plans off the net lol now if there were lots of versions and different people making them then.... to be honest I wouldn't have a problem with it. 

It will be a good learning curve for me as well..... and just at the moment I don't really have alot else to worry my self with lol 

[Cynic-al]

Material wise pick one and stick with it ir your making it hard for yourself. Abs is a proper plastic but not the easiest to work with. I think the printer would have to be well setup for a tight fitting spring clip. A bit of reverse taper on the clip will help pull them together. Tolerance the fit so the plastic isnt fully relaxed so it's still putting a bit of tension on the joint. Or could you make in 2 halves so it doesnt need to spring as much and you can have a bigger overlap? Or a screw down the middle so you can get them in position then clamp them tight together. Or just get a length of welding wire and bend that. 🤣

[landroversforever]

https://www.thingiverse.com/thing:1649639 I think it was one of these versions. They did a different one for the 3” stuff I printed. 

[FridgeFreezer]

I rekcon you'd be better off changing the design to suit the material, a tight clip-together thing like that from a precision injection mould is going to be hard to replicate to close enough tolerance - but there's likely other ways of doing it that would work just as effectively.

One thing I'd like to try is lost-PLA metal casting - print the thing and then plop it in sand, pour in molten metal and have a very solid part.

[landroversforever]

3 hours ago, FridgeFreezer said:

I rekcon you'd be better off changing the design to suit the material, a tight clip-together thing like that from a precision injection mould is going to be hard to replicate to close enough tolerance - but there's likely other ways of doing it that would work just as effectively.

One thing I'd like to try is lost-PLA metal casting - print the thing and then plop it in sand, pour in molten metal and have a very solid part.

It’s not actually too bad when the design is right... 

[De Ranged]

Thanks for that 

20 hours ago, landroversforever said:

https://www.thingiverse.com/thing:1649639 I think it was one of these versions. They did a different one for the 3” stuff I printed. 

I downloaded his design I'm going to use it to get the tolerance for the clips, but there is one thing I want to change and your photo shows this, if your doing tuned length headers you want to know the length of each runner.... the easy way is to work off the center line of each block just looking at his models the center line radius length isnt consistant.... I might be wrong I haven't pulled up dimensions on them and the link was the wrong size for me lol, I don't know if I'll go with clips or not.... but I'd like to know what the tollerance is for future projects lol 

21 hours ago, Cynic-al said:

Material wise pick one and stick with it ir your making it hard for yourself. Abs is a proper plastic but not the easiest to work with. I think the printer would have to be well setup for a tight fitting spring clip. A bit of reverse taper on the clip will help pull them together. Tolerance the fit so the plastic isnt fully relaxed so it's still putting a bit of tension on the joint. Or could you make in 2 halves so it doesnt need to spring as much and you can have a bigger overlap? Or a screw down the middle so you can get them in position then clamp them tight together. Or just get a length of welding wire and bend that. 🤣

I've been doing a lot of reading and watching.... I'm going to start off with PLA it is the strongest (not the toughest) and easist to learn on, I've bought all the bits I need except the enclosure which I'll build, when I have the printer setup with all the mods that the community suggest.... I'll play with ABS I think long term this is a better plastic for my applications 

I like the latteral ideas for securing

Screws nice n strong and can rotate.... catch is, say I want to change the one 26 blocks back and add in a straight, I then have to unscrew all of them to change that one 

Welding wire.... this isn't actually a bad idea, feed it down the center of all the blocks and kink it to stop them "pulling" apart its alot quicker than screws but it is still a bit auckward to chop n change in the middle..... I've thought about all sorts of ideas similar to this.... bungy cord, a mig gun liner with a locking clip I could slide down and lock etc but I keep thinking that if I start into too many bends this could collapse them and its still auckward to chop n change them 

What I was thinking is magnets, Neodymuim magnets but with a center pin...... possibly do the pins with a slight taper like a Morse spindle and the magnet at the end this would make them very ridged and still allow them to rotate 

7 hours ago, FridgeFreezer said:

I rekcon you'd be better off changing the design to suit the material, a tight clip-together thing like that from a precision injection mould is going to be hard to replicate to close enough tolerance - but there's likely other ways of doing it that would work just as effectively.

One thing I'd like to try is lost-PLA metal casting - print the thing and then plop it in sand, pour in molten metal and have a very solid part.

Lol I dont' have an injection moulder.... I might build a small one long term as part of the miss's recycling interest, there are some real simple plans on the internet 

Now the Lost plastic metal casting is something I'm interested in but not for this one lol I might end up with 400 pieces in this kit to do say a straight 6 header.... to cast them all lol I like making work for myself but thats over the top!

 

[FridgeFreezer]

12 hours ago, De Ranged said:

Lol I dont' have an injection moulder.... I might build a small one long term as part of the miss's recycling interest, there are some real simple plans on the internet

I meant the original kit you referenced is injection-moulded, hence the tolerances will be waaaay tighter than a 3D printed version - and the plastic type will lend itself to snapping together far better than PLA.

[De Ranged]

10 hours ago, FridgeFreezer said:

I meant the original kit you referenced is injection-moulded, hence the tolerances will be waaaay tighter than a 3D printed version - and the plastic type will lend itself to snapping together far better than PLA.

Given my limited knowledge on plastics...... I don't think it will be a internal "die" moulded, the hollow inside is too large I'd guess its blow mould or that one where you spin it... I think you'd be surprised on tollerances if it was internally die moulded then yes but if it is either of the other two your wall thickness is going to be inconsistant and therefore your shrinkage becomes a variable as for tolerances of 3d printers mine is advertised as within 0.1mm so that means a +/- 0.05mm tolerance..... most of my machining isn't to that level lol...... personally I see that as debatable lol, but these are sold and advertised with this tolerance in America, given there liabity law, they wouldn't be advertising it if it wasn't possible... and I've seen multiple examples of interference fits that have been modeled and printed on Utube and this is on a sub $200USD printer 

5yrs I've been drolling over these header blocks lol.... now back then the company that made them advertised on there website that they were 3D printed in ABS  

The PLA is mainly for me to learn on..... for most of the sort of automotive and shed tool stuff I see me doing with this, ABS is a better option, but for me to do this reliably on this printer the information I've found (generally, internet isn't the most reliable lol) is that I need to modify the printer for better heat control.... insulation, different bed, better fan control and an inclosure with the power source outside of the inclosure is required... sorting all of this I'm going to give it some time, time for me to learn some of the "slicing" and modeling limits playing with PLA

Then when I've got all the mods done for ABS I'll make these in ABS.... but I'll probably prototype in PLA since I can play with that pretty much straight away lol 

 

 

Edited April 11, 2020 by De Ranged

[Cynic-al]

If you had a central cord (bungee or welder liner etc) you could have them clip on so you could take the middle one out. 

The thing you might come across when trying to print fine tolerance clips is the layer thickness. How much it spreads as it comes out the nozzle and the layer thickness giving you a staircase on the taper and lower strength (as your putting hot material onto cold the joint isnt as strong, that's why some people heat the cabinet, that and to control shrinkage as it cools)

We injection mould hollow sockets at work with an internal socket range of 5mm yo 320mm and wheelie bins are injection moulded so it's quite common to do hollow parts. 

Edited April 11, 2020 by Cynic-al

[De Ranged]

 

2 hours ago, Cynic-al said:

If you had a central cord (bungee or welder liner etc) you could have them clip on so you could take the middle one out. 

The thing you might come across when trying to print fine tolerance clips is the layer thickness. How much it spreads as it comes out the nozzle and the layer thickness giving you a staircase on the taper and lower strength (as your putting hot material onto cold the joint isnt as strong, that's why some people heat the cabinet, that and to control shrinkage as it cools)

We injection mould hollow sockets at work with an internal socket range of 5mm yo 320mm and wheelie bins are injection moulded so it's quite common to do hollow parts. 

Got some models I'll throw up when I'm finished, just daft ideas other than clips, including a magnetic clip together to work around the bungee cord 

Yes I've noticed layering is a big thing on flater curves due to the step... I've also found slicer software that will reduce line height down in these situations to improve apperance/accuracy, I've also watched a fair bit on line width and playing with settings for different applications (there is an Aussy lad with a channel on Youtube that explains alot of it really well).... the diference heat makes in strength with ASA and ABS vs appearance, strength in line or by layer (roughly 50% less when pulling layers apart vs pulling a string) so orientation of your model makes a big differnce 

One thing that I'm curious about that I haven't been able to find is the use of CoreXY style printers (the bed only has vertical movement) vs H bot printers (the ender 3 I'm getting) regarding strength.... because they have less mobile mass the printing heads can move alot faster with less shake, does that increased speed change layer adhession

I think from a personal stand piont I think I'd throw more wall layers in at the start with my stuff.... if weight and material cost isn't an issue why not over engineer lol 

 

You wouldn't have any pictures of die's that are used for something like this, ones that can shrink down to come out a past a shoulder smaller than they cast.... the engineering on something like that would be cool.... I'm guessing an "iris" design that expands as you twist