Best Printer to Build from Scratch 2015

Hi,

I am looking to build a printer from scratch.... No kit.

I have access to a desktop CNC router 600x300mm approx, and an Ultimaker 2 at work.

I also have access to stepper motors and other electronics, sheet metals and plastics and extrusions rod and bar metals.

The enjoyment for me is mostly in the building of the printer from self sourced parts. However I would still like a printer relevant in 2015 as far as accuracy, quality, speed and availability of materials go. Ideally I would like the build area of the Ultimaker or bigger. I am considering both delta and Cartesian models.

What is the best up to date option for me?

Many thanks

Dave

My totally biased suggestions:

Cartesian: P3Steel (see the wiki)

Delta: Delta Steel (see the thread in the delta forum)

If going Cartesian route, Mendel90.

The MendelFlex with or without the flex would be wholly biased suggestion.

Unbiased would be a Herculien or Eustanthios <<< one of these will be my own next build

Go with delta design!

Just take one look at them printing and you know its what you gotta make one. [www.youtube.com]

I was thinking a Lulzbot. Though I am not sure how much help the CNC would be.

Having built both Cartesian and Delta, I recommend Delta. I found it more stable, and easier to set up to get accurate prints. The footprint is also smaller for the same print area.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Delta printers are fascinating to watch... they can also be a challenge to set up as they are harder to understand that the traditional Cartesian printer, with every motor moving when the effector moves. This can also be a challenge for the controller board, and they work best with a modern 32-bit board. They can be very quick. If you're wondering, I've just finished making one, and trying to iron out the final bugs.

CoreX-Y printers look interesting, but I can't say that I know much about them.

Nice thought David, CoreXY would be an interesting build for him and let him utilize the CNC machine. He might also look at RichRap's Sli3DR, though I am not sure if it is a true CoreXY variant.

Quote
David J
Delta printers are fascinating to watch... they can also be a challenge to set up as they are harder to understand that the traditional Cartesian printer, with every motor moving when the effector moves. This can also be a challenge for the controller board, and they work best with a modern 32-bit board. They can be very quick. If you're wondering, I've just finished making one, and trying to iron out the final bugs.

I may be accused of blowing my own trumpet, but a delta printer using Duet electronics with my firmware fork and a reliable Z probe is trivial to set up, because of the fast autocalibration function. The build still needs to be accurate in some respects; for example the diagonal rods must all be the same length as accurately as possible, and backlash in the joints must be minimised.

Delta printers have other advantages besides speed - for example, a smaller footprint for the same build area compared to Cartesian (but greater height). Also, the fixed bed of a delta makes it easier to use FSRs for Z probing, and makes it safer to use a mains-powered bed heater.

I can't comment on CoreXY as I have yet to build one.

Edited 1 time(s). Last edit at 04/23/2015 05:09PM by dc42.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Quote
dc42
Delta printers have other advantages besides speed - for example, a smaller footprint for the same build area compared to Cartesian (but greater height). Also, the fixed bed of a delta makes it easier to use FSRs for Z probing, and makes it safer to use a mains-powered bed heater.

The biggest difference I've seen already is that there is very little mass moving around in a delta, which means that the frame doesn't move when printing. Compare this to my Prusa i3 that has the wade extruder mounted above the hot-end on the X carriage - when everything's working hard it threatens to move the trolley it sits on around the room! It's a rickety trolley too, which means that all the joints are flexing when printing.

Edited 1 time(s). Last edit at 04/23/2015 05:19PM by David J.

Quote
dc42

I may be accused of blowing my own trumpet, but a delta printer using Duet electronics with my firmware fork and a reliable Z probe is trivial to set up, because of the fast autocalibration function. The build still needs to be accurate in some respects; for example the diagonal rods must all be the same length as accurately as possible, and backlash in the joints must be minimised.

Delta printers have other advantages besides speed - for example, a smaller footprint for the same build area compared to Cartesian (but greater height). Also, the fixed bed of a delta makes it easier to use FSRs for Z probing, and makes it safer to use a mains-powered bed heater.

I can't comment on CoreXY as I have yet to build one.

I must admit; I am more tempted by a Delta printer as I have seen some very good arguements for them, as long as I can use dual extrusion to print different materials and it as 'future proof' as an FDM model can be. What are you using at the moment if you don't mind me asking?

Quote
Dave_robinson_022
What are you using at the moment if you don't mind me asking?

I have two 3D printers. The first one was an Ormerod 1 kit from RepRapPro. I upgraded it in a number of ways, including a dual extruder conversion that I describe on my blog. Here is the Ormerod, and a print from it:



The second started out as a Mini Kossel built from a kit. I converted it to Duet electronics (also described in my blog) and added the colour touch screen control panel. I found it better than the Ormerod in most respects, except that it was single extruder and has a smaller build area. So I have recently increased the build area to more than 300mm diameter by using longer horizontal extrusions and fitting a 350W mains-powered heated bed.



It is printing well, but I have just ordered some 1m vertical extrusions so that I can increase the build height. After that, I intend to convert it to dual extrusion, using the E3D Chimera or Cyclops hot end.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Quote
dc42


The second started out as a Mini Kossel built from a kit. I converted it to Duet electronics (also described in my blog) and added the colour touch screen control panel. I found it better than the Ormerod in most respects, except that it was single extruder and has a smaller build area. So I have recently increased the build area to more than 300mm diameter by using longer horizontal extrusions and fitting a 350W mains-powered heated bed.

2015-04-15%2019.37.15.jpg

2015-04-20%2015.24.03.jpg

It is printing well, but I have just ordered some 1m vertical extrusions so that I can increase the build height. After that, I intend to convert it to dual extrusion, using the E3D Chimera or Cyclops hot end.

I find that idea very interesting; I have read some of your blog just now and there is some really great informations; thanks.

Will you be adding the upgrades to your blog?

Kossel Mini uses openbeam which is 15mm x 15mm isn't it? Is this still ok at larger size or will you look to move to a 2020 extrusion?

Also, I have noticed issues with airflow (drafts etc) around the Ultimaker at work so I was considering an enclosure; it occurs to me that if I fix the frame to a larger enclosure then this would improve stability and rigidity?

The only other issue I have is availability of some parts; moving from the UK where I had easy access to anything I needed for building things to NZ has been a bit of a shock in that respect. I was however considering a heated mat on aluminium with a PEI sheet for a heated bed as I was hoping to ho bigger than the 200x200mm PCB Heated beds anyway. What are your thoughts on this?

Cheers,

Dave

For deltas look at this designs for the bigger build area for deltas:

20x40 Extrusions Look at the cherry pi 4 xl design for the bigger size design -> [forums.reprap.org]
Or look at the cherry pi steel -> [forums.reprap.org]

All you can modify or make into your own delta, or even use the Mini Kossel design and upscale it.

Quote
Dave_robinson_022
Will you be adding the upgrades to your blog?

Probably, I have a draft entry already, just need to get around to finishing it.

Quote
Dave_robinson_022
Kossel Mini uses openbeam which is 15mm x 15mm isn't it? Is this still ok at larger size or will you look to move to a 2020 extrusion?

I'm keeping the 1515 horizontal extrusions, but switching to 2020 for the longer vertical ones.

Quote
Dave_robinson_022
Also, I have noticed issues with airflow (drafts etc) around the Ultimaker at work so I was considering an enclosure; it occurs to me that if I fix the frame to a larger enclosure then this would improve stability and rigidity?

The only way I managed to print the frame motor mounts in ABS for my Midi Kossel conversion on the Mini Kossel was to put a couple of plastic bags over the whole printer to increase the temperature around the print. So an enclosure is high on my list of priorities too. Whether it will hep with rigidity or not will depend on what it is made from. I am not planning to make the enclosure structural, I'm thinking of drop-in acrylic or polycarbonate panels, double skinned if necessary for thermal insulation.

Quote
Dave_robinson_022
The only other issue I have is availability of some parts; moving from the UK where I had easy access to anything I needed for building things to NZ has been a bit of a shock in that respect. I was however considering a heated mat on aluminium with a PEI sheet for a heated bed as I was hoping to ho bigger than the 200x200mm PCB Heated beds anyway. What are your thoughts on this?

I haven't tried PEI sheet. If you stick it directly to the aluminium, then the aluminium needs to be very flat. I suggest you try PEI on top of glass first, and only consider sticking it directly to the aluminium when you have more experience of it. Using a glass print bed has another advantage: you can have several of them, so that when one print has finished, you can remove the bed plate, substitute another one, and start printing again - while your first print is cooling down so that it can be removed from the bed. I don't bother with borosilicate glass, ordinary float glass is OK as long as the bed heating is even.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Quote
dc42

Probably, I have a draft entry already, just need to get around to finishing it.

I shall keep an eye out for it then and look forward to seeing it.

Quote
dc42

The only way I managed to print the frame motor mounts in ABS for my Midi Kossel conversion on the Mini Kossel was to put a couple of plastic bags over the whole printer to increase the temperature around the print. So an enclosure is high on my list of priorities too. Whether it will hep with rigidity or not will depend on what it is made from. I am not planning to make the enclosure structural, I'm thinking of drop-in acrylic or polycarbonate panels, double skinned if necessary for thermal insulation.

Would you heat the chamber as well or just use the bed to heat the area, what are your view on a heated bed and chamber?

I'd try just using the bed heater initially, with a thermistor inside the chamber to monitor the temperature. If I need more heat then I'll add a heater, probably a mains powered one driven by an SSR just like the bed heater.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

What temperature would you look at achieving inside the chamber and do you think this would have any issues with the electronics or stepper motors? Have you had any issues with the SSR Heatsink?

I'll bp looking for 60C in thr chamber. I don't expect any issues with the steppers, I am running them at around half their rated current, so they run cool at present. I will keep the electronics out of the chamber. The 10A bed SSR only passes around 1.5A @ 240V so generates about 3W of heat, which is not a problem. DC SSRs passing 20A of bed current at low voltages are a different matter.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

True, well it will be good to see how it turns out. I don't suppose I can see any rough work you have on your project? I am assuming you will need re-print parts to take larger extrusions?

Yes I will need to re-print the vertices to take larger extrusions. I reprinted the lower vertices once already.

I'm still thinking about the best way to make the enclosure. Drop-in panels would work well for my current machine, but not when it is 1m tall. I need to be able to remove the enclosure when I print PLA. So I'm now thinking of using 3 acrylic or polycarbonate panels, probably double skinned, permanently connected along the long edges by two piano hinges, and printed plastic parts on one of the remaining long edges with slots for the final edge to fit into. If I design this carefully, it can fold flat when not in use. Then a triangular hat to go on top, probably a piece of plywood with a few printed parts to help locate it on the side panels..

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Are you using clear plastics for visibility, else would insulated MDF not be easier at 1m with a clear window?

Could you make the case permanent but use a fan to actively cool for lower temperatures?

Yes, clear plastic for visibility. I might compromise and use MDF for 2 sides and clear plastic for just one. Not sure I want to make the case permanent, maybe I will if/when I stop tinkering with the printer.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Do you have a BOM and build instructions for the original kit you used?

The original kit was a Mini Kossel from Think3DPrint3D. You can find the build instructions on their site. One improvement I suggest (apart from better electronics) is to use a better carriage design that grips the belts without the use of cable ties. There are several such designs on Thingyverse.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

Have you had to increase the rod size for the arms yet when you incresed bed size? if not I take it you will for the extra height? If this is so will you increse the diameter of the carbon rods as well?

Thanks

Dave

I kept the same size of carbon fibre rods and they seem to be working ok. I think it may be a good idea to increase the spacing between the diagonal rods in order to increase rigidity, so I will do this when I redesign the effector. The height has no bearing on how thick the rods need to be, only the length does.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

I was just wondering if the increased length would decrease stability in the rod and that could be a reason for increasing diameter? especially if you were to add dual extruders. what diameter are they now?

Ok, following all of the great advice I have looked into things a bit further, a lot is limited by what is readily available here in NZ, especially when it comes to electronics.

I want to build a Kossel Style Delta (mini but bigger) I am looking at using 500mm Openbuild 2020 V-Slot for the horizontals and 1000mm Openbuild 2020 V-Slot.

I calculate this will give me 320mm Diameter printable circle with a 230mm printable square inside this up to a height of of 430mm. This should be more than enough for anything in need to print.

I also calculate needing 400mm arms, but I am hoping to redesign effector to allow dual extrusion from E3D Chimera (If I can get it here)

Does this sound over-ambitious in either calculations or stability?

Does anybody have experience building with Openbuilds V-Slot?

As a result of this I will need to redesign most printed parts; I have read somewhere that carriages should only run on 3 rollers to prevent inaccuracy but I cant remember where, or why? Openbuilds do a machined aluminium plate that hold 4 mini-wheels, I thought this may be better and then to bolt rest of carriage to these?

When I looked into electronics I looked at Duet and PanelDue but they are only availble in the UK. I get the feeling that 32bit processors are the way forwards and that Arduino based is very much outdated? Again I refer to my original criteria of future proofing the build. Does anyone have any experince with BeagleBone Black and Replicape? The processing power is there however I am unsure about the firmware Redeem and its options for configuration, which seem limited?

Many thanks,

Dave