looking for a good 3d printer

[Ross77]

Im in the market for a good quality 3d printer. £500 if possible

I don't need a large work area but does need to be accurate.

Whats the best around?

Thanks in advance

[EddyCurrent]

I see what you said about size but check this one out especially youtube review videos http://www.ebay.co.uk/itm/HICTOP-CR-10-3D-Printer-Aluminum-Pre-assembled-Large-Print-Size-300x300x400mm-/272622069677?epid=733428808&hash=item3f798aa7ad:g:BaQAAOSwA29Y55JL
I have one and it's working just great. ( I didn't have the time or inclination to build one, for a change I just wanted to plug it in and use)

It does not have auto bed levelling but I don't find that an issue.

[Zeeflyboy]

what's your budget?

[Ross77]

Thanks Eddy
Might be a bit tall but will have a look.

Is it possible to convert a router type machine with the z moving up and down to a 3d printer or does it affect the print head?

[Zeeflyboy]

Routers don't make good printers, they really have opposite requirements in most respects.

Printers ideally need to be fast and light in the moving parts. It is of course possible to put a print head in a router but it won't make an ideal printer.

[Ross77]

Thanks. i have an engraver that is solid and about the right size so thought I could start with that. guess it would be a pain swapping over the heat table and print head etc.

Maybe have to dig out the belt driven stages I have kicking around and see what will work. ;0)

[EddyCurrent]

I can see you are going where I didn't want to go :distress:

[Ross77]

I know, probably end up buying one one and upgrading it like triggers broom. :whistle:

[cropwell]

I see what you said about size but check this one out especially youtube review videos http://www.ebay.co.uk/itm/HICTOP-CR-10-3D-Printer-Aluminum-Pre-assembled-Large-Print-Size-300x300x400mm-/272622069677?epid=733428808&hash=item3f798aa7ad:g:BaQAAOSwA29Y55JL
I have one and it's working just great. ( I didn't have the time or inclination to build one, for a change I just wanted to plug it in and use)

It does not have auto bed levelling but I don't find that an issue.

That's a damn clever printer to print the example they put on the print bed, with one extruder as well !

I have a Wanhao 4DS which is twin extruders, but a small print volume. If I bought one now, I wouldn't bother with two extruders, I hardly ever do two colour prints and when I do it is a PITA to get it right.

[Boyan Silyavski]

For the money i would buy the Steel Prusa. Its cheaper than the Original Prusa and build with quality elements/ the electronics/ . It has unsupported rails, whats typical for cheaper printers, but is not like some funny rollers, rolling on aluminum profile. I dont know of something really better, from an educated point of view. Yeah, many will say this and that, but from CNC point of view, i see that as the best cheap printer around

[EddyCurrent]

Boyan,
As you might expect, I spent days weighing up one printer against the other in the required price range. Thought about building one but didn't really want to, I looked at the links to sellers in Spain that you posted in another thread and was just about ready to order the black steel Prusa when I came across the Creality CR-10.
Now in cnc terms there are what we might describe as undesirable elements, single Z drive, rollers for bearings, etc. but in reality on a 3D Printer I'm finding it's not a problem provided repeatability is there.
Also with the steel prusa it was interesting to see the price rise as options were added on, so after many hours of reading and watching videos, my choice was the CR-10, it's size being another positive for me.

[Ger21]

These are supposed to be very good for the money.
https://www.amazon.co.uk/QIDI-TECHNOLOGY-Extruder-Desktop-Structure/dp/B01CU2MCM4/ref=sr_1_10?ie=UTF8&qid=1500813141&sr=8-10&keywords=3d+printer

I didn't think I'd have any use for a 3D printer, but may buy one of these.

[Boyan Silyavski]

In 3d printers there are so many things that may look mediocre but if they are tested and working it just works. At the same time so many things could look engineered properly but with a hidden fault . So the decision is always difficult. I have the Original Prusa MK2. I laughed at it when i opened the box and started to respect it when i fixed some little obvious faults. It just works, and works and works and will work. I rarely stop it at all. Anyway, i am sure there are many options nowadays. But i say it again, crappy electronics is what i hate most.

[cropwell]

These are supposed to be very good for the money.
https://www.amazon.co.uk/QIDI-TECHNOLOGY-Extruder-Desktop-Structure/dp/B01CU2MCM4/ref=sr_1_10?ie=UTF8&qid=1500813141&sr=8-10&keywords=3d+printer

I didn't think I'd have any use for a 3D printer, but may buy one of these.

Yes Gerry, put it side by side with the Wanhao 4DS - you will see the differences are just cosmetic, but if the price is better.............

My friends (correction - friend) said to me 'What are you going to use that for' when I bought the printer. Two days later he asked me to design and print a bespoke cup to hold coins in his Jag.

You get to the point with some things where the question is 'Shall I buy an item - or print it?'.

Some things you just can't buy https://www.dropbox.com/s/yv6cbor60shigfm/WP_20170304_19_42_28_Pro.mp4?dl=0

Cheers,

Rob

[Ger21]

Nice.
Never seen a 3D printed Lithophane before. :beer:

[Ross77]

Thanks for all the responses, lots of options to look at. leaning toward the QIDI for a complete machine but also looking at E3d for extruders as they seem really good.

If i was to build, is there a wish list of extruders, heat plates and controllers? not to fussed on looks as long as it will be fast and and produce quality parts. Ta

[A_Camera]

Routers don't make good printers, they really have opposite requirements in most respects.

Printers ideally need to be fast and light in the moving parts. It is of course possible to put a print head in a router but it won't make an ideal printer.

I think this statement is wrong and a bit simplified. Yes, printers need to be fast but no, they don't need to be light in the moving parts, why would they? But... hopefully I have first hand experience about this soon because I have just recently started a 3D printer conversion project for my own router and just recently ordered some parts which I think may be needed. I have no experience in 3D printing, but what I have seen is that it is in fact very slow. Just made a model of something which a 22mm tall 70mm diameter cylinder and test printed in Slic3r and according to that, it would take over one hour to print it. To mill it will take about 10 minutes. Of course, that's because all the void must be filled with plastic, but having a look at the G-code, it clearly shows that even most Chinese routers with moving gantry and a maximum speed of around 4000mm/min can easily keep up the pace. My router with 9000mm/min will definitely outrun all the most popular 3D printers in both speed and acceleration. But this is just theory so far. I still need to build the necessary parts together to be able to easily switch between 3D printer and router, since I don't want to permanently convert but want to switch between the two.

The only real advantage of a dedicated 3D printer I can think of is that routers are normally noisier than 3D printers and also much larger and heavier. But I don't think size and weight can be a disadvantage for a 3D printer, except that if you only want a 3D printer then it is unnecessary to make it heavy and large because it only makes it more expensive.

Anyway, this is going to be my autumn project, currently just collecting parts and playing with different software. Maybe I'll start a thread later on about it.

[Boyan Silyavski]

The advantage of a dedicated printer is that the thing runs 24h non stop for months and the router soon you will need for sth else. Print jobs are ridiculously long if you want a quality piece as a result

[A_Camera]

The advantage of a dedicated printer is that the thing runs 24h non stop for months and the router soon you will need for sth else. Print jobs are ridiculously long if you want a quality piece as a result

Yes, a dedicated higher quality printer is of course better for longer jobs or if the main activity of a machine is printing. If in the end I decide that I like the results out of my conversion I will build a dedicated printer, but for occasional printing and for testing the concept, the software, finding out pitfalls and so on, at least for now, I will go on with the conversion. Converting is a pretty simple and cheap task and most of the parts are needed even for building one, and in fact, even if I bought one, most parts can be used as spares.

Though, if you can stand the higher noise of a router converted printer, I don't see why that can't run as long as necessary. Generally it isn't a good idea to let a printer work unattended for too long, so non-stop running is at least for me, out of the question unless I install special fire guards and a sprinkler system above or inside the printer.

[Zeeflyboy]

The point is that while it will work, a router just makes a poor 3D printer compared to something that is designed with 3d printing in mind.

If its all you can do for space/cost/trying out the idea/whatever then fine, as mentioned it will function but an ultimaker or something will run rings around any router conversion in actual day to day usability, productivity and reliability long term.

The reason why I say light is because it's much easier to build something that is fast and can run for hours on end efficiently if it is light. Note that light does not mean flimsy - a rigid printer is a good printer, but it just doesn't need the same bulk as a router, but needs to run for many hours on end with a lot of fast movements... personally I'd much rather have a light machine with belts vs a heavy machine with screws doing that, both for my sanity from noise and from a wear/tear and energy efficiency stand point.

As for leaving it printing, sometimes you have no choice. I've had 20hr+ print jobs and there really is no viable way of pausing in the middle. I personally set my printer up with a monitoring webcam and a smoke alarm tied to a smart power socket that will shut down the power if the smoke alarm goes off, but I've been printing for many years without any safety issues - there are several failsafes built into the printers themselves to prevent thermal runaways etc so they can be relatively safe left to their own devices.

[JaviPK7]

I'd recommend a prusa www.prusa3d.com
I have the previous to the last model and works like a charm. With the new one they have won best printer of the year and looks like a really good printer on a budget

And very good customer service/ support

Cheers

[cropwell]

My router with 9000mm/min will definitely outrun all the most popular 3D printers in both speed and acceleration.
My Wanhao 4DS is currently set at a safe 150mm/sec (9000mm/min !) and it doesn't chuck itself all over the shop. Print speeds have to be lower as you need to feed filament into the extruder and melt it before squirting it out all over the workpiece in some sort of controlled fashion.

[charlieuk]

i just got a prusa mk2s and while it was a long wait and a bit of a headache once it arrived it didn't take long to put together and was printing great quality stuff straight of the bat. its been running nearly solid since it arrived with a few 26 hour prints oh and to add to this the 3d bed level and all the other auto correction stuff is amazing and so simple and the heated bed is also very good and no need for glue sticks just wipe with window cleaner and that's it.

[A_Camera]

My Wanhao 4DS is currently set at a safe 150mm/sec (9000mm/min !) and it doesn't chuck itself all over the shop. Print speeds have to be lower as you need to feed filament into the extruder and melt it before squirting it out all over the workpiece in some sort of controlled fashion.

I didn't mean that 9000mm/min was not possible with 3D printers, of course it is, and more than that as well, but the printing speed is ALWAYS lower than that in any affordable printers I have seen, meaning that the speed I have is more than enough and in fact, most routers have more than enough speed. Noise is of course an issue, 3D printers are not as noisy as routers.

[Zeeflyboy]

While there are guys that push speeds to crazy levels with volcano nozzles and the like, typical print speeds for me are around 70mm/s or 4200mm/min, non print moves are 180mm/s or 10800mm/min... of course given the light weight it's fairly easy to have very high acceleration figures too. There can be a surprising amount of non-print moves (both long and short) so don't discount their acceleration and speed as a factor in overall print time.

I would also suggest that first of all, most of the bog standard routers out there are not capable of reliably doing anything near your 9000mm/s and secondly since yours is fixed gantry you also have an advantage in acceleration over typical routers with a much heavier gantry to throw around. There is also the factor of z-lift to consider - a lot of people, myself included, will have a printer set to lift before each non print move by a small amount (I usually set it to 0.25mm) which help stop scraping the nozzle on already printed regions. That is normally a pretty rapid move so if you are having a weighty spindle still attached to avoid alignment issues each time you replace it, the acceleration of the Z could be another factor slowing down a longer print.

Even still, I think that multiple 20+ hour sessions of your machine continuously throwing itself around at 9000mm/s with acceleration to match a typical much lighter 3d printer is going to wear out your kit much more quickly than normal use of a router/mill due to the weight/speed/acceleration demands, and it's going to be using quite a lot more power (and as your said a lot more noise) than a typical 3D printer.


I just think while it's an interesting side project, you are just better off in almost all respects building or buying a dedicated machine. Not least because tying your Router up with long prints means you can't use it for anything else at the same time!

[Doddy]

Happy QIDI owner here, having given away my home-built RepRapPro. Pretty rugged (by comparison) and the print far better. Dual extruder as standard, but as another has said - hardly ever use it and it's just another nozzle to snag on the work piece. But, worked out of the box with just a minor bed alignment. Did have one drive cable fail, but it came with a spare (!)

[Chaz]

Andrew Mawson has one, https://www.cetus3d.com/. Fairly well priced, imported machine but uses hiwin rails for the structure. Ive seen the prints it does, really impressive.

[Boyan Silyavski]

Andrew Mawson has one, https://www.cetus3d.com/. Fairly well priced, imported machine but uses hiwin rails for the structure. Ive seen the prints it does, really impressive.

Looks very nice. I don't know if its me only but i hate that Ultra cheap design. I would gladly pay 100 more for all to look normal gantry like

[Chaz]

Looks very nice. I don't know if its me only but i hate that Ultra cheap design. I would gladly pay 100 more for all to look normal gantry like

Agreed however IVe seen the prints this does, its impressive. I went as far as abandoning my own designs / ideas as for this money, nothing comes close.

[Ross77]

The Cetus3d looks really good but resisting the urge to say "I have all the parts to build that...."

the only down side is that it will only do PLA. Do you guys use nylon much or is it really difficult to do the engineering plastics?

[Chaz]

The Cetus3d looks really good but resisting the urge to say "I have all the parts to build that...."

the only down side is that it will only do PLA. Do you guys use nylon much or is it really difficult to do the engineering plastics?

Agreed, for that money, its not worth the time to design, buy, build and test IMHO.

[Ross77]

Agreed, for that money, its not worth the time to design, buy, build and test IMHO.

Any idea how much to import it though? not available yet either.



I'd recommend a prusa www.prusa3d.com
I have the previous to the last model and works like a charm. With the new one they have won best printer of the year and looks like a really good printer on a budge

That one is about £1k, are the clones as good?

[Chaz]

Speak to AndrewMawson on here, he imported one.

[magicniner]

Do you guys use nylon much or is it really difficult to do the engineering plastics?

I'm currently using PLA (and some ABS for tougher small parts) on my (£275) Pusa i3 clone kit, I'm in the process of modifying for Nylon and for larger ABS parts as I have product designs for parts which require the better engineering plastics with higher glass transition temperatures and better durability than PLA.

- Nick

[Ross77]

Hows this one if I go the Purusa route http://factory3d.co.uk/

[magicniner]

Hows this one if I go the Purusa route http://factory3d.co.uk/

That's the one I went for, I don't think you can beat it for the money and support is excellent via email and telephone, Neil from Model Engineer has that one too, have a look at his posts over there.

- Nick

[Ross77]

or else there are these? http://www.ebay.co.uk/itm/CTC-3D-Printer-Black-DUPLICATOR-4-Dual-Extruders-with-1-roll-of-abs-Makerbot/131407319751?_trksid=p2047675.c100005.m1851&_trkparms=aid%3D555018%26algo%3DPL.SIM%26ao%3D2%26 asc%3D20140122125356%26meid%3D94bc0d2f4fff40de98df e2d165929d04%26pid%3D100005%26rk%3D4%26rkt%3D6%26s d%3D321930591168

or http://www.ebay.co.uk/itm/3D-printer-Wanhao-duplicator-i3-V2-1-steel-frame-Fully-Assembled-/321930591168?epid=1962263201&hash=item4af48efbc0:g:ATAAAOSw5dNWtIzU

[Desertboy]

This Prusa is crazy cheap I'd just buy this if I were you.
http://www.ebay.co.uk/itm/2017-Upgraded-Full-Quality-High-Precision-Reprap-Prusa-i3-DIY-3d-Printer/252999597353?_trksid=p2047675.c100005.m1851&_trkparms=aid%3D555018%26algo%3DPL.SIM%26ao%3D2%26 asc%3D20140122125356%26meid%3D4bf9dd71734c4e56b122 7666b14b5a5f%26pid%3D100005%26rk%3D2%26rkt%3D6%26m ehot%3Dlo%26sd%3D272622069677

Add auto levelling for £2

[magicniner]

I like the look of the Wanahao.
I'd avoid plastic framed printers like the plague, Aluminium or Steel construction at least means you have a good base for a rebuild or upgrade project at some future point.
Nozzles and Throats which aren't commonly available as generic spares are a bad idea as are machines not running the most popular open-source firmware versions.
Printers with the hot end and nozzle very close to the extruder drive may well prove problematic should you wish to print one of the more flexible filaments,

- Nick

[A_Camera]

I'd avoid plastic framed printers like the plague, Aluminium or Steel construction at least means you have a good base for a rebuild or upgrade project at some future point.


Some of the cheap ones are not even plastic, but thin plywood.

[Desertboy]

Some of the cheap ones are not even plastic, but thin plywood.

Nothing wrong with a I3 Graber ;)

https://www.youtube.com/watch?v=3dUQHSLe5iY

I bought the frame for one because they look good especially when varnished.

They print very well mate has had one a few years working many hours day and still produces excellent prints.

most grabers are 6mm ply but if you have a router working you can easily cut a 12mm version with minor alterations in cad.

When you start worrying too much about rigidity in the printer then you need to ditch the shit 8mm smooth rod lol and go hiwin route. But for home printer stupid overkill.

Nearly all Prusa I3 based builds I've come across print well it's a good design there probably better stuff kicking around now but Prusa I3 printers are certainly not a bad choice.

I've seen Acrylic, aluminium & ply builds and the box frame works just as well made with a table saw lol.

[charlieuk]

I used a cheap clone one while I was away and it was a pain in the butt trying to get the bed level and messing about with setting and glue stick and lifting prints and ...........

when I got back I ordered the genuine prusa and within 10 min of turning it on it was printing faultless prints. if you put a price on your time the extra cost of the genuine will pay for it self almost instantly and with a lot less stress plus you get a 250 x 210 x 200 actual build area a proper Rambo and e3d extruder. down load slic3r and all the settings are all ready tweaked by prusa and can go down to 0.05 layer hight

[magicniner]

When you start worrying too much about rigidity in the printer then you need to ditch the shit 8mm smooth rod lol and go hiwin route. But for home printer stupid overkill.

Understanding the engineering and physics involved will show the elegance of the Prusa design, this is that the 8mm unsupported rods are not loaded dynamically the way they would be in a router or mill, the carriage acceleration loads are transferred by the drive belts, motors and motor mounts to the frame with the rails simply supporting carriage weight.
Failure to understand this might lead to over engineering the rails and under engineering the frame when it is the frame which supplies the required rigidity for this style of machine, if you sensibly move to a tube feed system with the reel off the machine the only significant loads experienced by the 6mm Aluminium plate upper frame are in the + & - X travel directions where a planar frame performs extremely well indeed.

- Nick

[A_Camera]

Nothing wrong with a I3 Graber ;)

I bought the frame for one because they look good especially when varnished.

They print very well mate has had one a few years working many hours day and still produces excellent prints.

most grabers are 6mm ply but if you have a router working you can easily cut a 12mm version with minor alterations in cad.

When you start worrying too much about rigidity in the printer then you need to ditch the shit 8mm smooth rod lol and go hiwin route. But for home printer stupid overkill.

Nearly all Prusa I3 based builds I've come across print well it's a good design there probably better stuff kicking around now but Prusa I3 printers are certainly not a bad choice.

I've seen Acrylic, aluminium & ply builds and the box frame works just as well made with a table saw lol.

I have no doubt that it can be made to work, especially for printing toys, but I was referring to the thin simple designs. The one you show is really large and is more like a CNC router than a 3D printer, at least as far as the base and the beam is concerned. Also those rods look more rigid, something like 12mm (1/2") in diameter, not just 8mm. Then again, it depends on what you want to use it for. Personally I see no reason for using wood or anything similar, but I know some people disagree.

[Desertboy]

I think it must be perspective because it's 8mm smooth rod, standard prusa build size.

I have the same printer it's not particularly big I had a mendel and just bought the ply frame and swapped the parts over.

Was a massive upgrade from the original mendel lol.

This was the mendel
22343

One of the early FDM 3d printers for the home

[Boyan Silyavski]

Original prusa would have been the kill with Hiwin bearings and sturdier bed. I dont believe square supported bearings are overkill for a printer, i believe they are a must for a machine that moves non stop. My Prusa growlingly agrees from the garage. My wife will tell you that too.

[Zeeflyboy]

The Cetus3d looks really good but resisting the urge to say "I have all the parts to build that...."

the only down side is that it will only do PLA. Do you guys use nylon much or is it really difficult to do the engineering plastics?


In my personal opinion, a heated bed is a huge plus. I'm not a big fan of ABS but there's a cornucopia of options out there and most are much easier with a heated bed.

My personal favourite material for functional parts is XT-CF20 but it's a bit of a pain in some ways. Simply the best material I've ever used for strength, layer adhesion and lack of warp. As a cheaper option for normal use I am getting on very well with e3d's edge.

[cropwell]

Understanding the engineering and physics involved will show the elegance of the Prusa design, this is that the 8mm unsupported rods are not loaded dynamically the way they would be in a router or mill, the carriage acceleration loads are transferred by the drive belts, motors and motor mounts to the frame with the rails simply supporting carriage weight.
Failure to understand this might lead to over engineering the rails and under engineering the frame when it is the frame which supplies the required rigidity for this style of machine, if you sensibly move to a tube feed system with the reel off the machine the only significant loads experienced by the 6mm Aluminium plate upper frame are in the + & - X travel directions where a planar frame performs extremely well indeed.

- Nick

I bought the Wanhao 4DS because of the steel frame, everything else is Makerbot, which has been cloned a lot, so it must have some plus points. The filament reels are on the frame, tube fed to the extruders. The machine has a heated base, which I have a borosilicate plate on. I only print in ABS and on glass I find a thin wash of ABS solution in DMK (acetone ) or MEK (methyl ethyl ketone) gives me good adhesion and easy separation when the bed cools. The other thing it has is a hood to maintain the temperature in the enclosure, which reduces (but not always eliminates) corner lifting. There was mention of a £259 Makerbot clone, and if this has a steel frame, I would consider it a good buy, if I were looking for a 3D printer.

I haven't seen many Hiwin railed machines, probably for a good reason :peaceful:

Cheers,

Rob

[Ross77]

Thanks for all the advice, starting to get an idea of what I need. certainly lots of choice.

Is there any advantage of a 24v system? I'm assuming lower current so less stress on the heater and faster stepper motors?

Thanks again for all the input

[Zeeflyboy]

24v better all round. My printer was originally 12v, they later changed newer models over to 24v and I upgraded - steppers happier, bed heats up much faster, hot end a bit faster.

[Ross77]

Ok so Im coming to the conclusion that I will have to build one as Id probably end up rebuilding whatever I can buy.

If i was going to buy one I think the Wanhao I3 plus looks the best as it is 24v and touch screen, that said people still upgrade the extruder and have to brace the gantry to improve stability.

Ive had a dig around and I have all the mechanical bits to make a 500x300x200 machine. x and y belt driven and z on 4mm pitch lead screws. seems a no brainer as then I can get the best electrics to make it work.

Probably go for a 24v system and 32bit controller to get the best resolution

E3d aero extruder and hot end looks good quality and light weight

Any recommendations for the contoller and heat bed?

[Desertboy]

I'd look at a rapsiberry pi solution, cheap as chips and can add 4" colour touch screen for a £10

As for 24v
The difference you will notice giving the nema's 24v is not like with nema 23's lol, you'll get 30% more torque but in the real world it will make almost no difference.

But the heated bed you will notice the difference I would just get a cheap mk3 aluminium heated bed, they work well at 24v and heat up in less than 2 mins, at 12v it can take over 10 mins! If you're only printing PLA I find at 12v the bed will heat to 60c in a couple of mins but to get to 110c at 12v takes forever! So 24v makes much more difference when printing ABS.

[A_Camera]

I'd look at a rapsiberry pi solution, cheap as chips and can add 4" colour touch screen for a £10

As for 24v
The difference you will notice giving the nema's 24v is not like with nema 23's lol, you'll get 30% more torque but in the real world it will make almost no difference.

But the heated bed you will notice the difference I would just get a cheap mk3 aluminium heated bed, they work well at 24v and heat up in less than 2 mins, at 12v it can take over 10 mins! If you're only printing PLA I find at 12v the bed will heat to 60c in a couple of mins but to get to 110c at 12v takes forever! So 24v makes much more difference when printing ABS.

The higher voltage on the steppers is mainly for speed not the torque. I think it makes a difference there.

[magicniner]

I've just upgraded to a Mk3 bed on a 12v printer and although it's not quick it's not as slow as some have experienced.
The Mk3 bed has two 12v heater circuits, connect them in parallel and it runs on 12v, connect them in series and it runs on 24v but in either case each circuit only gets 12v, for heating the problem is not the voltage it's the voltage drop experienced with inadequate power supplies often shipped with basic 12v machines.
I am planning to upgrade a separate bed heater PSU with a torroidal transformer, a rectifier, a bit of smoothing and a DC-DC solid state relay.
I have a larger format 3D printer build in the pipeline for which I've decided to use 240v Silicone heater pads using a solid state relay to interface with the control board's standard heater output.

[Desertboy]

The higher voltage on the steppers is mainly for speed not the torque. I think it makes a difference there.

I run my printer at 24v I never noticed a difference in print speed when I upgraded from 12v but I did notice the heated bed, maybe it makes more a difference with a 32bit controller (I modded a ramps board to 24v)

The nema's do seem a little quieter at 24v.

[Ross77]

I'd look at a rapsiberry pi solution, cheap as chips and can add 4" colour touch screen for a £10

Ive been looking for the Pi but cant find many options, Any specific ones you know of?


As for 24v, The difference you will notice giving the nema's 24v is not like with nema 23's lol, you'll get 30% more torque but in the real world it will make almost no difference.
if i make the machine then the motors will be nema 23-24s, might have to go for separate drives any way.

I like the look of the Duet as its all in one but could be an issue if a section stops working. At least separates can be replaced

[Desertboy]

Ive been looking for the Pi but cant find many options, Any specific ones you know of?


if i make the machine then the motors will be nema 23-24s, might have to go for separate drives any way.

I like the look of the Duet as its all in one but could be an issue if a section stops working. At least separates can be replaced

Nema 23's for a 3d printer unless it's crazy over sized is probably too much like you don't want to use nema 34's on smaller routers, I built my printer with 0.44nm nema 17's (This is at the high end) £10 each from Uk seller and they can whip along at a very fast rate I run my machine a lot slower than I could because I only print occasionally so print time is irrelevant to me and I've found slower speeds are more forgiving.

I only print ABS which is a lot more hassle than PLA which is a joy in comparison.

[Zeeflyboy]

nema23 I would also say are overkill for pretty much any 3d printer, except maybe if you are lifting a fairly heavy solid bed - I could see the Z-axis benefiting there. If you have spare nema23's lying around anyway then of course no reason not to use them.

I was actually toying with the idea of a servo driven 3d printer using small 100w servos or similar for X and Y, could be a fun experiment if I can find some cheap enough.

[Zeeflyboy]

Ive been looking for the Pi but cant find many options, Any specific ones you know of?


if i make the machine then the motors will be nema 23-24s, might have to go for separate drives any way.

I like the look of the Duet as its all in one but could be an issue if a section stops working. At least separates can be replaced


I use a raspberry pi with the official 7" touchscreen and octoprint as my printer's control interface. Very neat solution - also allows a webcam feed and remote access.

astroprint is another good print server, I think they are adding easier touchscreen support. Less technical than octoprint but looks easy to use and well presented.

[Ross77]

Nema 23's for a 3d printer unless it's crazy over sized is probably too much like you don't want to use nema 34's on smaller routers

Yes, sorry I meant to say for the z-axis, I have a pair of complete slide/lead screw assemblies. the x,y will be belt driven so plenty fast enough.


I use a raspberry pi with the official 7" touchscreen and octoprint as my printer's control interface. Very neat solution - also allows a webcam feed and remote access.

Is the PI just running the printers driver or does it control the steppers and heating mosfets direct?

guessing i will still need a duet or RAMPs with the PI

[Zeeflyboy]

In my setup it's just acting as the computer that interfaces with the ramps board through USB.

[Washout]

In my setup it's just acting as the computer that interfaces with the ramps board through USB.

I use a similar setup on my Hictop 3DP-12 (Prusa clone with aluminium extrusion frame) - it has a great benefit of being able to run the Pi with Octoprint remotely across my LAN and I use a Pi camera, which is integrated to keep an eye on the print whilst its doing it. So that's PC in office with web browser interface---->Pi with Octoprint---->Ramps based control board from Hictop running Marlin firmware----->3DPrinter.

I've just finished upgrading the printer this weekend with a filament sensor, although I struggled with using the firmware's M600 filament change routine, which kept timing out Octoprint, so rolled my own GCode to emulate it without the timeout issues.

[Ross77]

In my setup it's just acting as the computer that interfaces with the ramps board through USB.

are you using the standard Ramps or the 32 bit Radds board?


>Pi with Octoprint---->Ramps based control board from Hictop running Marlin firmware----->3DPrinter.

have you got any wiring diagrams for that set up?

[Washout]

are you using the standard Ramps or the 32 bit Radds board?



have you got any wiring diagrams for that set up?

Not as such - as with Zeeflyboy the Pi is connected to the Ramps board via a short USB cable and the Ramp board uses 4 wire stepper wiring, which is likely the same as you would use for a CNC router.

The board is shown here (scroll down a bit for the schematic): https://www.hic3dprinter.com/collections/3d-printers-parts/products/hictop-3d-printer-control-board-mpx-3-reprap-arduino-compatible-mother-board

I mounted my Pi onto the printer by using a 3D printed case (shameless channel plug) and I have my property flood wired for CAT5e so it just becomes another device on it accessible from any PC/tablet etc. with a browser:


https://youtu.be/95eNVlc3N1w

[Boyan Silyavski]

Wow, i would never do that :hysterical: My printer prints non stop and why i would need to spend time watching its whats its doing, as if all is setup right it works flawlessly. last time i had a crash was when filament went from 1.75 to 2.5 at almost end of bobin...

[Desertboy]

Wow, i would never do that :hysterical: My printer prints non stop and why i would need to spend time watching its whats its doing, as if all is setup right it works flawlessly. last time i had a crash was when filament went from 1.75 to 2.5 at almost end of bobin...

What materials do you normally print with? I've had trouble with larger parts and ABS lifting up at the corners, anything that's over 6cm tall has caused me problems in the end I have to encase the printer to get it to print larger parts (I do this with a cardboard box lol) but when my router works going to build this enclosure to add to my printer because it looks nice lol.

https://youtu.be/wt7bQXyLSL4

Never had an issue with PLA found it very easy to print if you have a fan blowing over the part correctly obviously I don't use the fan with ABS because it shrinks one of it's design features (So it drops out the injection molds easier).

[Ross77]

Not as such - as with Zeeflyboy the Pi is connected to the Ramps board via a short USB cable and the Ramp board uses 4 wire stepper wiring, which is likely the same as you would use for a CNC router.

The board is shown here (scroll down a bit for the schematic): https://www.hic3dprinter.com/collect...e-mother-board

Ah that makes sense now. The ramps I was looking at didn't have usb, only a shield for the Aurduino.

More options to look for now, thanks

[Washout]

Wow, i would never do that :hysterical: My printer prints non stop and why i would need to spend time watching its whats its doing, as if all is setup right it works flawlessly. last time i had a crash was when filament went from 1.75 to 2.5 at almost end of bobin...

Well I record the timelapse for one reason. The other is my printer is at the other end of my property and occasionally the "filament out" sensor trips out when it shouldn't. I can glance up at one of the bank of monitors in my office and tell what's up and decide whether to make the trek down there or just hit the resume button. I would have thought anyone with an interest in CNC would have also been interested in maximising automation?

[Boyan Silyavski]

What materials do you normally print with? I've had trouble with larger parts and ABS lifting up at the corners, anything that's over 6cm tall has caused me problems in the end I have to encase the printer to get it to print larger parts (I do this with a cardboard box lol) but when my router works going to build this enclosure to add to my printer because it looks nice lol.

https://youtu.be/wt7bQXyLSL4

Never had an issue with PLA found it very easy to print if you have a fan blowing over the part correctly obviously I don't use the fan with ABS because it shrinks one of it's design features (So it drops out the injection molds easier).

I print ABS in an enclosure, rafts and supports always. Plus some extra coins that help against lifting. PLA same. Nowadays i use mainly PETG for functional parts/ up to 70C/ .

Whatever they tell you each one of the materials lifts, shrinks and so on. I print the dust shoe from my signature and there is not even one material that it will print ok due to the specific shape, if i did not knew how to counter effect that. All else is marketing.
See example below of the coins, how i situate them, or all hacks up...

3d printing makes a lot of sense after the CNC and once you understand what is happening and solve the common beginner problems, its easy to become an expert when you are using your brain and figuring stuff. But on Facegroup groups and so i was tired of repeating the steps to understand all, nobody cared. So i dont visit them now.


22428

[Zeeflyboy]

Whatever they tell you each one of the materials lifts, shrinks and so on. I print the dust shoe from my signature and there is not even one material that it will print ok due to the specific shape, if i did not knew how to counter effect that. All else is marketing.

XT-CF20 and Carbonfil (I can't tell much difference between the two in how they print or the end results) are the only materials I've ever used that literally just don't seem to suffer from any warping even on very large parts... very dimensionally stable stuff, excellent overhang ability and very strong layer adhesion.

Easy enough to print with, but you have to used hardened steel nozzles as they destroy brass or even normal steel ones very quickly. I also had a problem with it sticking to the nozzle and building up until finally a big blob gets stuck on the part and potentially ruins the entire print... I never managed to get rid of that entirely and prints would need frequent babysitting with a pair of needle nose tweezers to de-blob the extruder nozzle, a massive pain in the arse. I intended to experiment with a nozzle wipe action on each layer change at some point using a silicon wiper to see if it could address the babysitting issue, but haven't got around to it yet to see if it works.

Due to that I've switched over mainly now to Edge for most stuff, doesn't hurt that it's cheaper too. It is a lot better than ABS in regards to warping, but not as good as the two I mention above.

I also use a print-bite bed and that stuff is bloody brilliant. I've printed ABS, PLA, Edge (a PET type plastic iirc), XT-CF20, Carbonfil and ninjaflex on it. Stuff sticks very well when heated and releases easily enough when it's cooled a bit... no more messing around with tape/slurry/spray etc. That is one of the best things I've ever bought for the printer.

[Ross77]

I also use a print-bite bed and that stuff is bloody brilliant

Looks like they are out of stock! What would be the best make up for a bed if starting from scratch? Ali plate - heater - glass - print bite ?

[Zeeflyboy]

Best is probably debatable.

Personally I just have heater/alu plate/print bite. Its nice and light (my printer moves the bed for Y axis so lighter the better really), good heat transfer and quick to warm up.

If you want a removable bed then of course you need another layer using something flat. Personally I'd probably use a thin steel plate and use high temperature magnets to retain it.

[Desertboy]

I print onto Kapton tape or straight on to a mirror with ABS juice (Acetone and ABS dissolved together). Since I rebuilt my printer and now it's level I rarely have one that doesn't stick to the bed. At least for me 99% of sticking to bed problems were to do with the nozzle being too high or too low.

If I got another one whatever I bought/built I would make sure the Z axis has a contactless solution for setting the Z height. Inductive probe works well as long as you have an aluminium heated bed I use an 8mm probe and it will sense the aluminium bed through the mirror glass which is nice. Optical end stops are good too and hall effect switches. I found microswitchs unreliable for the Z axis.

Never ever use stainless for a print bed stainless has very poor heat conductivity. Tooling aluminium is the best choice for a bed because it's light and has excellent heat conductivity that's why they make heat sinks out of it. The super flat is just a bonus.

[A_Camera]

Desertboy, so you mean that the inductive sensor will sense the aluminium table? I thought it only senses magnetic material. Also, how thick is your table?

[Clive S]

Desertboy, so you mean that the inductive sensor will sense the aluminium table? I thought it only senses magnetic material. Also, how thick is your table?
Yes they will but you need the 8mm one

[cropwell]

I am always wary about axial sensing with inductive sensors, after crushing one. The coil at the end is quite delicate, even though it has a plastic shield.

[Desertboy]

Yes they will but you need the 8mm one

Yes the 4mm doesn't sense through the mirror and you also need to feed it 12v so if you have ramps you will need to wire 2 resistors to drop the output to 5v otherwise you fry the Arduino.

[AlexDoran]

I was thinking about releasing the designs to my printer, i had planned to build and sell these, but with so many cheap units flooding the market i decided it was probably not worth it, if anyone is interested i will go to the effort of updating the BOM's and finishing the documentation.

Cronos-FDM.uk

Thanks

Alex

[Greeny]

Inspired by this thread, and on a bit of a rush of blood, i decided to buy a Wanhao Duplicator i3 Plus.
http://www.ebay.co.uk/itm/Wanhao-I3-Plus-3D-Printer/222201963657?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2057872.m2749.l2649
It arrived yesterday :-)

I'm a total 3d print noob, so the reasons i went for this particular one are:


Was mentioned as a good choice in this thread.
Price, £379.
Generally positive reviews on internet & You Tube.

Overall I'm very impressed, for the money.
As I had already watched quite a few YouTube guides, assembly & setup was a breeze.
It took me about 45 mins from the box coming through the door to starting my first print. (the 'ok' hand that comes on the SD card)
The print quality is pretty good. Again i'm more than happy with it for the money paid.

I've printed a lens cap & gimbal mount for my camera using STL files downloaded from the internet(Thingiverse) and then used Cura to generate the G-Code
It was all relatively straightforward & again very happy with results

Next step is to try some of my own stuff drawn in Fusion 360.

It's worth pointing out that I have never even seen a 3d printer 'in the flesh' before this one, so don't take my impressions as worth much. I don't have anything to compare to!

The photo's are of my first few prints. They look better in real life than in the photos! (The white plastic doesn't seem to photograph very well)

I'm stoked :thumsup:
Cheers

224442244222443

[Ross77]

Inspired by this thread, and on a bit of a rush of blood, i decided to buy a Wanhao Duplicator i3 Plus.
http://www.ebay.co.uk/itm/Wanhao-I3-...72.m2749.l2649
It arrived yesterday :-

Nice one, Kind of wishing I had just bought one now...

[Greeny]

Nice one, Kind of wishing I had just bought one now...
Swings & roundabouts i suppose :smile:

In terms of getting going quickly and ease of use, it's a winner.
However if building your own you will almost certainly end up with a better machine.

For the money paid I'm very happy with it, but it's definately built with parts supplied by the lowest bidder :greedy_dollars:

[Ross77]

Does this look like a good controller? 32 bit with screen and replaceable drives

http://www.ebay.co.uk/itm/Motherboard-32bit-ARM-Controller-WiFi-3-5-Touch-Screen-DIY-for-3D-Printer-TE722-/162591981233?hash=item25db3cbeb1:g:d6EAAOSwpHFZaYj 9

[AlexDoran]

Does this look like a good controller? 32 bit with screen and replaceable drives

http://www.ebay.co.uk/itm/Motherboard-32bit-ARM-Controller-WiFi-3-5-Touch-Screen-DIY-for-3D-Printer-TE722-/162591981233?hash=item25db3cbeb1:g:d6EAAOSwpHFZaYj 9

What firmware is it supposed to use?

If you want a solution like that i am personally using MKS products, the MKS SBase and MKS TFT32. The S Base is a smoothieboard clone which obviously uses smoothieware. The TFT32 is a host controller and can also be fitted a Wi-Fi addon, smoothieware also has a web interface if you wanted to connect to it via ethernet.

Thanks

Alex

[Desertboy]

Does this look like a good controller? 32 bit with screen and replaceable drives

http://www.ebay.co.uk/itm/Motherboard-32bit-ARM-Controller-WiFi-3-5-Touch-Screen-DIY-for-3D-Printer-TE722-/162591981233?hash=item25db3cbeb1:g:d6EAAOSwpHFZaYj 9

The firmware for this can be found here
https://github.com/St3dPrinter/Marlin4ST

[AlexDoran]

The firmware for this can be found here
https://github.com/St3dPrinter/Marlin4ST

Can't view that very well on my phone, I didn't think Marlin had been ported to 32bit yet.

Alex

[Ross77]

The firmware for this can be found here
https://github.com/St3dPrinter/Marlin4ST

Looks like a a really good option as its ARM processor and can run Marlin but I cant find anybody using it or support.


If you want a solution like that i am personally using MKS products, the MKS SBase and MKS TFT32. The S Base is a smoothieboard clone which obviously uses smoothieware. The TFT32 is a host controller and can also be fitted a Wi-Fi addon, smoothieware also has a web interface if you wanted to connect to it via ethernet.

Is the smoothie a standalone or can I use a pi to controll it? looks like hard work to set up the smoothie tho.

[magicniner]

Smoothieware has a reasonably well commented text config file on the SD card which you simply edit in a text editor on your PC then reboot the smoothie.
That makes smoothieware even simpler to tweak than Marlin on an Arduino, which again is editing a text config file but then compiling and uploading from the Arduino development environment - that took me a few hours to work out and complete from scratch when I swapped to higher micro-stepping drivers and needed a higher hot end max temp ;-)

[AlexDoran]

Smoothie can be independent or host controlled, depending on what LCD / TFT controller you want to use with it; i believe it can be operated from octoprint, it does have its own Web Interface to be used with Ethernet, never used it so can't comment on how well it works.

Config is pretty simple, educating yourself on the variables within the config file can be pretty steep, the documentation on their wiki is very complete, but some of it is confusing to read, not all variables come in the config file as standard, you have to add some of the uncommon ones in yourself, setting up a basic 3D Printer config is easy though. When you download the firmware it has various template files setup already.

When you want to update the firmware you keep your config and just copy the firmware file to the SD and thats it job done, not like Marlin where they release a new RC Version, change half the variables and their names so you have to reconfigure the whole firmware again.

The SBase is a great clone i am using several of them already, had no problems what so ever, if you want to use the TFT32 touch panel controller, bear in mind that it is classed as a host controller, it has its own firmware and you cant control the SBase from other devices (PC) whilst using the TFT.

There are some other 32 bit controllers around(Re-ARM, RAADS etc) trouble is none of them support common 32bit firmware, Smoothie has been 32Bit from day one, it's not being "ported" to work with 32Bit, and i believe the motion algorithms it uses are some of the best.

Marlin is very good, i use it on one of printers that has a RAMPS controller (an old gMAx 1.5), but now half of their dev team has moved on to trying to port it to 32bit, im not sure how well each branch of the firmware is progressing because i dont really follow it anymore.

Also worth mentioning, do you really NEED a 32bit controller? I only use 32bit because my printers are driven by ballscrews and the steps per mm is higher than using belts, so performance wasnt great for me until i went 32bit. If you want to use ballscrews or build fast Delta Printer i would stick to Ramps or similar, Mega cheap and very easy to setup.

The only gripe i have with Smoothieware is that the guys who write are, for lack of better words - arrogant bastards, how dare you ask for help or suggest changes LOL

Thanks

Alex

[Zeeflyboy]

Aren't there some new smoothie board options coming out soonish?

I haven't looked into it for a while but was considering one of the v2 boards for my own-design machine, hoping they will be around by the time I get to actually build it.

[Fred]

I jumped in and contributed a little bit of code on Smoothie and thought they were very helpful, especially as I was a C++ newbie.

I think V2 has been on the cards for quite a while so I wouldn't hold your breath.

[Ross77]

Also worth mentioning, do you really NEED a 32bit controller? I only use 32bit because my printers are driven by ballscrews and the steps per mm is higher than using belts, so performance wasnt great for me until i went 32bit. If you want to use ballscrews or build fast Delta Printer i would stick to Ramps or similar, Mega cheap and very easy to setup.

Im coming to that conclusion and may just get a cheap mega/ramps set up to start with while I learn. the belt drives I'm using are from a vinyl plotter and are geared so lookes like I will have plenty of resolution (20mm/200=0.1mm with no microstepping)

I do want to be able to print small high detailed parts tho.

Other than the fact that i can use a 1/32 mircostepping on the swappable boards is there any advantage of going with the soldered drives limited to 1/16th?

[magicniner]

Iis there any advantage of going with the soldered drives limited to 1/16th?

Yes, if something dies you get the pleasure of soldering to fix it ;-)

[AlexDoran]

If you want high resolution, you can still use TMC2100 stepper drivers, expensive compared to the more common drivers used on a RAMPS but can provide upto 1/256 microstepping iirc, operate almost silently too.

Thanks

Alex

EDIT: Although DRV8825's still provide excellent performance and resolution.

[m_c]

If you want high resolution, you can still use TMC2100 stepper drivers, expensive compared to the more common drivers used on a RAMPS but can provide upto 1/256 microstepping iirc, operate almost silently too.


Microstepping does not improve resolution. Due to stiction/friction in the system you can only guarantee positional accuracy to within 1 full step, regardless of the number of microsteps.
When microstepping, the simplest way to describe how the motor's rotor is being held, is opposing coils are pushing or pulling against each other and holding the rotor like it's being pushed/pulled by a couple springs. Then depending on the position of the rotor relative to the cogging, you've effectively got another spring trying to push the rotor either way. And that's before you allow for any stiction/friction affecting the forces.

The only real benefit microstepping has is it improves how smoothly a stepper motor rotates at low speed. Once you get above a certain speed, it has no benefit which is where good quality drives will gradually reduce output stepping to full step mode as motor speed increases. Good drives will also let you adjust the point at which that happens, as it helps avoid motor stalling due to resonance between the stepping speed, motor cogging, and load.

[AlexDoran]

So you are saying that if the transmission was provided by a ballscrew with a 16mm Pitch, at 1 to 1 stepping - that i would have the same resolution as if i was using 1/32 microstepping.

Thanks

Alex

[A_Camera]

So you are saying that if the transmission was provided by a ballscrew with a 16mm Pitch, at 1 to 1 stepping - that i would have the same resolution as if i was using 1/32 microstepping.

Thanks

Alex1/256 microstepping sounds totally pointless. Even 1/32 sounds too much to my ears. I run my CNC with 1/10 microstepping and I think that's more than enough for both speed and smoothness.

As far as I know, there is no ball screw with 16mm pitch, unless it is very large in diameter. Even 10mm pitch is normally 20-25mm in diameter and is very heavy. Nothing you can run with NEMA17 motors, which are common in 3D printers. My CNC is using 1605 ball screws, which means it is 16mm in diameter and has 5mm pitch.

I think if I build a 3D printer I will use dual ball screws (probably 1004) on the Z with one NEMA23 and belt connecting the two screws. For X and Y I will use direct belt drive.

[AlexDoran]

1/256 microstepping sounds totally pointless. Even 1/32 sounds too much to my ears. I run my CNC with 1/10 microstepping and I think that's more than enough for both speed and smoothness.

As far as I know, there is no ball screw with 16mm pitch, unless it is very large in diameter. Even 10mm pitch is normally 20-25mm in diameter and is very heavy. Nothing you can run with NEMA17 motors, which are common in 3D printers. My CNC is using 1605 ball screws, which means it is 16mm in diameter and has 5mm pitch.

I think if I build a 3D printer I will use dual ball screws (probably 1004) on the Z with one NEMA23 and belt connecting the two screws. For X and Y I will use direct belt drive.

I'm sorry but you could not be further from the truth, 1/256 microstepping is widely used in the 3D Printing sector, mostly for improved quality and almost silent operation (Mainly with Delta Printers), you think people want to listen to their printer go for 40 hours when its in the next room, making a noise when you're trying to sleep? The needs for a 3D Printer are totally separate from a Router regardless of how similar they actually are.

Well my printer uses a SFU1616 ballscrew on both the X & Y and a SFU1204 on the Z Axis, the X Axis is driven by a 1/32 microstepped Nema17, DRV8825 driver.

22487

A belt is fine for most applications, but if you printing engineering plastics where the Bed Temperature needs to be 120c+ you will soon see they stretch and deform from the heat, screws are also far more robust.

Thanks

Alex

[m_c]

So you are saying that if the transmission was provided by a ballscrew with a 16mm Pitch, at 1 to 1 stepping - that i would have the same resolution as if i was using 1/32 microstepping.


Re-reading what I wrote, I should of used accuracy instead of resolution.
You can have as much resolution as you want, but you still can't guarantee accuracy. I'm aware 3D printers will have less stiction than a typical CNC, but the same problem will still be there. Apply a bit load to the screw, or even try twisting it by hand, and unless you have a system with next to zero friction, it won't return to the exact position.
However with a 3D printer, you should get good repeatability, as the load is pretty constant. It's not like a typical CNC machine where you'll have a cutter pushing/pulling things away from where you want them.

The best way to see how much accuracy you do, or don't gain, would be to fit a high resolution encoder to a stepper, and see how even the encoder pulses are compared with the pulse train at slow speed. Once things are moving inertia will help smooth things out, and drivers will typically reduce the output microstepping anyway.

It's a bit like those who say a 10'000 count servo on a directly driven 5mm pitch screw has a resolution of 5 microns. In reality a 10'000 line servo is only likely to hold position within 20 counts with a pretty good tune, so your realistic accuracy is 0.01mm.

It's worth mentioning, that more advanced microstepping drives, will still microstep the output, even if you are using full step input. Without microstepping the output, you get rough movement at low speed.

Sound is down to a mix of the switching frequency of the drive, and the sound of the motors physically moving/cogging. Get something that switches above 20KHz, you're not likely to hear any buzzing/humming, but you'll still get the noise from the motors above a few RPM (6RPM if my calculation is correct assuming a typical 200 step motor and a lower hearing frequency of 20Hz - to get 20 steps a second, you need 20*60 = 1200 steps per minute, divided by 200 steps/rev to give 6RPM)

[AlexDoran]

Re-reading what I wrote, I should of used accuracy instead of resolution.
You can have as much resolution as you want, but you still can't guarantee accuracy. I'm aware 3D printers will have less stiction than a typical CNC, but the same problem will still be there. Apply a bit load to the screw, or even try twisting it by hand, and unless you have a system with next to zero friction, it won't return to the exact position.
However with a 3D printer, you should get good repeatability, as the load is pretty constant. It's not like a typical CNC machine where you'll have a cutter pushing/pulling things away from where you want them.

The best way to see how much accuracy you do, or don't gain, would be to fit a high resolution encoder to a stepper, and see how even the encoder pulses are compared with the pulse train at slow speed. Once things are moving inertia will help smooth things out, and drivers will typically reduce the output microstepping anyway.

It's a bit like those who say a 10'000 count servo on a directly driven 5mm pitch screw has a resolution of 5 microns. In reality a 10'000 line servo is only likely to hold position within 20 counts with a pretty good tune, so your realistic accuracy is 0.01mm.

It's worth mentioning, that more advanced microstepping drives, will still microstep the output, even if you are using full step input. Without microstepping the output, you get rough movement at low speed.

Sound is down to a mix of the switching frequency of the drive, and the sound of the motors physically moving/cogging. Get something that switches above 20KHz, you're not likely to hear any buzzing/humming, but you'll still get the noise from the motors above a few RPM (6RPM if my calculation is correct assuming a typical 200 step motor and a lower hearing frequency of 20Hz - to get 20 steps a second, you need 20*60 = 1200 steps per minute, divided by 200 steps/rev to give 6RPM)

Now that makes perfect sense to me, thanks for explaining it i have learnt from your post, you obviously have a much higher understanding than me in general about the technology involved. I think the TMC2100 can take a 1/16 input, and re-pulse (is that even a word?) it to 1/256, but in a way that you still get the 1/16 resolution.

I believe the "Silent Mode" the TMC2100's provide is only really suitable for smaller printers / Delta Printers, and don't handle higher accelerations well, but just for context this is a good video.


https://www.youtube.com/watch?v=oNAHOOolHWw

Thanks

Alex

[A_Camera]

Now that makes perfect sense to me, thanks for explaining it i have learnt from your post, you obviously have a much higher understanding than me in general about the technology involved. I think the TMC2100 can take a 1/16 input, and re-pulse (is that even a word?) it to 1/256, but in a way that you still get the 1/16 resolution.

I believe the "Silent Mode" the TMC2100's provide is only really suitable for smaller printers / Delta Printers, and don't handle higher accelerations well, but just for context this is a good video.


https://www.youtube.com/watch?v=oNAHOOolHWw

Thanks

Alex

It's silent when it is hardly moving. No wonder it must run 40 hours to make anything... and when it is not in silent mode the high pitch noise is unbearable, at least for my ears. Also, didn't you just dismissed belt drive?? I can see a belt in that video and that is very common in 3D printers. I also don't think heat is a problem for the belt. You know that there are different qualities, don't you?

[AlexDoran]

It was moving at 10mm/s because that is the loudest it is going to be causing the most vibrations, are you saying you preferred the noisier DRV8825 compared to the TMC2100?

You are completely missing the point, that video was only intended to display the noise difference between the two drivers, nothing else. Yes i did dismiss belt driven printers for specific applications, you dismissed Ballscrews for all 3D Printers but actually they are required for specific applications - as i stated before, if you are printing in materials such as Nylon or Polycarbonate, you need to have a bed temperature of 120c+, and almost certainly need the printer in an enclosure to maintain the correct ambient temperature.

Now if you try and print a large object, where the first layers are going to be printed very slowly, the toolhead and the belts connected to it would be exposed to high ambient temperatures from the Heatbed, these will cause standard rubber belts to distort, i've seen it happen. Maybe you could get some steel belts or whatever, but whats the point when ballscrews are so cheap and offer far superior performance.

Look im sorry, but go and do some research, LARGE 3D Printers, depending on their architecture, will print at an acceptable quality from 40 - 60mm/s. Now if im printing a full size object on my printer (400 x 400 x 400mm), at a high resoluton maybe 150 - 250microns it will take 40+ Hours, thats standard - not a big deal.

For just one example, check out this video, i have set it to start just before he tells you how long it took to print - Just under 100 Hours.


https://youtu.be/OTXPU2P-ElE?t=3m58s

Thanks

Alex

[m_c]

Now that makes perfect sense to me, thanks for explaining it i have learnt from your post, you obviously have a much higher understanding than me in general about the technology involved. I think the TMC2100 can take a 1/16 input, and re-pulse (is that even a word?) it to 1/256, but in a way that you still get the 1/16 resolution.

I believe the "Silent Mode" the TMC2100's provide is only really suitable for smaller printers / Delta Printers, and don't handle higher accelerations well, but just for context this is a good video.

I've got to admit, people who claim microstepping increases accuracy, are one of my pet hates.

The term you're probably looking for is gearing or scaling. With modern drives, there's a lot of internal scaling going on. The input simply tells the drive the distance you want the motor to move, be that a 256th microstep, or a full step. The drive then takes that, and runs it through however many switching phases it takes to move to that position, which in itself will depend on the speed things are moving at, as once you increase speed, the benefits of microstepping diminish, so the number of switching phases is usually reduced to ensure maximum motor performance (often called morphing).

[Ross77]

Thanks for all the good input. I was all set to build one and then a Wanhao D5 mini came along at the right price so Im now the proud owner of this little beast.22494

also has the heated glass bed and some PLA+ to play with. Should be up and running soon.

Thanks Again.

[spikey47]

Hi I have a Cubex trio 3d printer, triple extruder ideal if you want to use dissolvable filament for support material when printing complex parts. Very easy to use comes with some material & and tools to get you started. Good condition. £400
22534

[A_Camera]

It was moving at 10mm/s because that is the loudest it is going to be causing the most vibrations, are you saying you preferred the noisier DRV8825 compared to the TMC2100?

Of course I'd prefer a silent driver.


You are completely missing the point, that video was only intended to display the noise difference between the two drivers, nothing else.

OK.


Yes i did dismiss belt driven printers for specific applications, you dismissed Ballscrews for all 3D Printers

That's nonsense.


I think if I build a 3D printer I will use dual ball screws (probably 1004) on the Z with one NEMA23 and belt connecting the two screws. For X and Y I will use direct belt drive.

Which part of that sentence was not clear? The only thing which is wrong there is that I meant 1204 ball screws. I also said that there are different quality of belts on the market, not necessary steel belts, but also not need to buy the cheapest rubber band, which of course will cause problems, not only because of heat but also stretching. Regardless of which I did NOT dismiss ball screws for 3D printers, I said thet I THINK I'd use belt (not said rubber belt!!!). Whatever will be is written in the starts right now. I will probably only use my CNC to start with and then decide if I want to build one at all, or if I just use my CNC for 3D printing occasionally. I will definitely not print large parts or parts which takes 40+ h to print.


... but actually they are required for specific applications - as i stated before, if you are printing in materials such as Nylon or Polycarbonate, you need to have a bed temperature of 120c+, and almost certainly need the printer in an enclosure to maintain the correct ambient temperature.

OK


Now if you try and print a large object, where the first layers are going to be printed very slowly, the toolhead and the belts connected to it would be exposed to high ambient temperatures from the Heatbed, these will cause standard rubber belts to distort, i've seen it happen. Maybe you could get some steel belts or whatever, but whats the point when ballscrews are so cheap and offer far superior performance.

Look im sorry, but go and do some research, LARGE 3D Printers, depending on their architecture, will print at an acceptable quality from 40 - 60mm/s. Now if im printing a full size object on my printer (400 x 400 x 400mm), at a high resoluton maybe 150 - 250microns it will take 40+ Hours, thats standard - not a big deal.

For just one example, check out this video, i have set it to start just before he tells you how long it took to print - Just under 100 Hours.


Thanks

Alex

I don't plan to start and industry... so it is not relevant for me and don't have time to watch, and I believe most DIY 3D printer builders/users never print anything that takes 100 hours and never use or build a LARGE 3D printer anyway.

[A_Camera]

Thanks for all the good input. I was all set to build one and then a Wanhao D5 mini came along at the right price so Im now the proud owner of this little beast.22494

also has the heated glass bed and some PLA+ to play with. Should be up and running soon.

Thanks Again.

Cool. Have you printed anything? Are you happy with it? What's your impressions so far?

[A_Camera]

I'm sorry but you could not be further from the truth, 1/256 microstepping is widely used in the 3D Printing sector, mostly for improved quality and almost silent operation (Mainly with Delta Printers), you think people want to listen to their printer go for 40 hours when its in the next room, making a noise when you're trying to sleep? The needs for a 3D Printer are totally separate from a Router regardless of how similar they actually are.

Quality is NOT improved if you are micro stepping a stepper motor at 1/256, that is a total misunderstanding. Also, you don't need to microstep at 1/256 to make it silent and even running, that largely depends on the quality of the driver, as well as the mechanical quality of the printer or router.



Well my printer uses a SFU1616 ballscrew on both the X & Y and a SFU1204 on the Z Axis, the X Axis is driven by a 1/32 microstepped Nema17, DRV8825 driver.
OK, so I was wrong, there exists 1616 ball screws. Anyway, I don't think it is a good idea to use them because the 16mm pitch is just too much in my opinion. I would prefer something with MUCH less pitch.

[magicniner]

I've been printing some Nylon parts with my Factory 3D Prusa i3 clone kit ;-)
This is my PLA part cooling duct with a 40mm fan.
22536

- Nick

[Ross77]

Cool. Have you printed anything? Are you happy with it? What's your impressions so far?

Yes lots of failures but I guess its a learning curve. I had bed leveling issues and poor prints with layer shift, since found out that the 2 or the 3 screws holding the main x-y carriage where missing and the 3rd was loose.

tightening them has solved the layer shift but Im still getting 1mm variation on the bed calibration. guys on the forum think it is the micro switch but it seems to work fine when I tested it with a meter.

I think it will be a good machine but needs a few tweeks and lots of wear, the previous owners must have used it a lot.

biggest issues that i missed when researching are the bowden extruder and being stuck with the wanhao maker slicer program. (firm ware upgrades to 3d partry but not brave enough to try)

[A_Camera]

I've been printing some Nylon parts with my Factory 3D Prusa i3 clone kit ;-)
This is my PLA part cooling duct with a 40mm fan.
22536

- Nick

How long did it take to print it? Which temperatures did you have on the table and the extruder?

[A_Camera]

Yes lots of failures but I guess its a learning curve. I had bed leveling issues and poor prints with layer shift, since found out that the 2 or the 3 screws holding the main x-y carriage where missing and the 3rd was loose.

Too bad... never the less, at that price I'd expect that something needs to be fixed. Glad you managed to fix those small and simple things. Levelling issues are not difficult to fix but depending on the results you expect and the instruments you have to help you, can take time. Anyway, regardless if it is a CNC or a 3D printer, I think that spending time on levelling is beneficial in the long terms.


tightening them has solved the layer shift but Im still getting 1mm variation on the bed calibration. guys on the forum think it is the micro switch but it seems to work fine when I tested it with a meter.

Not good. On the other hand, 1mm variation should be easy to reduce considerably. If you have an instrument to measure than you could see if the problem is mechanical or electrical. The mechanical variation takes some time to remove or reduce, but normally should not be too difficult.


I think it will be a good machine but needs a few tweeks and lots of wear, the previous owners must have used it a lot.

biggest issues that i missed when researching are the bowden extruder and being stuck with the wanhao maker slicer program. (firm ware upgrades to 3d partry but not brave enough to try)

What do you mean is the issue with the bowden extruder? The printer I am planning to build is planned to have bowden type of extruder, so I'd be interested if there is an issue I did not think about.

[magicniner]

How long did it take to print it? Which temperatures did you have on the table and the extruder?

I think the print was about an hour and 40 minutes, 0.4 nozzle, 0.2 layer, 60C bed 170C hot end.
It's in Taulman Bridge which is lovely to work with compared to standard Nylon 6 filaments.

[Desertboy]

I think the print was about an hour and 40 minutes, 0.4 nozzle, 0.2 layer, 60C bed 170C hot end.
It's in Taulman Bridge which is lovely to work with compared to standard Nylon 6 filaments.

I found PLA a lot easier to print with than ABS I only recently changed my hot end to an E3d clone so can now print Nylon, etc. The only issue with PLA is it's absorbs water so make sure you store in with silica gel and if it has absorbed moisture an hour in the oven at 80c will sort it out.

I bought 10kg's of ABS when I first bought my printer so I make it work lol but my life would have ben much easier if I had bought 10kg of PLA instead ;)

I find ABS is strong than PLA but there's not a lot in it.

[magicniner]

The only issue with PLA is it's absorbs water so make sure you store in with silica gel and if it has absorbed moisture an hour in the oven at 80c will sort it out.

Nylon can absorb up to 20% of it's weight in water and can need 1 to 4 hours at 80C to dry a full reel to the core, PLA at 80C for 1 to 4 hours will yield a nice bit of modern art but no useful filament ;-)


I find ABS is strong than PLA but there's not a lot in it.

Print a mount for your phone in PLA and leave it on the dash in a hot car on a sunny day, then tell me that :D


I'm interested in making functional plastic components and while PLA is OK for a few bits it's properties limit it's usefulness in any applications which require greater permanence.

- Nick

[Desertboy]

Nylon can absorb up to 20% of it's weight in water and can need 1 to 4 hours at 80C to dry a full reel to the core, PLA at 80C for 1 to 4 hours will yield a nice bit of modern art but no useful filament ;-)



Print a mount for your phone in PLA and leave it on the dash in a hot car on a sunny day, then tell me that :D


I'm interested in making functional plastic components and while PLA is OK for a few bits it's properties limit it's usefulness in any applications which require greater permanence.

- Nick

I forgot the heat thing, I print everything in ABS apart from the first prints I did so never noticed it lol.

If you want to make injection mold strength parts maybe consider Nylon or Polyurethane casting especially if you want produce a lot of one part, you can use a router or 3d printer to make you master molds.

I definitely want to have a go at casting with polyurethane and also aluminium casting but need to make a furnace first. Vacuum former is the next project but cnc style lol.

[AlexDoran]

Bowden can be trickier to get the retraction settings dialled in properly, and rules out using flexible filaments as well.

You guys should certainly look at PETG filaments, as easy to print as PLA but as strong as ABS, doesn't absorb moisture either; some types can be stringy but good retraction settings handle that. I have been using "Real Filaments" (Check amazon) and they have been working very well for me, £15 - £20 a spool depending on colour.

Thanks

Alex

[magicniner]

You guys should certainly look at PETG filaments

Thanks for the suggestion, I have some samples of PETG to try once I've got the Nylon parts in production.

[Ross77]

Too bad... never the less, at that price I'd expect that something needs to be fixed

Its always a risk buying secondhand but yes it was cheap enough. Having said that I've only replaced a few screws, oiled it and changed a few setting and the print quality is much better. (mainly slowed it down and dropped the hot end temp 15 degrees) At least in happy with it now. Still using the PLA that came with it so hopefully it will get better with better filments.


What do you mean is the issue with the bowden extruder? The printer I am planning to build is planned to have bowden type of extruder, so I'd be interested if there is an issue I did not think about.

I was looking at the e3d aero which is more controllable and can be used for all materials, I think the main issue is that for this machine it is propriety so not easy to change. its really slow on the warm up which is bugging me already.

Having done some quick prototyping tonight I can definitely see the potential so may look at up grading all the electrics, The machine is rock solid so I tthink it will be a good base which will save me a lot of time.

Might even get the lathe finished this year... :0)

[Greeny]

You guys should certainly look at PETG filaments, as easy to print as PLA but as strong as ABS, doesn't absorb moisture either; some types can be stringy but good retraction settings handle that. I have been using "Real Filaments" (Check amazon) and they have been working very well for me, £15 - £20 a spool depending on colour.

Thanks
Alex

I agree, I'm experimenting with a roll of PETG (coincidentally from real filaments!)
I have found it pretty easy to print with, although it can sometimes have small stringy artifacts, but these are easy to scrape off.
It is definately tougher than PLA, but i couldn't quantify by how much.
I think it will be my "go to" filament when strength is a factor, otherwise PLA . I haven't tried ABS yet as I've heard it's smelly and I'm enough trouble as it is!!

Pic of 16mm tube motor mount for my upcoming X8 copter printed in PETG
Cheers

22571

[Zeeflyboy]

ASA is another good option, it's basically got all the properties of ABS but significantly less warp, less odour when printing and better UV resistance. I've tried formfutura's ApolloX with pretty nice results once I added a fan at 30-40%

I've also been having very nice results with formfutura TitanX again with the fan. Basically a modified ABS that seems to have very minimal shrinkage and warping.

Carbonfil is still by some significant margin the most impressive filament I've used in terms of making accurate structural parts with very high strength and dimensional accuracy. Its still expensive and a pain though, so still looking for my ultimate filaments... Edge (PETG type material), Apollo X (ASA) and Titan X (modified ABS) are about as close to ideal as I have yet found.

[Boyan Silyavski]

I print mainly PETG and am looking to print only Nylon in near future due to the fact that for parts that will heat more than 70C that is the only way to go. Talking about functional parts. ABS typically is not strong enough to my liking and is best suited for pieces that will be polished and must have a nice look at the end.

I would have printed only Nylon and Carbon filled Nylon if not for the price... . Ridiculous price i say for the Carbon filled Nylon. All the above are only 15-20 euro per kg top quality so that matters.

[magicniner]

Boyan,
I know what you mean about cost and was initially a bit miffed about the cost of the Taulman Bridge Nylon but I have to say that it is well worth the price for small fine detail using a 0.2 nozzle, with 0.4 and 0.9 nozzles I'm getting good results with Nylon at the normal price per Kg,
Regards

[Boyan Silyavski]

Boyan,
I know what you mean about cost and was initially a bit miffed about the cost of the Taulman Bridge Nylon but I have to say that it is well worth the price for small fine detail using a 0.2 nozzle, with 0.4 and 0.9 nozzles I'm getting good results with Nylon at the normal price per Kg,
Regards

Great to know. Thanks!
I always have a stock from PETG and ABS, as i buy 5x at least, so am still shy to buy Nylon :tan:
Thing is that for functional parts in a car or a sth that will be hit by sunlight here in spain or left in a car or similar, nylon is the way.

[m.marino]

For Sunlight issues Nylon or ASA in a FDM printer will work very well indeed. I have to say for the money a Prusa MK2S or the MK3 is a good kit for the money. I own and run 2 MK2S's and will be upgrading them to the 2.5 model. Don't know IF I will be buying the MK3 as I am working on a design of my own and will see how that goes (larger build volume and using some of the ideas from the Lutzbot/Prusa).

Michael
MM0MSU

[Doddy]

As an observation about - particularly PLA - and "sunlight", albeit here in the UK, I've read about PLA and the low temperature resilience, also its biodegradable nature. Three years ago, I printed in PLA a button stack - a barrel setting for a captive M4 nut to engage with a greenhouse tee-bolt (i.e. hot, humid environment) and an extended lug with a hole to take twine - so that many of these could be set into the framework of a greenhouse to suspend guide wires, etc, to support tomato and other heavy set plants. At the same time, this being an allotment setting, we were required to present plot numbers on gates, etc. Again, PLA plastic, printed numeric plates set onto gate posts. So, despite all the concerns about PLA, these have survived expose direct to the elements and in hot environments (admittedly not Spain) PLA has been surprisingly robust.

[A_Camera]

Doddy, that's good to hear. I think your climate and mine is about the same....

I had plans on modifying my CNC to be able to use it as a 3D printer as well, so I bought all the parts necessary, together with two large rolls of PLA to start with. After some considerations I decided not to continue with the activity, mainly because I don't want to have a "plastic melter" machine in the house just side by side with my sleeping room. Having a CNC in that room is fine, but I am reluctant to have a 3D printer there also, due to the fumes. So I finally decided today, and actually bought a crazy cheap 3D printer.

http://www.ebay.com/itm/Duty-free-Geeetech-Acrylic-Reprap-Prusa-I3-Pro-B-3D-imprimante-MK8-LCD2004/231911254318?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2057872.m2749.l2649

I know it is a toy... :encouragement: ... a plastic fantastic acrylic machine... :yahoo: but it is good enough to test the concept and see if I want to use it more seriously or not. Anyway, if I want to continue using it, which I guess I will, then I know almost certainly how to make it much better, which parts to change and so on. This way I can move the printer out of the house and let it work for hours if needed. Maybe I'll set up a camera so I can watch and supervise it during the printing. Anyway, I'll get it next week and will see if it was worth the money or not.

[AlexDoran]

As an observation about - particularly PLA - and "sunlight", albeit here in the UK, I've read about PLA and the low temperature resilience, also its biodegradable nature. Three years ago, I printed in PLA a button stack - a barrel setting for a captive M4 nut to engage with a greenhouse tee-bolt (i.e. hot, humid environment) and an extended lug with a hole to take twine - so that many of these could be set into the framework of a greenhouse to suspend guide wires, etc, to support tomato and other heavy set plants. At the same time, this being an allotment setting, we were required to present plot numbers on gates, etc. Again, PLA plastic, printed numeric plates set onto gate posts. So, despite all the concerns about PLA, these have survived expose direct to the elements and in hot environments (admittedly not Spain) PLA has been surprisingly robust.

I agree, a lot people who report problems with PLA and its thermal performance seem to be over the pond in the states - phone holders printed for their cars melting in the summer heat etc. I use PLA for a lot of stuff, all of the components on my printer are done in PETG simply because i wanted it all the same colour, and parts on the Extruder Carriage needed to be PETG to survive the heat. I also bought a roll of "Reflect-A-Gold" tape from the states, off the top of my head it can reflect 80% of the heat upto ~500c. Good stuff and seems to work well, my old Blower Fan Shroud used to touch against the hotend, and this tape stopped it melting for well over a year.

Alex

[Boyan Silyavski]

PETG works for cars. This summer tested in Spain, i have printed a holder for the mobile.

I have a digital temp controller oven so i have found that yes, 75C is the temp that the PETG starts to lose rigidity and is able to be reshaped. But one very important thing to understand is that in an oven 75C means the whole piece is heated from all sides so there is no way to cool. While in a car it could be only partly heated or one side heated. So its not the same. In reality i expected a fail but that was the only material i had at hand at that moment
PETG will work also for printer parts inside enclosure, as the desired and achieved in reality temperature of the enclosure is around 40-45C, no more.

But PLA is definitely a fail in the car dashboard or directly hit by sunlight here in Spain. neither it works for printer parts in enclosure.

Hi Temp PLA is absolute crap. tried all the famous brands. Heat treated and so on. No, no and no. Very difficult material. Once you try to treat it it warps tremendously >5% so is unusable in real life.I have f%%d at least 50 pieces but to no avail. So i stopped using it.

So heating to certain temperature in an oven is not the same as in normal use. But FYI under the car hood is an oven so there only Nylon and at worse case ABS. maybe PEEk also.

[AlexDoran]

PETG works for cars. This summer tested in Spain, i have printed a holder for the mobile.

I have a digital temp controller oven so i have found that yes, 75C is the temp that the PETG starts to lose rigidity and is able to be reshaped. But one very important thing to understand is that in an oven 75C means the whole piece is heated from all sides so there is no way to cool. While in a car it could be only partly heated or one side heated. So its not the same. In reality i expected a fail but that was the only material i had at hand at that moment
PETG will work also for printer parts inside enclosure, as the desired and achieved in reality temperature of the enclosure is around 40-45C, no more.

But PLA is definitely a fail in the car dashboard or directly hit by sunlight here in Spain. neither it works for printer parts in enclosure.

Hi Temp PLA is absolute crap. tried all the famous brands. Heat treated and so on. No, no and no. Very difficult material. Once you try to treat it it warps tremendously >5% so is unusable in real life.I have f%%d at least 50 pieces but to no avail. So i stopped using it.

So heating to certain temperature in an oven is not the same as in normal use. But FYI under the car hood is an oven so there only Nylon and at worse case ABS. maybe PEEk also.

You could also try the Colorfabb HT, however i have found anything other than PLA from Colorfabb to be a nightmare to stop warping. I think PEEK might need hotend temps of 350+.

Typically i use Carbon Nylon for extreme stuff, its very expensive, but is very easy to print compared to traditional Nylon as the Carbon fibre makes it very stable and not very prone to warping. I was also sent a sample of a Polycarbonate filament, however it was TOXIC - and the print also de-laminated very quickly as it cooled down.

Alex

[A_Camera]

Doddy, that's good to hear. I think your climate and mine is about the same....

I had plans on modifying my CNC to be able to use it as a 3D printer as well, so I bought all the parts necessary, together with two large rolls of PLA to start with. After some considerations I decided not to continue with the activity, mainly because I don't want to have a "plastic melter" machine in the house just side by side with my sleeping room. Having a CNC in that room is fine, but I am reluctant to have a 3D printer there also, due to the fumes. So I finally decided today, and actually bought a crazy cheap 3D printer.

http://www.ebay.com/itm/Duty-free-Geeetech-Acrylic-Reprap-Prusa-I3-Pro-B-3D-imprimante-MK8-LCD2004/231911254318?ssPageName=STRK%3AMEBIDX%3AIT&_trksid=p2057872.m2749.l2649

I know it is a toy... :encouragement: ... a plastic fantastic acrylic machine... :yahoo: but it is good enough to test the concept and see if I want to use it more seriously or not. Anyway, if I want to continue using it, which I guess I will, then I know almost certainly how to make it much better, which parts to change and so on. This way I can move the printer out of the house and let it work for hours if needed. Maybe I'll set up a camera so I can watch and supervise it during the printing. Anyway, I'll get it next week and will see if it was worth the money or not.

Just an update in case someone is interested...

I bought the above printer, it arrived a few days later, no issues. Installation took a few hours with some help from the Geeetech official installation videos found on Youtube. Of course, based on my CNC-building activities, I knew that spending time on squaring and alignment worth every minute, so I spent about a week (not 24/7) aligning everything possible. Here is a short video of the assembly and some initial printing. It is not an instruction video, just showing some steps of my progress.


https://youtu.be/XcHv82eBWo4

Printing was fine but slow, so just a week or two after I assembled the printer I made some quick and simple modifications. Actually, some of the modifications were implemented already during the original assembly, like replacing the PSU and the heat bed, but the major work was done during a weekend. I also made a short video of the modifications after they were done.


https://youtu.be/aq4qBsM0jPs

The printer is now "as good as it gets". It is fast and prints really nice, so considering the money and the time I spent with it, it was well worth the efforts. However, I would NOT recommend to anyone who expects a perfect machine right out of the box and is not ready to spend time on improving it. But it is a very good starter machine for a person who is handy and can make improvements, or a person who just want to have a 3D printer to play with and to learn some principles of 3D design and printing.

So the next step...

Currently I am in the process of building my own from scratch. It is a 100% aluminium frame, very robust and rigid design. Perhaps overkill for a 3D printer, but there is no harm in building it stronger than absolutely necessary.

[cropwell]

I printed a couple of dozen shaped washers for a customer/friend and the last two I left to print overnight and went to bed. In the morning I found the print had come unstuck and the ABS just kept extruding, but due to the unstuck component hanging round the extruder, a great mass of ABS had accumulated
2429524296
So I am just preheating the nozzle and will leave it for half an hour to see if it gets liquid enough to come away without damaging the heat blanket.

I made a couple of mods to the printer. The first one was to fit stainless steel extruder gears, but this proved wrong. The stainless does not have the self cleaning property that the original brass has, so any minor filament jam just grinds a notch in the filament and the dross just stays on the gear and requires disassembly to get to it to clean.

I have also replaced the 40mm M3 bolts that hold the extruder motors, extruder, heat sink and fan on to the bar with the nozzles. I used 50mm M3 studs and wing nuts, so that I can remove the heat sink to get to the extruder gear without having to take the extruder motor off the frame (cos it is a bugger to get it back).

The other thing I did was make a jig for setting the height of the nozzles with the nozzle bar out of the printer. It saves a hell of a lot of hassle later on in assembly, when you can't easily get to the nozzle height clamping grub screws.
24297.

I think I may have gotten away without having to replace the heat blanket :joyous:
2429824299

Cheers M'dears !

[Chaz]

So, IVe got one of my original Mendelmax printers still but its a bit tired.

Can anyone suggest what I'd look at today for a 'decent' budget printer? Dont mind a bit of fettling to get it 100% right as long as things like the extruder work reliably (this always seems to be the main issue on most printers).

Thanks

[AlexDoran]

So, IVe got one of my original Mendelmax printers still but its a bit tired.

Can anyone suggest what I'd look at today for a 'decent' budget printer? Dont mind a bit of fettling to get it 100% right as long as things like the extruder work reliably (this always seems to be the main issue on most printers).

Thanks

Im slowly compiling parts to build a "Voron", http://vorondesign.com/voron2.1

Alex

[Chaz]

Im slowly compiling parts to build a "Voron", http://vorondesign.com/voron2.1

Alex

Thanks, ill look them up, are they good? What type of cost?

[AlexDoran]

Thanks, ill look them up, are they good? What type of cost?

It's a premium DIY build, iirc (probably i dont), the BOM is going to set you back close to ~800, most of the guys building them dont have access to a Mill so i assume you will save a decent amount on having someone machine the bed etc. If you have a discord account they have a good community there.

Alex

[Chaz]

It's a premium DIY build, iirc (probably i dont), the BOM is going to set you back close to ~800, most of the guys building them dont have access to a Mill so i assume you will save a decent amount on having someone machine the bed etc. If you have a discord account they have a good community there.

Alex

Thanks. I have a few machines so can make stuff as needed.

[AlexDoran]

Thanks. I have a few machines so can make stuff as needed.

The main attraction to me is that these guys are printing close to 140mm/s with good quality, and have a decent build volume (350mm2).

[Chaz]

Does anyone sell these as ready to go? I dont have the time to source the hardware and possibly do the build.

[AlexDoran]

Does anyone sell these as ready to go? I dont have the time to source the hardware and possibly do the build.

to my knowledge no, might be worth jumping on the reddit or discord and asking the question though.

[AndyUK]

So, IVe got one of my original Mendelmax printers still but its a bit tired.

Can anyone suggest what I'd look at today for a 'decent' budget printer? Dont mind a bit of fettling to get it 100% right as long as things like the extruder work reliably (this always seems to be the main issue on most printers).

Thanks

I'm reasonably happy with the 2 Creality CR-10S printers in the family. Circa £350 6 months ago. Printing within an hour of opening the box. Only fault we've had so far is the SD card slot can be a bit dodgy recognizing the card, which can start a session with 10mins of faffing taking the card in and out (I probably just need to reflow the solder on the control board) - but easily solved by using a raspberry pi and octoprint.

Otherwise I've also heard good things about the Ender3 and ANET A8.