3D Printer On sale in Currys for under £1,200

[John Stevenson 1Participant @johnstevenson1]

In the red corner we have Andrew Johnson with 1446 forum posts to his name and 208 photo's, most of which we have all seen in his many posts.

He has posted the results of a 25mm cube as 25.00 x 24.99 x 25.03. A mean difference of 0.04mm or one and a half thou. A measurement I'm sure most of the readers on here would kill to get off their milling machines repeatedly.

In the blue corner we have Peter Edwards with a total of 6 forum posts, all in this thread, no pictures and no practical examples but who thinks that these results are utter rubbish.

I will leave you all to form your own opinions but I think that Jason and myself will have to do a bit of checking of ISP address in the background because call me a suspicious person.

SUSPICIOUS PERSON !!

Who said that ?

But I think I can smell a troll 5 bridges away.

[jason udallParticipant @jasonudall57142]

John I don’t know .
But the point made of difficulty in finding output examples esp. from “manufactures” is true.
If any one goes into 3d printing (fused filament ) expecting what they see of plastic components in every day life..well they will be suprised if not disappointed. .but imagine milling a cube with a ball mill say 1mm dia…well you get the picture..for me one of the disappointment s is the faff when changing filament..different stuff spreads differently and at different temperature. .and some stock shows the filament more….
But the joy of printing two parts that fit each other just so…

[Anonymous]

Overall I think the results are pretty good, and are certainly better than I was expecting. As Muzzer has pointed out injection moulding and diecasting are not an exact science and require tweaks, as does CNC if you're looking for the last bit of accuracy. So I see no reason why additive printing should be any different. The nozzle is 0.4mm diameter, so essentially the part is built up with 0.4mm wide lines, a bit like icing a cake. There is also shrinkage to take into account; getting a feel for that is part of the learning process. I haven't done any quantative measurements yet, but subjectively I think that the repeatability is pretty good. That's encouraging, as absolute errors can then be corrected for; it's a bit more difficult when errors are random!

There are also some wavy patterns on some of the flat faces, I suspect this is to do with the dynamics of the axis controls. Compared to small CNC mills the head is moving pretty fast. So the same issues to do with acceleration and deceleration that affect CNC mills will also affect the printer.

There is a small stereolithographic printer 'available' for the home market, the Form1. I decided against it for the following reasons. It is quite a bit more expensive than the printer I eventually bought. The print volume is quite small, and smaller than I need. The resins needed are expensive, and I suspect have finite shelf life. I can see it being a right pain having to clean out the system in case the resin goes off. Finally it is not quite available yet: the company are asking for cloud funding. I needed a printer now, not at some unspecified point in the future.

This is the way I see it:

Will my 3D printer compete with a stereolithographic printer? No, but then its a lot cheaper, and the better resolution and finish is of no consequence to me for my applications.

Is the dimensional accuracy, finish and repeatability of the printer adequate for my needs? Very definitely, I don't need super accuracy, I can compensate for repeatable errors, and the skill in engineering is in designing parts that go together without needing tight tolerances.

Am I pleased with the printer so far? You bet! Definitely one of my better buys and I already have several ideas for parts to help with my traction engine build, irrespective of what I might design for work.

I don't think I'm breaking any confidences by saying that it was Peter who PM'd me and asked if I could show some close ups of the surfaces. I thought it might be a 'feed', and from the response so it would seem.   However, since I was going to do the measurements anyway I thought I might as well publish the results.

Regards,

Andrew

Edited By Andrew Johnston on 09/01/2014 21:52:03

[John Stevenson 1Participant @johnstevenson1]

Well i haven't printed a cube to check it, I may in the future just for my own sake but I'm sure i won't get better than Andrews, in fact I'm guessing it won't be as good.

His machine seems to have more control than mine, for instance I have layers in 0.15, 0.2, 0.3, 0.4 and 0.5 and then the software seems to make most of the other choices like temperature etc.

As regards finding working examples it's not hard. i researched this subject for about 2 – 3 month off and on. In that time I saw 3 repraps actually working, results from 2 more, results from the UP! and one working [thanks Steve ] and one of the RS units over at Coventry so they are out there if you do your homework.

Anyway over the last couple of nights I have sorted the file out for the part I want, took me longer than I thought but had some licensing issues with Alibre requiring help from the States.

Anyway got the file, printer said 2 hours 45 minutes to print it in fine mode so printed it.

Came out very well and to be honest I couldn't have made such a nice job milling it. decided to leave the surface finish as is because the first alloy part after cleaning up will be the actual pattern. Took it up to the foundry this afternoon, only a small foundry and you deal with the owner.

He was mega impressed, first time he's seen a printed pattern,plenty of wax patterns but not a straight print.

Now this is where it gets interesting, remember the post where I reckoned up if the laser would pay for itself ?10% work horse, 90% John's toy ?

After seeing the pattern he dug round in the store room and come up with some bronze badges a bit like horse brasses but very worn, rounded edges, text nearly worn off.

Did I want the job to print some patterns for these, good paying work.

I declined at the moment for two reason, one is i have some family commitments at the moment that limit my time and secondly and more important i don't feel competent enough at the moment to develop the drawings. After so long on 2D CAD I find 3D very hard but that's just me.

Anyway i asked him if he could work to one and a half thou doing the casting. I won't repeat his reply

[Peter Edwards 5Participant @peteredwards5]

Moderator – thanks for suspecting me of 'Trolling' . Andrew says his surface finish is '10 times worse' than machining. I would go along with this. For some applications the results he shows will be OK, but for small parts for modelling would that be true?. I am sincerely trying to establish the practicality of the present range of relatively cheap desktop 3D printers in the absence of a practical demonstration.

I have long experience of solidworks and stereolithography (which produces excellent results). I also have 40 years experience in engineering and university lecturing in the subject, so moderator perhaps you should think before you go in for your accusations.

[Anonymous]

Here are the results of some more trials. I printed a 50mm cube last night, I measured the height just after the print had finished at 50.01mm. An hour later it had shrunk. The final measurements (W x D x H) were:

49.81 x 49.75 x 49.88

I had a gut feeling this may be due to shrinkage. So this morning for each axis I calculated the change needed to bring the part to a 50mm cube, altered the scale factors in the slicing software, and re-ran the part. The results I got were:

49.99 x 49.94 x 49.99

Not quite exact, but nevertheless not bad. The percentage changes needed were 0.35%, 0.5% and 0.24%. I found some commercial injection moulding data on the internet for PLA which gives a shrinkage range of 0.3% to 1.1%.

I also tried a couple of new materials. Changing material was much easier than I expected. It is all driven from the front panel. When you select change material the machine heats the nozzle and then automatically retracts the filament. You then install the new filament and the machine advances it and extrudes until you say stop, ie, when the old material left in the nozzle has been extruded. I tried a PLA/PHA mix which is supposed to be a bit better than straight PLA, and in particular less brittle. I also tried a plastic called XT, which is transparent, and has a higher Tg than PLA, which may be important for my application. I made the same part in each, a 25mm square box:

Sorry about the garish colour of the part on the left; I used the samples that I was sent by my supplier. Possibly the magenta isn't a good seller, which is why they're using it for samples. I haven't done any quantative tests on these parts. However, judging by the way the material change went the PLA/PHA is a bit more prone to stringing than the PLA. I'm not sure if that says anything useful about its properties.

Peter: Out of curiosity which university are/were you at? In my business life we've been dealing with a couple of universities (on gas sensing) that do extensive work with 3D printing for sensors. Certainly the stereolithographic parts I've seen have had an excellent finish. I found an interesting paper on the net by academics at USC and Harvard about a mathematical model of the shrinkage in stereolithographic resins. If I've read the paper correctly the shrinkages seemed quite large, tens of thou on a 3" diameter test piece. What's your experience of this; does it exist, and if so is it a problem?

Regards,

Andrew

[Peter Edwards 5Participant @peteredwards5]

Andrew. I have PM'd you on your latest post.

[Anonymous]

This morning, while out for a cycle ride I let the printer make another 50mm cube, with the same code as the previous corrected cube. I measured the cube after removing it, but before it was properly cool. Results (W x D x H) were:

50.08 x 49.97 x 50.03

After an hour or so I measured it again and got:

49.99 x 49.90 x 49.89

Compared to the previous cube made with the same code the differences are 0.0, 0.04 and 0.1mm. I'm pretty happy with the W and D; I don't expect my CNC mill to hold much better repeatability. I'm a bit mystified about the error in the height. However, I did change the filament (and plastic type) before printing the cube. So it is possible I didn't clear the nozzle properly. If the first layer was messed up it could easily have an effect on the height. Tomorrow I'll run four 25mm cubes at the same time, the slicing software gives the option to pattern the model across the build platform, to see what the repeatability is like. If nothing else it's something else I haven't tried before.

Regards,

Andrew

[John Stevenson 1Participant @johnstevenson1]

Andrew,

I think the results you are getting are very good, in fact better that I would have expected.

Biggest decision I can see is "can you make use of the technology" For some, Andrew and myself, I think we have done our homework and the answer is yes.

There will be a lot of owners thought who buy one without thinking it out given all the media hype and this is where a bargain is to be had when they realise this and part with it.

[MuzzerParticipant @muzzer]

My company (of which I am the engineering director) designs quite a few diecast and moulded components, connectors and housings. We use these in the manufacture of finished products in volumes of over 150k parts per year so we like to have a good idea that the parts we design are fit for purpose. Currently if we need prototype samples of intended volume components we have to have them machined up from solid (either metal or plastic) – or printed by an external partner. It often takes the best part of a week for the CNC machines to be programmed before they even start to make the parts.We certainly wouldn't find it easy to justify buying our own full blown stereo lithography machine but at this price point it makes sense to go for one of these 3D printers in-house. It's very helpful to see Andrew and JohnS giving them a good workout and sharing their thoughts and findings with the rest of us – for which many thanks.

We could be handling and using a usable connector, cover or winding bobbin within an hour or so of completing the 3D model. Certainly the tolerances and surface finish are perfectly adequate for our purposes. Used appropriately, I see a lot of real benefits for a very reasonable cost. Obviously it may also come in handy for turning out our own parts – we will be happy for our work colleagues to do so (within reason), after all we like to have a bit of fun!!

Muzzer

[John Stevenson 1Participant @johnstevenson1]

Forget the making of a part for a pattern and go to the finished process.

 

http://www.youtube.com/watch?v=LFGnjkoeDng

 

OK I know, not home shop but then again neither were CNC's 10 years ago

 

edit, clickable link added.

Edited By John Stevenson on 11/01/2014 22:30:22

[V8EngParticipant @v8eng]

Just an idle thought:

I wonder if there is some future for 3D printers in the ceramics industry, i.e. CAD a master, 3D print it, use that for producing the plaster moulds.

Sorry not at all ME related, just a bit of late night musing.

Edited By V8Eng on 11/01/2014 22:47:06

[BazyleParticipant @bazyle]

On Saturday I was talking to someone who until last year had been working at a non-ferrous foundry near London and he said printed patterns were already quite common because of the phenominal cost of hand made woodern patterns.
I imagine also it means that design modifications are also much easier and cheaper. This in turn probably means that in due course you will find spare parts while still form fit function are less likely to look like the original as they cost reduce the part on each batch.

[David JuppParticipant @davidjupp51506]

Posted by V8Eng on 11/01/2014 22:46:16:

Just an idle thought:

I wonder if there is some future for 3D printers in the ceramics industry, i.e. CAD a master, 3D print it, use that for producing the plaster moulds.

Sorry not at all ME related, just a bit of late night musing.

Edited By V8Eng on 11/01/2014 22:47:06

3D Systems launched the 'Cerajet' printer for ceramics at the CES last week – it's covered in the same publicity video as the food printers and a host of other stuff…

This one is actually direct 3D printing of the clay, not the moulds – certainly suggests your idea is perfectly feasible.

[V8EngParticipant @v8eng]

Looks like I was at least one stage behind, such is the pace of progress!

I will have to look this one up, having worked with clay on an amateur basis I am amazed at the idea.

The downside of course is likely to be an even greater loss of jobs.

[richardandtracyParticipant @richardandtracy]

It may mean a greater loss of jobs.

But where?

I suspect that companies in the UK that take on the process will at least keep their staff while production ramps up. It's in areas that don't adopt that are likely to loose staff in the future. Maybe, due to the cost of investment, that China won't be picking up the production jobs.

This technology will, at some point, lead to a total disruption of manufacturing industry as currently set up, and I really can't read how it's going to go. Wish I could. It would be nice to be on the crest as it happens! (for once).

Regards,

Richard

[Russell EberhardtParticipant @russelleberhardt48058]

Posted by richardandtracy on 14/01/2014 08:27:31:

Maybe, due to the cost of investment, that China won't be picking up the production jobs.

I wouldn't be too sure. China can afford the investment better than the UK. It depends more on the end manufacturing cost. I once visited an electronics factory in Shenzen province that had a room containing about 50 Japanese pick and place machines all standing idle. The next room was full with hundreds of girls of about 16 years doing the job by hand because it was cheaper that way!

Russell.

[Involute CurveParticipant @involutecurve]

I personally think high res 3D printing will remain illusive a bit like CNC milling was, and yes this has become available but only to enthusiasts and technical hobbyists, or geeks, if you prefer, I know the model engineering fraternity etc. etc. etc. will embrace this stuff but I still cant see it becoming mass produced, in anywhere near the numbers needed to bring the price down much lower than now, I do think the ones available will get better, I know my grand kids would look at the current ones with disdain, not because of resolution but because of speed or lack of it, if it cant print in a couple of minuets it will burn, unless of course it can print a Big Mac to feed there fat little faces with…………… ;–)

I do hope I'm wrong about some of the above, and I do hope they end up on everyone's desk, I just don't see it, I currently use IPF for all my 3D printing this is the higher end of this technology and very good but expensive, a look around there website http://ipfl.co.uk/ gives a little insight into the current capabilities of this very interesting technology, I have no affiliation to this company other than as a customer.

Shaun

[richardandtracyParticipant @richardandtracy]

Russell,

You could be right. And there are all the other costs that exist in the UK.

Planning costs, building costs, bank reluctance to approve investment etc. all increase the hidden costs behind such things to a surprising degree. It feels as if the entire system it trying to stifle enterprise because things will be much easier for the bureaucrats if no-one does anything.

Hmm. Reminds me of a very silly & salutary story. Last year the company I work for had two main accounts with the bank. One an 'income account' and another the 'outgoings account'. The company is quite profitable at the moment (>£20k/year.employee, which isn't too bad for an engineering company in the current climate) and had a 7 figure surplus in the 'income account' when the bank closed the 4 figure overdraft on the 'outgoings account' on the basis there was no prospect of it ever being paid off. The bank also started insolvency proceedings against the company to try to recover the 'unpayable' four figure debt – both the insolvency action and account closure occurred on the same day. 6 hours later the bank had lost its customer and the company had changed bank to a more amenable organisation – which has its headquarters in a country outside both Scotland in particular and the UK in general because UK based banks seem to have a catastrophically shortsighted view on investment and loans.

Regards,

Richard

Edited By richardandtracy on 14/01/2014 10:39:26

[MuzzerParticipant @muzzer]

BTW, if you have an hour or two spare, here's an interesting free book(?) in PDF that has a whole load of stuff about 3D printers and where things are going. Mostly for the amateur but some interesting information also about CNC micromills, laser cutters etc. **LINK**

Dunno what the guy does but he seems to have spent a few hours typing stuff out! Pretty comprehensive state of the nation information though, nicely researched.

As these printers are essentially lightweight 3-axis CNC machines running g-code and come with the necessary controllers and compilers, there is a lot of interesting development going on here. This can only speed up the development of CNC technology and its availability to the likes of us – and the broader masses in general. Bring it on.

Muzzer

PS – Since this morning, the PDF has been removed for moderation. Hope it returns soon, otherwise I can post a link to a copy on my Google Drive or Skydrive.

Edited By Muzzer on 14/01/2014 20:14:25

[Gone AwayParticipant @goneaway]

Posted by Muzzer on 14/01/2014 20:08:19:

PS – Since this morning, the PDF has been removed for moderation. Hope it returns soon, otherwise I can post a link to a copy on my Google Drive or Skydrive.

 

It's here by the look of it but you need to create a login to get it.

Edited By OMG on 15/01/2014 01:07:49

[MuzzerParticipant @muzzer]

I've put it on my Google Drive here **LINK**

Once you've clicked on the link, you will see a download button at the top left next to the printer icon. It'll be easier to browse a local copy after downloading it from Google Drive (unless you are too paranoid to do so).

It's 13MB, whether it's zipped or not. Quite a bit of interesting stuff in Part II (page 145 onwards). This guy knows how to Google!

Edited By Muzzer on 15/01/2014 04:14:09

[Michael GilliganParticipant @michaelgilligan61133]

Muzzer,

Thanks for posting the link

I've only skimmed through it, but it looks well worth studying.

… His previous publication is downloadable here.

MichaelG.

[MM57Participant @mm57]

Very interesting doc – thanks for finding it. Plenty of bedtime reading there…

[BazyleParticipant @bazyle]

Key factor is timescales.

In '45 one computer existed in the UK and it didn't know how to play teletype startreck. By '77 I bought a 6800 micro for £300 (inlation adjusted) that could do more faster. I can now buy that processing for £1 or get a ten thousand times more powerful one for my £300.

But it took 60 years.

I won't be here to see super printers but good luck you ten year olds.

[Author]

Posts