Printed 3D metal better than a Casting

[JasonBModerator @jasonb]

As it has been a bit quiet here for the last week or so I thought I would post this example of 3D metal printing.

One of the members over on MEM recently started building a Stuart Williamson Engine. He did not get too far as he decided to start with the fluted column. As you can see the aluminium casting is not one of the finest examples of the foundryman’s art with poor definition and more holes than a lump of Swiss cheese. Things did not improve much after several hours of fettling.

I mentioned to him that I already had the column drawn in CAD for a possible future “Big Willey” build. After rescalling back down to the original size and adding a 0.5mm machining allowance to the top and bottom I e-mailed the .STP file to him.

It was an easy job to upload the file to Craftcloud and get an instant quote which was 2/3rds that of a replacement stuart casting and no gaurantee that the replacement would be perfect.

I have just seen some photos of the finished column, which looks to have come out extreamly well.

I had intended to fabricate the larger version of the column if and when I decide to make it but 3D printing is now looking a very attractive alternative and price wise comprable to thick wall tube etc needed to make it.

[BazyleParticipant @bazyle]

Looks very clean if that is the print rather than a CAD rendition – sometimes can’t tell these days.
Do any of the Craftcloud respondents come from the UK? It seems postage is often up to half the bill and a local supplier would be great for people who live close enough to pickup. Maybe in 20 years time every town will have one like they all used to have foundries.

[Jim NicParticipant @jimnic]

Two things spring to mind after looking at these pictures

The first is that the casting as supplied is a disgrace and no self respecting company should send it out to a paying customer.

The second is that the printed item seems to be the way forward for model engineering.  From Jason’s previous post on metal printing the prints are dimensionally accurate and from the picture above the finish is superb.   All we need is, as Bazyle says, more suppliers.

Jim

[Clive FosterParticipant @clivefoster55965]

Looks very good.

I think it’s almost inevitable that 3D metal printing will almost universally sufficiently approach price parity with castings to the final purchaser and displace them. Small castings are something of pain when it comes to quality at the best of times due to inevitable issues with metal flow and cooling rate from the small size. Small numbers and occasional orders as opposed to series production with optimised processes just makes it ten times worse. Lost wax and similar give good finishes and accurate dimensions at the cost of extra faffing around which adds to the price. The amount of annual labour in casting with no realistic way cutting down inevitably keeps prices up.

Realistically the metal costs of our sizes of casting are only a small part of what you pay for. It’s all the handling, faffing around, patterns, storage costs involved in keeping stock et al that eats up the money. Send file, print, send back minimises faffing not to mention the cost of holding stock. How many castings of one design does a supplier sell in a year anyway? Even for popular designs it can’t be that many by normal retail supply standards. I’d lay odds that folk like Blackgates, Reeves et al with a goodly number of designs on offer may sell only single digit numbers, possibly less, of several in an average year. Which really isn’t a sustainable business model.

Acting primarily as an order taker for a 3D print shop on a fortnight, or possibly less, delivery looks far more sustainable. Compared to a small, old established, casting firm a 3D printer ties up vastly more capital but it doesn’t care what they ask it to do and doesn’t need all the manual work. Basically just machine time and investment capital amortisation.

A 3D print via CAD is much less work than patterns too. For legacy designs conversion to 3D will be needed if printing is to be possible but I imagine AI is approaching the stage where it could do the heavy lifting by scanning original drawings or maybe simply scanning the patterns would do.

Clive

[Russell EberhardtParticipant @russelleberhardt48058]

Just wondering how strong it would be as opposed to a traditional casting.  As it would appear to be a form of sintering I would expect it to be a bit more brittle – but what do I know?

Russell

[JasonBModerator @jasonb]

Yes that is the print as recieved, this is a quick 3D render of the STL file

I’m not sure more local printing would be any cheaper. On only needs to look at the range of quotes that Craftcloud gives you to see that the current source is by far the cheaper. He could have opted to have it printed in Germany but at about £450.00  not what most people want to spend on their hobby. Labour, rent and overheads are all going to be a lot more locally. Turnaround would have been about the same from either source unless an express service was chosen adding more to the cost.

Depending on the part 3D printing can provide a saving as in a case like this no core is needed to form the main vertical hole or the windows. Also as there is no need to pull a pattern from the sand draft angles become less of an issue, that column would not pull out of a two piece mould, more likely it would need splitting into four. All extra work at the foundry = cost. Though it does depend on how the foundry prices work. I was using one that for simple parts charged by the Kg but they went down hill. Th elatest seem to have a charge per item which is the majority of the cost and the rest is made of with material cost so small parts can seem more costly than large.

[Russell Eberhardt]

On Russell Eberhardt Said:

…would appear to be a form of sintering I would expect it to be a bit more brittle – but what do I know?

Funny you should mention that. I am currently doing machining and tensile strength trials on some 3D printed stainless steel test pieces. The plan is to write up the results for ME&W, editor willing of course.

All I am going to say at this point is that I am astounded, in a good sense, by the outcome of the tests so far; not at all what was expecting.

Julie

[JasonBModerator @jasonb]

Look forward to reading that Julie.

From my own rough and ready pull out tests on an M4 x 0.5 thread in printed aluminium I too was quite impressed with the strength.

The figures given on the craftcloud site for the alloy used are better than those of solid 6062 T6. Other suppliers give similar figures. Even printed 6061 is given as beingas strong as solid 6061 T6

Stuarts are currently casting these in iron, possibly from the same pattern. I bet if you dropped the printed one and an iron replacement from the same height the iron would be more brittle and likely to break.

[John HaineParticipant @johnhaine32865]

A participant in the last SMEE digital group zoom showed some loco parts he’d had printed by Craftcloud.  From what I recall the cost was competitive with castings (but made in stainless).  When delivered they had an extra rod which was presumably added as a support for printing – he sawed it off and (with a lot of trouble) bent it in a right angle in a vice, with no cracking.  If I was an ME casting supplier I think I’d either give up or get all my designs converted to 3D models and strike up a relationship with a 3D metal printing company in China.  The only value they have is the intellectual property in the designs, and (hopefully) their reputation.

[JasonBModerator @jasonb]

Well Reeves are scanning all their patterns and also offer a scanning service.

 

Not sure I would want to go down that route as some of the patterns may be a bit tired by now and going buy castings I have had from many sources they can be a bit out such as not quite symmetrical or with bosses in the wrong position. Also others where detail had been left off no doudt to make casting easier and more economical.

So possibly better to start from scratch with a new 3D CAD file and hopefully end up with a part that is even better than the castings. At the moment one of the problems is they don’t seem to be able to offer iron as a material. Brass and bronze have to have “waxes” printed and are then cast so cost is a bit higher due to the two operations though it may still be worth while depending on the part. Certainly easy enough to upload a file and get an instant idea of what the costs would be and make a decision based on that.

[David JuppParticipant @davidjupp51506]

Mechanical properties of direct 3D prints should be very good because the size of the instantaneous melt pool is very small.  Not unusual for properties to be close to those of forged parts.

The range of materials that can be printed increases rapidly, those with high thermal conductivity were problematic, but becoming less so.

[Julie Ann]

On Julie Ann Said:

ooFunny you should mention that. I am currently doing machining and tensile strength trials on some 3D printed stainless steel test pieces. The plan is to write up the results for ME&W, editor willing of course. […]

This could be a ‘defining moment’ in the history of Model Engineering … and the magazine.

MichaelG.

[DiogenesParticipant @diogenes]

Be nice to think that the current advances in both hard- and soft- ware we are witnessing really will result in another ‘Golden Age’ for making / modelling / home-engineering – it certainly has exciting potential..

The biggest bugbear I can see is how the authorities will respond to the problems of policing ‘mail order’ engineering..

[Diogenes]

On Diogenes Said:

 

The biggest bugbear I can see is how the authorities will respond to the problems of policing ‘mail order’ engineering..

Probably my introducing taxes.  France has just imposed a €2 “Taxe sur les petits colis” (TPC) on non-EU, low-value (under €150) parcels, targeting Chinese e-commerce like Shein and Temu. This mandatory, temporary, and separate administrative fee applies to each item category, not just the package, with an additional €8 fee for non-compliant shipments.  I’m sure others will follow.

Russell

[JasonBModerator @jasonb]

Certanly in the case of Craftcloud the price includes VAT, as you are dealing with a German company not the Chinese (or other) print house. Not sure how that would work out in France as Germany is in the EU.

Well I sent another file off last night, this time something else for me.

The current engine needs so pipe fittings for the cooling water to suit 6mm pipe. I could have used the PM Research cast ones but the advantage of these printed ones are that they will include smooth sweeping internal bores and will also be counterbored at the ends ready to thread M6 x 0.75. rather than being solid castings

The larger 135deg bends are for the exhaust. I had wasted money buying these along with some 1/8″ BSP brass elbows. The Elbows were fine but it seems that due to lack of demand for the 135 bends they jusy use the larger casting for 1/4″ BSP but machine it to 1/8″ BSP which gives an ugly oversize fitting. These will actually be threaded M10x1 so are counterbored 9mm. ( I rethread the BSP to M10)

Pricing was quite interesting. At first I just drew a single 6mm 90deg elbow and got a quote for one, 5 and 10 all of which were quite expensive.

I then joined a few together with a simple 2mm wide spacer between each and the prices started to look better. Adding the Tee and two sizes of 135 bends completed what I needed so I uploaded the STL as shown above. Clicking on the quantity button the cost to have four sets printed was only £2 more than a single set so it would have been rude not to get my monies worth. I also had a discount code from my previous order so got the four sets for just under £50. Better than buying PM Research 1/4″ trees and the BSP from BES plumbing supplies.

[Roderick JenkinsParticipant @roderickjenkins93242]

Very impressive, both the Williamson column and Jason’s elbows.  This surely is the way forward.  Stuart at Bridport foundry only seem able to do traditional sand casting, not the shell moulding that the original Stuart Turner company in Henley used to use which gave a much finer finish.

I’m not sure “Inspector Meticulous” would have approved though.  I’m guessing he would have said “If the original was cast then the model version should also be cast”.  However, Inspector Meticulous and the judging he used to do at exhibitions is pretty much gone and good riddance too.  Personally, I don’t think model engineering is a competitive sport.

Rod

[JasonBModerator @jasonb]

That would be a sure fire way to discourage people from showing their models.

It would be interesting to know how 3D prints would fall in the judging criteria? I feel they should fall some way between making your own patterns & doing your own casting at one end of the scale and buying in readily available castings at the other. Certainly more effort on the part of the builder to do the CAD work without which there would be no 3D print than just buying a casting.

At the moment I’m not sure it is the way to go with larger models, I recently drew up this pump that stands 275mm with twin 9″ fast & loose pullies to the ctr of the crankshaft and 3D printing was really putting the cost beyond what the majority of people would want to spend. I have not done the sums for a fabricated version which may be an aternative as there are a lot of similarities to the Cameron pump I di dsome years back.

Though as can now be seen in the Model Figure painting world you can easily alter the scale at the click of a button.

[JasonBModerator @jasonb]

As another example of 3D metal printimg, Sanjay, who I gave the column file to also has a box bed printed for a very old Stuart 10H which is a different casting to the current one and something that you are unlikely to ever find to replace a missing one.

He also had this brass nameplate done which Craftcloud produce by a combination of a 3D printed “wax” which is then investment cast in brass. It will be interesting to see some of the results from the Cambridge Casting Trials particularly if Julie tries some printed “waxes” using the likes of Polycast filament.

 

[JasonB]

On JasonB Said:

Certanly in the case of Craftcloud the price includes VAT, as you are dealing with a German company not the Chinese (or other) print house. Not sure how that would work out in France as Germany is in the EU.

Well I sent another file off last night, this time something else for me.

 

I have ordered a new 3D printer from Germany and they have charged me VAT at the French rate of 20% as opposed to the German rate of 19%.  I believe they have to charge at the rate of the country they deliver to and credit the VAT to that country.  When I buy from the UK I am not charged UK VAT but when the parcel arrives I am charged French VAT, customs duty, and an exorbitant collection fee by the Post Office!  Well done Brexit.

Russell

 

[JasonBModerator @jasonb]

Just a  quick update as Sanjay who I sent the file to has got his Williamson to the stage where it could be test run. Most of the other parts were made while waiting for the print to arrive. I’d be tempted to leave the column in bare metal and maybe black paint for the other parts.

[DiogenesParticipant @diogenes]

Nice..

[DiogenesParticipant @diogenes]

..certainly going to revolutionise our kind of model-making..

Jason, if you are freed from the tyranny of fabrication you will be be able to ‘do’ an engine a day!

[Roderick JenkinsParticipant @roderickjenkins93242]

The whole thing does indeed look very nice.

Rod

[Andrew CrowParticipant @andrewcrow91475]

A lovely looking and well made engine, the 3D printed column is first class.

Andy

[Diogenes]

On Diogenes Said:

..certainly going to revolutionise our kind of model-making..

Jason, if you are freed from the tyranny of fabrication you will be be able to ‘do’ an engine a day!

I don’t know about tyranny, I still quite enjoy a bit of fabrication along with cutting from solid so I don’t think I need worry too soon about running out of shelf space.

Just like the CNC if you look at 3D metal printing as just another tool and then decide if it is the best option for the job in hand. I see it being a good option where a part may have a lot of internal cavities that would traditionally have been formed by cores as they are not easy to machine if you can’t get access to the inside. So things like IC engine crankcases where you may want transfer ports or water passages.

[Author]

Posts