3D printing

[wefalck]

Just phantastic ... I see the days coming, when I have to scrap all my lathes and milling machines ...

[Fliger747]

I believed that we earlier made the comparison to 3D printing to casting techniques. For display models it is a good method to produce good surface replicas. For stuff that actually needs precision to operate well, the other stuff isn't obsolete. Currently hybrid techniques remove some limitations as to size or material strength. Really only being able to print 5 sides of the six sided box is a limitation. The technology evolves quickly so time will tell!

Cheers: Tom

[wefalck]

Yes, for items with simple rotational-symmetric geometries and in machineable dimensions, I would probably still prefer the (manual) lathe. More complex geometries could be done on a CNC-lathe (if you have one). The machining would probably result in a crisper and dimensionally more precise shape.

[Fliger747]

CNC would be fun to experiment with. However industrial concerns designing such items as jet engine fuel control units have discovered considerably greater freedom in design by being freed from machining constraints, ie having to have voids accessible from the exterior.

This small 1:120 D-7 I made could certainly be made more precisely by assembling a hundred CNC micro machined parts. It's a deck cargo for a model ship...

Prints as a single piece with considerable interior detail you can't see.

Tom

[DrPR]

I do prefer machining - mill and lathe - to 3D printing. When I was working I had access to a shop with large and small lathes and mills. If we had a slow period I would go into the shop and play, creating "mag wheels" for model cars and even machining a steam engine front end and drive cylinders out of a block of aluminum.

To me there is far greater craftsmanship in precision work on a lathe or milling machine than with 3D printing. Here are a couple examples of extraordinary ship models made without 3D printing. These are long posts because these fellows show how they made all the intricate parts for their models. Superb craftsmanship!

Schooner Germania Nova:
https://modelshipworld.com/topic/19848- ... 05204_menu

Cruiser Varyag 1901:
https://modelshipworld.com/topic/19333- ... ent=590551

The Varyag is by a friend in Zaporizhia, Ukraine - near where the Russians shelled the nuclear power plant. The last we heard from him was on 28 February. His extraordinary work is/was destined for a naval museum.

And of course we have Wefalck's outstanding machine work for another example.

Phil

[Fliger747]

Phil:

I wish your friend and family in Ukraine good luck in rather desperate times. Destruction ensues on a large scale. Flying form Hong Kong to London we would stop in Baku for cheap fuel! Thence across the Black Sea and talk to the Ukrainian and Romanian controllers or chat with the lovely Moldovian Gal's in Dubai.

I'll sometimes get into a mood to go out to the shop and make metal shavings on the lathe, yes wheels are fun. Being physical and tactile it's an enjoyable work. Yes the items are quite crisp and "metal" looking but there are limitations in what can be produced with these machines. Surprisingly it my be easier to dimensionally control 3D items.

Theoretically most readers are interested in producing models and parts for models. The deck load D-7 I produced above may not be perfect, but it's dimensionally pretty close and has a lot of hidden details as to transmission, undercarriage, drive train, suspension etc. It would be a considerable challenge to produce one with milling machine and lathe, maybe hundreds of hours? Achieving the internal details would require many parts and watchmaker's assembly. I am certain that this could be done as a traditional tour de force of subtractive technology! Someone have at it!

Regards: Tom

[DrPR]

Tom,

I agree that the purpose of our activities is to make models. Or maybe i should say we do it because we enjoy it. In the long run it doesn't matter much how we accomplish the goal of making the model if we are satisfied with the results.

Your model of the APL is beautiful. And Pascal's entirely 3D printed models are works of art! So in the end the method doesn't matter as much as the results.

I am enthusiastically jumping into 3D printing as a means to create some of the really complex objects on the Okie City model. It would take many more hours to make the parts by hand or on a milling machine or lathe. And considering that I don't have a lathe or milling machine (or a workshop to put them in) they aren't an option right now. So the 3D printer is the means to the end.

And the 3D printer allows me to take direct advantage of the 14 years research and design I put into the CAD model. It wasn't designed for printing, and I am amazed that I can get the results that I am seeing.

But there are two great drawbacks to 3D printing that I can see. First, the parts are extremely fragile, at least in the fine detail. I just broke off part of the antenna support for the Mk37 director by accidentally picking it up the wrong way. I have broken off hand rails, ladder rungs and lift rings from several other parts while trying to work on them. But there is the advantage that I can always print another!

The other problem is that 3D printing is an unknown as far as longevity is concerned. It is not recognized as a suitable method for making museum quality models. I keep thinking of the first kitchen appliance produced with a plastic case. It was a hand mixer, and in less than a year after it was released (with a lot of fanfare as a product for the modern kitchen) the plastic disintegrated into little bits and pieces. My mother had one of the earlier models with a cast pot metal case and it lasted forever.

But I guess the real source of my bias favoring metal comes from the beautiful brass builder's models in the hallways of the classrooms and dorms at Officer Candidate's School. I have spent more than a decade planning how to make the tools and dies for machining and stamping parts for the Okie Boat model. I expected to spend another decade creating all the hundreds of tiny bits and pieces. And now I am turning them out in just a few days, and it just seems too easy!

Phil

[wefalck]

Acrylic resin is actually quite stable, acrylic parts have been now around for decades, approaching some 90 years now - ok not yet 300 or 400 years.

Any plastic that contains plasticisers is a problem, as they will gas out over time and make the material brittle and that indeed over surprisingly short time - I remember a pair of skiing boots that were about a dozen years old at the time when they dropped from my wardrobe and shattered into hundreds of pieces ... btw something similar can also happen with zinc die-cast parts.

I gather our 'mechanical' is going to fight a retrenching battle in certain areas of model-making. It just cannot compete with the level of detail one can achieve with 3D-printing.

Of course, there are other areas of model-making, where not only the final painted piece counts, but that have an aesthetic quality through the materials used and their treatment, such as for most models depicting 'historic' sailing ships. But even there, 3D-printing creeps in.

[Fliger747]

Well yes, metal replicates metal the best! Ah, brass! I remember in the early 60's I was into model railroading and at the local hobby shop that had the fabulous Japanese hand made brass steam locomotives with lost wax castings and beautiful workmanship. I never owned one, an adult toy, price wise, perhaps North of $50? I guess people painted them, so sad to cover up the beautiful brass. I remember seeing for sale some rather beautiful (at least in the ad's) Russian built Iowa Battleships.

As to handling fragile items, we are fortunate that the general strength of materials in miniatures is on our side. As something scales up, the strength of the section more or less follows the square of the increase, but the weight and design load items increase by the cube. If a huge King Kong fist was to pick up a real MK37 all sort of destruction would occur. With 3D, usually (like turtle eggs) I make a few to ensure a survivor. We learn ways to handle them, perhaps holding with tweezers or forceps. This is an advantage of home printing, we can afford a loss rate whereas jobber printed one's, we usually can't afford this luxury.

Traditional works, especially machine work, will always be admired!

Tom

[DrPR]

A while back someone suggested using TransTint dye to dye anchor chains black. I got a bottle of the black dye and ran a quick experiment.

$24 for 2 ounces, but you dilute it 1 part dye to 32 parts water or alcohol.

I used some of the chain that I printed that had fused links. The original is on top and the dyed chain is below. For reference a chip of black plastic is between the two chains. In this run the chain was immersed in the dye for about two hours. As you can see the chain is darker, but far from black. This really isn't black like the chains were painted.

I will let another piece of the chain sit in the dye overnight and see if it gets any darker. If it is just adsorbed as a surface coating it probably won't get any darker, but if the dye somehow is absorbed into the cured resin it might become black.

To be continued ...

Phil

[Fliger747]

Some advantage to using black resin, though mine is translucent in thin sections. I wonder how much the type of resin and curing process might matter? Missouri has some of the FDD (no longer available) chain, which I think I printed more or less in place? Alaska will have it's chain replaced with some of my homemade product so I will be paying attention to your experiments.

As a historical time marker, "Ethyl" (remember her?) was $4.69, up a buck from last week.

Cheers: Tom

[DrPR]

The Mk 37 director saga continues ...

Some of the more experienced members on this forum suggested printing the parts tilted at an angle to reduce the forces involved when the horizontal steps and rails are separated from the plastic film at the bottom of the resin vat.

The cross sections would be smallest if the parts were tilted 45 degrees, and I first looked at that option. However it would be necessary to add a lot of supports on the visible surfaces in order to achieve enough support contact area to withstand the separation forces, and that would leave the exterior visible surfaces pockmarked with attachment points. And since there are a lot of features on the exterior surfaces it would be difficult to sand around them to remove the attachment artifacts without breaking off something.

An angle of 15 degrees was suggested and I tried that. This allowed most of the attachments to be connected to the hidden surfaces of the director shell and barbette. As before, I placed as many supports as possible on these surfaces:

I will describe the pros and cons of printing the barbette first.

This is the Chitubox image of the supports on the part tilted 15 degrees on the Y axis. All supports were added manually.

Keep in mind that the part was printed with the top surface in this photo connected to the print plate by the supports. The print was much more successful than my attempts to print the part vertically. Everything printed, even the very thin steps, rails and pipes. But as you can see, not everything was perfect. The two circular railings around the top of the barbette came out wavy. Even after 35 years of service the rails on the ship didn't bend out of shape this much! This photo is of the cured piece. I did notice that they straightened a bit after printing and before curing. So even at a 15 degree angle the thin rails stretched a bit as they separated from the plastic film. And the cross section between supports was about circular, and not flattened as in the vertically printed parts.

The biggest disappointment was the 1MC speaker cone. I have printed this many times now and this is the only one that failed. When the part was tilted the top surface curved upward and these had never failed to print. Maybe it was just a random failure. But after it is painted bright brass it will be a very noticeable part.

The water washdown pipe printed, but it is hanging at an odd angle (the horizontal part is supposed to be horizontal). It had two supports along this area, one near each "elbow". It looks like the attachment at the left end failed, allowing the pipe to bend "down" with each successive print slice. The bottom end also didn't print exactly right, and this needs some work.

And I forgot to add the six dogs that secured the armored water tight door. Operator problem!

The rear side printed OK - even the fragile washdown pipe, although it is a bit crooked. I suppose I can use a heat gun/hair dryer to soften it and try to straighten it. Has anyone tried this with printed parts?

Here is the top surface. You can see the many pockmarks from the attachment points, and see why I wanted them to be in a hidden location. It was very easy to sand this surface flat - except it is tricky handling the part because of all the thin steps and piping.

This is a closeup of the surface of the cylindrical part. It was printed with 0.03 mm steps, and the anti-aliasing features were turned on in the Chitubox "Advanced" settings (Anti-aliasing Level 2, Image Blur Pixel 3). This seems to fill in the "jaggies" resulting from the 0.05mm pixel size. But it also filled in the sprinkler heads on the washdown pipes. Without "Image Blur" enabled these very tiny parts have printed open.

All 20 steps printed OK.

Notice the faint angled lines across the part I have no idea what caused these, but they are at certain slices. Also notice that the right end of the steps at the curve seems to be a bit thinner than the left end. The left end was the lower end during printing.

Now look at a part that was printed vertically. Notice the prominent vertical "jaggies" grooves from the 0.05 mm pixels. I do not think I was using "Image Blur? because the washdown sprinkler head printed open. One of the steps broke off during printing/washing and another has one end broken loos. Also notice the prominent bulge at the bottom where the 3 mm thick bottom layer meets the 3 mm thick vertical walls. This bulge was absent on the angled part shown above.

All in all I would say that the angled orientation produced a superior part. Thanks to everyone who suggested it.

Now on to the director shell ...

Phil

[DrPR]

The director shell had many problems in the vertically printed parts. For this test I rotated the part 15 degrees around the X axis and 15 degrees around the Y axis. This ensured that none of the railings would be horizontal during printing.

All of the supports shown in these Chitubox images were added by hand. Notice the radar cabinet box hanging on the rear of the shell in the second picture. It has three flat vertical bars on the bottom, and I placed a support at the "lowest" end (nearest to the print plate) that would print first. I thought that if these points were supported the rest would print correctly. Wrong! All three failed to print completely. More supports needed!

In this photo of the rear of the part you can (barely) see how the support brackets didn't print correctly. However, other than a slight wave in the lower rear foot rail, all else printed nearly perfectly. I am especially pleased that the nuts and bolts came out quite visible. I spent days poring over blueprints and photos to get the locations of these fasteners correct! Of course they are far too small for most people to see without a magnifier, but I have the satisfaction of knowing they are there, even those on the bottom surfaces.

The front of the shell printed very nicely - as good as I can expect. There is a bit of aliasing, or jaggies, on the optical rangefinder arms, Oh well, nothing is perfect!

This last photo shows the part printed at an angle on the left and a part printed vertically on the right. Notice that the angled part didn't have the prominent "shrinkage" line just below the top surface, and the hand rails do not bulge in the center on the angled part. But the three vertical bars below the radar cabinet did print correctly on the vertically printed part!

There is another subtle difference. Notice on the vertically printed part on the right that you can see reflections in the very smooth top surface. You don't see these reflections in the part printed at an angle. In fact, all of the surfaces on the angled part have a smooth "matt" finish with no jaggies or vertical grooves. It is ready for painting.

I did have one problem with the shell. The "key" part on the bottom that fits into the "lock" on the top of the barbette came out a bit oversized. In the CAD drawings I left a 0.125 mm gap between the key and the hole it fits into. But I had to file down the sides of the key a bit to get it to slide into the lock. So next time I will make the key a bit smaller, or the lock hole a bit larger.

And those who suggested I try the Phrozen Fast Black resin will be interested to know that it is due to arrive here in a couple of days and I will repeat these prints for comparison. But first I need to read up on this new resin to get the right exposure settings for the Photon Mono printer.

Phil

[BB62vet]

Phil,

I was the one who suggested you try the TransTint Dye - as I stated in my original posting, I mixed approx. 1/8" of dye to the same amount of water in a 1 oz jigger and had excellent results using the gray resin chain. On the last go-round, I simply put a small amount of dye in a plastic cup UNDILUTED and got the same results - sorry, I don't have pix to illustrate my results. Personally, I think that just enough dye to cover the bottom of one of those little "pill" cups with the chain curled in the bottom should do it. But hey, I'm no expert, only reporting what I've achieved, etc. Besides, at the price of the dye, why use more than necessary and why NOT use it full strength? I do realize your chains will be larger and require more dye - but you're only building one model - 2 chains, right?

As for your Mk. 37 FC director - I think it came out rather nice, as well the barbette. I'm sure you'll work out the minor problems with exterior details, etc.

Hope this helps,

Hank

[DrPR]

Hank,

Thanks. I guess that's what I deserve for following the instructions on the bottle! It probably works fine for porous material like wood or cloth.

I'll try it full strength and see what happens. A milliliter or two in the bottom of a shot glass should be enough.

Do you recall how long you left the chain in the dye?

I let it dry for a few hours on a paper towel, and the dye doesn't rub off.

Phil

[BB62vet]

Hank,

Thanks. I guess that's what I deserve for following the instructions on the bottle! It probably works fine for porous material like wood or cloth.
I'll try it full strength and see what happens. A milliliter or two in the bottom of a shot glass should be enough.
Do you recall how long you left the chain in the dye?
I let it dry for a few hours on a paper towel, and the dye doesn't rub off.
Phil

Phil,
I left it in the dye for maybe up to an hour, I guess....I should have made a note regarding the time, but didn't - sorry, it would have helped to have that info. Drying time on a paper towel - I left overnight and that seemed to work ok.

Hank

[Fliger747]

Phil:

You had mentioned somewhere about some diagonal lines across the printed item. Are these parallel to the platen? Several things can contribute to this: (1) small supports are good for helping with sagging and other basal defects, but a few larger one's are needed to prevent the printing object from having a shift. (2) shifting in a layer can result from vibration, the table should be solid and isolated (3) the jack screw should be cleaned and lubricated so as to prevent a jerking from windup which can occur if the shaft gets dirty. If this get bad you will hear a "wonk" and see the printer jump as it breaks free.

For the foot and hand rail rings on the barbette for the MK 37 I printed these separately for Alaska and Missouri (1:192). sometimes it works better to have choice of the best of several rather than endanger a large item at the hazard of some detail not printing. The Kingposts on the APA have several such rings, the masts being brass, had to be printed separately. If necessary they can be made as split rings for installation. Loud Speakers also print much better as separate items as they tend to flatten on top or bottom otherwise.

Good luck! Tom

[DrPR]

I followed Hank's instructions and used the TransTint black dye at full strength. I allowed the anchor chain to sit in the dye for one hour and then dried it in open air over night.

I also used the dye diluted 1:32 according to the label instructions and let a piece of anchor chain sit in it overnight. Then it was dried in air.

The top piece in the photo is the grey resin print with no dye.

The second from the top is a piece of fused chain that was dyed for one hour in the 1:32 diouted dye.

The third down was dyed overnight in the 1:32 diluted dye.

The bottom is a section of freely articulating chain that was immersed for one hour in the undiluted dye and then dried overnight.

The middle two aren't much, if any, different. So the dye is not absorbing into the resin, and time is not a factor. Some of the dye pigment is adsorbing onto the surface, and my guess is that how much of it sticks to the resin is purely a matter of pigment concentration in the dye liquid.

The bottom chain looks pretty good. Some of the links were stuck together slightly, but a gentle twist of the chain freed them. I couldn't get the dye pigment to rub off the chain with a dry paper towel, but some did rub off on my fingers.

Then I rubbed the chain with a wet (water) paper towel for a few seconds and a LOT of pigment came off onto the towel. However, after I dried the chain it still looked very good. Here is a close-up of the washed links.

After the brief washing I no longer got pigment on my fingers when I handled the chain. However, I wouldn't push your luck and wash the dyed chain with alcohol, or leave it standing in water.

Thanks again to Hank for this tip. I think it is a good way to blacken anchor chain without gluing the links together with paint.

Phil

[BB62vet]

Phil,

Thanks! Glad that this method paid off for you. I would say at this point that immersing perhaps in the full strength dye for let's say...1-3 hours and then drying for a day and perhaps a gentle water wash might be the recipe for a black 3D printed chain. Yours looks "crusty & worn" which means it should work well in use!

I think one thing that might effect this method of "painting" 3D resin parts is the type of resin being dyed. When I first came up with this I was trying to blacken the old Shapeways white plastic type resin chain links and perhaps they took the dye better than this newer SLA resin does. In any event, this is still experimental stages in finishing these parts, so nothing cut & dried at this point.

Hank

[Fliger747]

Phil:

With the Rapid Black resin some items such as the foot rail on the MK 37 will come out looking like cooked spaghetti when removed from the platen, dry these fully and they will straighten out nicely. I have printed many a separate vertical ladder, even in 1:192, that came out quite nicely after a thorough drying before curing. You may find that the shrinkage with drying may also reduce the rod diameter from the designed diameter, so you might be able to design slightly oversize. As to cylinders such as your barbettes, they will be plenty strong without internal reinforcement. To avoid a pinching of the external wall from shrinkage of the partitions I made the partitions and decks on my gun tubs separate items, lodging against a designed ledge. One can print these but I designed and cut them on my stencil machine.

Once printed, your items will look very convincing!

Regards: Tom