3D Printed Metal Engine Parts

[ddmckee54]

The difference between resin printers and filament printers was summed up quite well in another thread, I don't remember where right now. If you need detail use a resin printer. If you need durability for a hands on prototype use a filament printer, resin prints tend to be more brittle.

I'm still on the fence about getting a resin printer, they're typically messy and smelly. But both of those objections are curable, or at least getting more tolerable.

 

[GreenTwin]

Wow what an inspiational post. Probably got through 56 of the 57 pages

At the trade guilde we are playing with resin printers. Early efforts at printing wax models have been successful. After seeing the magnificent examples in the 57page thread my mate modelled up a globe valve with minature writing on the side that we hope will be a part of his ME beam engine build (it certainly changing the modelling landscape)

Obsessed with casting fine detail I recently did a jewerly casting workshop. I was the odd one out - 5 young mums and me the obvious old man

Example of process…


Cheers jeff

The video shows sand molds made using an oil-based sand such as Petrobond (tm).
You can copy fingerprints with Petrobond, and some use it with iron, but I have found that it iron generally causes mold failures due to the high tempertures involved.

The 57 page locomotive work is one of the more impressive hobby casting things I have seen.
I am still reading the 57 pages, and should be done reading in the next 20 years or so.

.

 

[GreenTwin]

keep us posted of how the valve turns out.

This may be of interest if he wants some pipe elbows and flanges to go with it

https://www.modelenginemaker.com/index.php/topic,11630.0.html

The resin printers certainly seem to be changing the plastic model world, now you don't buy an injection moulded kit or aftermarket PU resin parts. You just buy the file and print out the it yourself be that a figure (historic or fantasy), body shell, turret, etc and you can make them in whatever scale you want.

I can see some of the photos in this link, but not the last post photos.

After reading the link, it sort of reinforces what I am thinking about trying to make patterns using resin prints that are burned out.
Years of struggling is not something I want to do again.
It took me six years (on and off) to figure out resin-bound sand and gray iron, and I don't want to go through another learning curve like that again.

3 days of burnout.......no thanks, not for me.

Surface finish still seems to be a problem with both lost PLA and lost resin patterns.
I still think using ceramic mold wash in some way to coat the mold interior would help, but with tiny castings that have intricate text and details, the mold wash wold wash out those details.

3D modeling and 3D printing is radically changing the hobby, as is the CNC thing.

What about a micro-CNC machine ?

.

 

[jirvin_4505]

The video shows sand molds made using an oil-based sand such as Petrobond (tm).
You can copy fingerprints with Petrobond, and some use it with iron, but I have found that it iron generally causes mold failures due to the high tempertures involved.

The 57 page locomotive work is one of the more impressive hobby casting things I have seen.
I am still reading the 57 pages, and should be done reading in the next 20 years or so.

.

Yes the delft clay is oil bound similar to petrobond. The jewellers I was talking to at the workshop say the delft is much finer and gives a finer finish when picking up detail

Comparing the feel of the delft to the petrobond I have suggests it is more like silt. I have seen reference to 200mesh grade.

I have no access to cast iron so am unaware of how it would perform

Craig dabbler on YouTube has developed sandcasting techniques using pre warmed sand Molds and vacuum assisted Molding that look like they would be interesting crossover techniques

His 6prong ring cast series shows some interesting techniques using cores etc.


Results of cast about 2 1/2 minutes in

Sorry for the thread derailment

Cheers Jeff

 

[Jasonb]

Pat that 3 days is the whole process from printing waxes, making the tree, investment, burn out and cast

"Each pour, between print, tree build, investment setting and burn out is about three days, obviously not full time but there is lots to attend to so you can't if there is lots on over the weekend."

You would not have to sit there watching the printer, waiting for the investment to set or watching the burn out.

CNC now that is a good idea

 

[GreenTwin]

He mentions a 3 day burnout schedule.

.

 

[GreenTwin]

That makes me wonder if cast iron could be melted on a jewelry scale.

Cast iron needs a slight amount of ferrosilicon added after reaching pour temperature, and also needs to be de-slagged prior to pouring.

The oxygen in the oxy-actet torch could cause excessive slag formation.

.

 

[Jasonb]

Read it again he says he uses a 24hr cycle for the burnout and the thread above I quoted his words that the whole thing is 3 days.

"What worked is X-One castable resin, 873 bronze, 510C flask temp that was taken to 740C (about 24 hour cycle, first to 150, then 350 then 740 with very slow ramps)"


"Each pour, between print, tree build, investment setting and burn out is about three days," obviously not full time but there is lots to attend to so you can't if there is lots on over the weekend.

 

[GreenTwin]

And as if the 3 day period were not bad enough, then if the mold cracks, or the casting does not turn out well, then you get to start all over again.

And so that is why I have not gotten into the investment process; it is very tedious, time consuming, and requires special equipment and materials that I don't have.

Hats off to those who are able and willing to do it.
Some impressive work for sure.

.

 

[ddmckee54]

Yup, tedious, time-consuming, need special equipment and special materials... But OH MAN, the detailed complex parts you can get.

 

[Jasonb]

Yes , you need to consider that investment will do items that would need several core boxes making or printing and then all the cores making. Then you have to try and work out where to split the pattern and so on. All adds up time wise so not always as simple as bound sand is quicker.

The resin printed waxes will need far less work to the surface as the layer heights are a lot less than filament. So also add in the time to fill, sand, paint etc all with drying times between.

 

[ddmckee54]

You really shouldn't need to fill a resin print, other than to repair the areas where support material was attached. The layers of a resin printer should be about an order of magnitude thinner than a filament printer, down to the point where you need magnification to clearly see the layer lines. Where the layer lines from a filament printer are clearly visible to the Mk. 1 eyeball. I typically print at 0.2mm, or 200 microns. For me this is the best compromise between print quality and print speed. I have printed at 0.1mm, and have heard of people printing at less. 100-200 microns is are on the UPPER end of a resin printers ability, not the lower end of around 10 microns.

You can vapor smooth most of the printable waxes with isopropyl alcohol to a glass smooth finish. From both a resin printer, and a filament printer. The styrene resins will need acetone or MEK for vapor smoothing, where PLA takes something a lot nastier.

 

[Jasonb]

I did not word it well, what I meant was resin printed waxes are just about ready to go. But Pat wants to stick with PLA which will need work to the surface before he can get a decent bound sand mould from it. Eg his gears are likely to need some sanding/filling to the teeth and as yet he has not come up with a good method for more intricate surfaces. This will add to the overall time from print to casting.