Making the cylinder head, design, patterns and casting.

[Alyn Foundry]

As a footnote to the pictures above I forgot to mention the two “ slip blanks “ placed either side of the patterns. This helps to prevent slip between the two halves of the finished flask.

I wish I had taken the time to video the guys at the foundry, they were using Greensand on a special moulding machine. They started with the plate, mounted on the machine then placed a wooden hinged box around the first half. Virgin sand would be sieved over the plate to make an 1” thickness layer over the patterns. Then a layer of backing sand would be added along with a rectangular Steel band that nearly touched the sides of the wooden box. More sand would be added before a second band was placed near the top. The whole box would then be compressed by a pneumatic plate before a vibrator helped loosen the plate from the sand. The moulder would then place the flask down onto the foundry floor, remove the wooden box and place it around the other side of the plate. The same procedure would be used for the Cope but in addition the sprue would be cut. Any cores would be placed in the Drag before closing the two halves together. The finished mould was basically two large, bare sand rectangles held together by the Steel bands. Amazing to think about the way they actually held themselves together but it worked non the less.

   Graham.

[A7er]

The new head design, same size as the old head, different shape. The new head has a slight curve to the outer face to aid removing the pattern from the mould. And I think it looks better! Old head first.




The new design.




And the water core moulds.





The head pattern. The square protrusions are for the water core support, the round ones for the inlet and exhaust cores.



The pattern is split horizontaly.



Lee

[A7er]

I think I have the design of the new head sorted, so I am 3D printing the patterns and moulds. Most of the moulds that don't require a perfect surface finish can be printed in pla on my Ender3. Some of the patterns I will use my resin printer, especially the head pattern. I have "hollowed out" the prints in the cad programme to create walls 2mm thick, rather than doing it in the slicer programme. Locating sockets have been added for 6mm dowels. One of the sockets is round in each half, the other is slightly elongated in one of the patterns and round in the other half. This I find takes care of slight misalignment if printing in pla. I am undecided at the moment whether to use a matchplate for the head pattern.







The outer shell of the water core mould is printed in two halves.



The core for creating the combustion chamber also locates the inlet and exhaust cores.



This is the mould for making the core, two required, mirror image.



Lee

[AlexS]

Interesting project!

Are you also printing a mold riser? Do you have pictures or maybe a video of casting? I am wonder what kind of heater you use!

[A7er]

No pictures or video of a casting, yet. I would like to video a pour, it could be an interesting project in itself. I will have to find out how to embed, or link to a video I make.

My metal melting foundry, some call it a furnace, is made from several high temp lightweight fire bricks with home wound electric elements of about 1800 watts. I made the whole thing myself. It has an electronic temperature control that when set to a certain temp will keep the foundry to within a few degrees of that temp. It will reach 1000 degrees C, but I don't need to go higher than 750C. I could do with making a bigger foundry to hold a bigger crucible running at about 2500 watts. Another future project!
Lee

[A7er]

I used the mould for the combustion chamber core to make the two halves of the core using just under 1 gram sodium silicate to 17 gram fine sand for each half. Approx a 5% mix. This was then firmly compacted into the mould and a fine wire used to poke holes in the top surface so that the gas could get into the sand. The mould is small enough to fit under an empty yoghurt pot with a hole in for the gas tube and another hole to let the air out. 10 seconds of gas, turn it off and block the air hole with my finger, sit there for two mins and test the hardness of the sand before removing the core from the mould and doing it again to get both halves. Both pieces are quite firm and not crumbly, plus, several hours quicker than using expoxy resin for the binder. I will see if there is an epoxy resin that sets in about 30 mins with a use time of at least 10 mins.





I have the file for the two halves of the water core components ready for a 5 hour print in the Ender3, and at the same time I will use the resin printer to print the two halves of the head pattern, one half at a time because the build plate is small. Three and a half hours each half. It will be a long day!

This what the head patterns look like with supports added in the 3D slicer.



Lee

Edit. Whoops! I forgot the size, the sand core is 35mm square by 15mm deep, the combustion chamber is 30mm dia by 5mm deep.

[BillTodd]

Are the  under sides of your resin mould well supported?

Better too much than to little ;⁠-⁠)

[A7er]

Perhaps a few more?
I sliced the parts in Prusa slicer and added supports at 200%, then added at least 40 more. As someone cleverer than me once said, "Better too much than too little". 

[A7er]

The patterns for the head and core have been 3D printed and I am in the process of making them nice. The resin prints look perfect, but they are in fact slightly warped. I will get over this by bonding them to a matchplate. They require very little sanding, unlike the pla prints from the Ender3. They need a fair bit of sanding but the new filament makes the task easier.









Lee

[Kim]

Those prints look great A7er!

So, I don't know much about SLA printing, but what's the reasoning for printing those parts at that jaunty angle?  I think the print bed hangs upside down and prints that way. So in your picture, the print bed has been flipped over after the print completed, right?).  But why didn't you just print the parts with one of the flat ends toward the print bed?  You could have had the big open end facing down to let the resin drain out.  I'm just curious.

Thanks,
Kim

[ddmckee54]

A7er:

Couple of question about the resin prints:
It doesn't look like the supports pulled away from the build-plate, did the part pull away from the supports?  Any theories/fixes for this?

What printer are you using?  I've got a couple of FDM printers and I've been thinking about pulling the trigger on a resin printer, just because of the amount of detail that you can get.  My biggest hurdle with a resin printer is the smell of the typical resin.  Have you tried the water-wash resins, and if so do they really have less odor like they claim?.

Don

[BillTodd]

Those prints look great A7er!

So, I don't know much about SLA printing, but what's the reasoning for printing those parts at that jaunty angle?  I think the print bed hangs upside down and prints that way. So in your picture, the print bed has been flipped over after the print completed, right?).  But why didn't you just print the parts with one of the flat ends toward the print bed?  You could have had the big open end facing down to let the resin drain out.  I'm just curious.

Thanks,
Kim

As each layer is printed, it has to be peeled away from the FEP film , this tension has to be minimised to prevent distortion or even pulling the job off of the build plate. So slanting the job can help reduce the area of each layer.

, I don't know much about SLA printing, but what's the reasoning for printing those parts at that jaunty angle?  I think the print bed hangs upside down and prints that way. So in your picture, the print bed has been flipped over after the print completed, right?).  But why didn't you just print the parts with one of the flat ends toward the print bed?  You could have had the big open end facing down to let the resin drain out.  I'm just curious.

Thanks,

Further to the above. Another problem is the so called 'suction cup'  if you have a periphery that encloses an area ,  fluid (air or resin) has to be able to get into that space , or a low pressure zone will form and, since the model is actually quite soft,  this can distortion, puncture or rip the whole job from the build plate.

[Kim]

Those prints look great A7er!

So, I don't know much about SLA printing, but what's the reasoning for printing those parts at that jaunty angle?  I think the print bed hangs upside down and prints that way. So in your picture, the print bed has been flipped over after the print completed, right?).  But why didn't you just print the parts with one of the flat ends toward the print bed?  You could have had the big open end facing down to let the resin drain out.  I'm just curious.

Thanks,
Kim

As each layer is printed, it has to be peeled away from the FEP film , this tension has to be minimised to prevent distortion or even pulling the job off of the build plate. So slanting the job can help reduce the area of each layer.

, I don't know much about SLA printing, but what's the reasoning for printing those parts at that jaunty angle?  I think the print bed hangs upside down and prints that way. So in your picture, the print bed has been flipped over after the print completed, right?).  But why didn't you just print the parts with one of the flat ends toward the print bed?  You could have had the big open end facing down to let the resin drain out.  I'm just curious.

Thanks,

Further to the above. Another problem is the so called 'suction cup'  if you have a periphery that encloses an area ,  fluid (air or resin) has to be able to get into that space , or a low pressure zone will form and, since the model is actually quite soft,  this can distortion, puncture or rip the whole job from the build plate.

Thank you for the detailed explanation. This makes a lot of sense!
Kim

[A7er]

Hi Kim,
BillTodd answered your questions better than I could. I will add though that printing at an angle should give a better surface finish to the faces at an angle because resin trying to print across two supports tends to sag. That's what I've read and discovered by experience. The ideal angle is supposed to be just under 45 deg. There is a formula for working it out, I don't use it myself. If I printed one of the head patterns flat, say 5mm off the build plate, it would print quite quickly because there would be less layers to process. 2 or 3 hundred-ish. If I printed the same part vertically, there would be several hundred layers to process, each layer is exposed to UV light for 3.5 seconds. The first 6 layers need to be "baked" onto the build plate, so those 6 layers are exposed for 40 seconds. All the exposure times and the speed at which the build plate is pulled out of the resin before being lowered again are adjustable. If I printed the same pattern flat on the build plate, no supports, the first 6 layers would have distortion because they have been exposed to UV for longer, and the light creeps into the resin setting it. On fdm printers, as well as resin printers, this is called elephants foot. There are settings in most slicers to take this into account, my slicer doesn't.

I am still getting distortion on my resin prints which is a shame as the photographs I post don't do the surface finish and detail justice. On one piece I did it printed a quarter mm square line, text half a mm high and wide that was so sharp.

I will try a trial print with one half of the pattern 5mm from the build plate with supports, drain holes etc and a layer time of 5 seconds, see if I can get less distortion.

Don,
The supports didn't pull away from the build plate or the print. The part of each support that connects with the part is very thin to make it easier to break away and leave less damage to the surface of the part. The settings concerning the supports can be altered, finer, thicker, smaller head into print etc.

I am using an Anycubic Mono 4K. It is hard work! Once the print has ended I have to wash it in water, then carefully remove the print from the build plate, wash the bits I missed the first time, dry it, put it into the wash and cure machine to sit under UV light on a turntable to make sure all the resin has set. It could be put outside in the sun shine to do that, and then after a certain time turn it over and do the other side. In the meantime I drain the unused resin back into the bottle using a filter to catch any loose bits, use the plastic spatula to remove the resin from the fep film in the vat, wash the vat and film with clean water, dry the film and vat then clean it again using IPA. I haven't used anything but water washable so far. The water based resin I use is almost odour free. The IPA smells a lot worse. I have my resin printer in a small room just off the kitchen, I can't smell it at all.

I have my Ender3 printing the same head patterns shown in the images to see what difference, if any, there is between resin prints and fdm prints when it comes to distortion. If the resin print had no distortion and just needed tidying up a bit, it would be the perfect option for producing 3D prints for casting.


Lee

[BillTodd]

Lee,

Is your print room temperatures warm enough?   

Another possibility is to reinforce the back of the model  with a grid or similar to help hold it in shape .  The distortion is likely caused by shrinkage