Very impressive work, my friend! I’m retired now, but I owned and ran a small foundry and machine shop for over 30 years. I’ve only done sand casting with petrobond sand, and a few failed experiments with lost foam. At the height of my business I went through about 2000 lbs of 319 alloy a year. Maybe that’s why my back is shot! I sold all of my foundry equipment 7 years ago, but your videos have inspired me to try my hand at lost foam. I’m going to get one of those small eBay furnaces to get started. Thank you for sharing your methods!👍 Tony BTW, what alloy are you casting? I chose to use 319, because it age hardens.
I use A356 ingot. It's a good all-around casting alloy as far as fluidity, mechanical properties, weldability, and responds well to grain refiners and enhancers. Only downside compared to 319 is the need for heat treating to obtain optimal mech properties. 319 would be my close second choice, but it's only practical for me to buy one alloy in my ingot purchases. Best, Kelly
You need to be able to melt and handle/pour molten metal and make the foam pattern. Outside of that you need a bucket of dry sand. You can buy a cheap foundry furnace for a couple hundred dollars. You can carve a foam pattern with a razor knife, hot wire, or machine it with wood working tools. I have dedicated videos on pattern materials and pattern construction methods if you search my channel. Best, Kelly
For those really repetitive patterns have you looked into lost wax casting? I image you could 3d print a master pattern and cast silicon around it to create a negative mold which you can then pull a wax positive from.
Making the molds and patterns for lost wax for this part would not be a trivial task. Then you would still have all the downstream costs and labor mentioned in the post immediately above. You'd just need to have used both processes on this scale to appreciate the differences, but I'd compare timecards with anyone that wanted to have a crack at trying lost wax on this one 😏. Best, Kelly
I really enjoy your videos. What CNC program do you use to design the water necks etc? I’ve been developing a manifold for an LS engine using BMW ITBs and 3D printed them but really like the idea of machining the foam for a lost foam casting.
I use CamBam for about three years now and think it's tremendous value for money. Does everything I need it to including 3D machining. They have a forum with very helpful moderators. It's one time purchase and you own the license with no further fees. It has a rudimentary CAD capability and if I was only going to do 2.5D modeling, I probably would not have needed an additional CAD program. It pains me to say, but the CB developer is somewhat absent and the website to purchase and forum is frequently down as it is now, which is a bit concerning. I have heard favorable things about MeshCam. There are of course many other offerings at higher price points. Best, Kelly
@ thank you. I’m still trying to learn how to use these programs (I’m an old fireman trying to learn new things 😂). I appreciate your help and I look forward to seeing more videos. Thank you.
Lost foam castings vent through the refractory coating along the entire pattern surface as the mold fills. In general, I find patterns with high surface area/volume ratios cast best because they vent the best. Best, Kelly
Hey Kelly, with conventional core casting, you'ld get some external flash at the joints, but with the foam method, there's potential for internal "gaps" at the joins. I would have thought waxing all joins, would be advisable, but then, you wouldn't be able to do the internal ones. Have you had any problems with that, or is the aluminium not that "sensitive" to the joins?/You are careful to have full glue coverage.. (I basically have zero experience casting, for reference purposes..)
I do strive to have gapless joints and for the most part succeed, but I do inspect and fill any gaps with wax. It's not so much that they are sensitive to the molten aluminum, but moreover the refractory coating, which will penetrate the gap and result in a fissure in the casting. I don't coat the interior and the sand grain size is too large to penetrate internal gaps. I have used four different glues over the 8 years for joining the tube halves. Hot melt: too low of working time resulting in thicker and inconsistent joint thickness. Shellac: Very thin and more sandable joint but doesn't burn out cleanly. Polyvinyl Acetate (PVA white) glue, better if used sparing, but slow drying and more dense than preferred. Rubber cement: Seems to be best at achieving thinnest joint with clean burn out. I still often use PVA and hot glue depending upon application. You never fully get rid of visibility of joint unless you do a build of wax and even then, you see the wax build on the casting, but you can polish that out of the casting if in a critical area. Here it's just cosmetic and just not warranted on these IMO. Best, Kelly
I'm sure you've been asked this 1000 times before but any reason why you aren't 3D printing the cores and lost PLA casting them? I like the vacuum jig. Smart.
The short answer is lost foam is overall less laborious and much less expensive. If you have used both processes this can be easily appreciated. Sure, you can 3D print a pattern, and that is much less labor than assembling machined foam pieces, but the down stream processes and costs are strongly in favor of lost foam. Though this pattern is small for me, most would consider it to be large for lost wax (statuary art casters aside). I cast them in a 5 gallon bucket and an investment flask could not be much smaller (say 90-100lbs of investment). Besides the cost of printable media and investment, you would need to mix, degas, invest, and then burn out ~4-5 gallons of investment. A mold that size would need to be burned out for a long time, likely days, would require a large kiln, lifting equipment to manage the preheated hot mold before pouring, and still much more labor to demold and dispose of 90lbs of investment. The pattern and mold material cost for lost foam by contrast is literally pennies, requires no burn out, the mold can be vibratory packed and ready to pour in 5-10 minutes, and demolded by literally just dumping it on the ground……and the dry sand mold media can be reused. My larger intake manifold castings would have 400-500lbs of investment so amplify the above by 5x. It’s a different ball game than casting small parts in 70mm-100mm flasks. This part is by far my most laborious. The larger intake manifolds are all cnc run time, much like print time. These are much more “artful”. Best, Kelly
@kellycoffield533 I hear you. That would indeed be a lot of investment to heat up to off gas etc like you say. What about ceramic dip casting 3d prints? Viable? 1 good thing with 3d printing is you can define the infil, which is how solidly you want the print to be. If I set it to be 0% it will just print the outside shell of the object. Massively reducing the amount of filament used, burn out time etc. One day I'll make a forge and get into it and try it myself. I've got too many other projects on the go already.
@@ratgreen By dipping, I assume you mean ceramic shell, and it's pretty much the same story there on cost and labor, and downstream processes. Zero infill isn't very helpful if the pattern can't be hung to support 30-40lbs or ceramic mold weight. Best, Kelly
Yes, retail extruded polystyrene foam insulation board 1.3 - 1.5lb/ft3 density. I have an entire video on pattern materials if you search my channel. Best, Kelly
This is ART. Please do not die before you make hundreds of videos.
I'm trying to get 365 days older every year.......LoL!
Every time you post a video I get excited about learning and seeing more cool parts made!
Me TOO 😁. I get so inspired after watching every video Kelly releases.
I would never have thought you could bend eps. Always nice when you post a video Kelly. Thanks👍
This is absolutely amazing work, thanks for sharing.
grat job, great patience...
Nice job i’ve learned a lot.
Very impressive work, my friend!
I’m retired now, but I owned and ran a small foundry and machine shop for over 30 years. I’ve only done sand casting with petrobond sand, and a few failed experiments with lost foam.
At the height of my business I went through about 2000 lbs of 319 alloy a year. Maybe that’s why my back is shot! I sold all of my foundry equipment 7 years ago, but your videos have inspired me to try my hand at lost foam. I’m going to get one of those small eBay furnaces to get started.
Thank you for sharing your methods!👍
Tony
BTW, what alloy are you casting? I chose to use 319, because it age hardens.
I use A356 ingot. It's a good all-around casting alloy as far as fluidity, mechanical properties, weldability, and responds well to grain refiners and enhancers. Only downside compared to 319 is the need for heat treating to obtain optimal mech properties. 319 would be my close second choice, but it's only practical for me to buy one alloy in my ingot purchases. Best, Kelly
What does it take to get started in doing this kind of casting? Iam very impressed great video
Patience, elbow grease and finesse.
You need to be able to melt and handle/pour molten metal and make the foam pattern. Outside of that you need a bucket of dry sand. You can buy a cheap foundry furnace for a couple hundred dollars. You can carve a foam pattern with a razor knife, hot wire, or machine it with wood working tools. I have dedicated videos on pattern materials and pattern construction methods if you search my channel. Best, Kelly
You are just too damn clever!
For those really repetitive patterns have you looked into lost wax casting? I image you could 3d print a master pattern and cast silicon around it to create a negative mold which you can then pull a wax positive from.
Making the molds and patterns for lost wax for this part would not be a trivial task. Then you would still have all the downstream costs and labor mentioned in the post immediately above. You'd just need to have used both processes on this scale to appreciate the differences, but I'd compare timecards with anyone that wanted to have a crack at trying lost wax on this one 😏. Best, Kelly
I really enjoy your videos. What CNC program do you use to design the water necks etc? I’ve been developing a manifold for an LS engine using BMW ITBs and 3D printed them but really like the idea of machining the foam for a lost foam casting.
I use CamBam for about three years now and think it's tremendous value for money. Does everything I need it to including 3D machining. They have a forum with very helpful moderators. It's one time purchase and you own the license with no further fees. It has a rudimentary CAD capability and if I was only going to do 2.5D modeling, I probably would not have needed an additional CAD program. It pains me to say, but the CB developer is somewhat absent and the website to purchase and forum is frequently down as it is now, which is a bit concerning. I have heard favorable things about MeshCam. There are of course many other offerings at higher price points. Best, Kelly
@ thank you. I’m still trying to learn how to use these programs (I’m an old fireman trying to learn new things 😂). I appreciate your help and I look forward to seeing more videos. Thank you.
Is there a relationship between the the amount of foam used per item to be cast, and the venting needed?
Lost foam castings vent through the refractory coating along the entire pattern surface as the mold fills. In general, I find patterns with high surface area/volume ratios cast best because they vent the best. Best, Kelly
Hey Kelly, with conventional core casting, you'ld get some external flash at the joints, but with the foam method, there's potential for internal "gaps" at the joins. I would have thought waxing all joins, would be advisable, but then, you wouldn't be able to do the internal ones.
Have you had any problems with that, or is the aluminium not that "sensitive" to the joins?/You are careful to have full glue coverage..
(I basically have zero experience casting, for reference purposes..)
I do strive to have gapless joints and for the most part succeed, but I do inspect and fill any gaps with wax. It's not so much that they are sensitive to the molten aluminum, but moreover the refractory coating, which will penetrate the gap and result in a fissure in the casting. I don't coat the interior and the sand grain size is too large to penetrate internal gaps. I have used four different glues over the 8 years for joining the tube halves. Hot melt: too low of working time resulting in thicker and inconsistent joint thickness. Shellac: Very thin and more sandable joint but doesn't burn out cleanly. Polyvinyl Acetate (PVA white) glue, better if used sparing, but slow drying and more dense than preferred. Rubber cement: Seems to be best at achieving thinnest joint with clean burn out. I still often use PVA and hot glue depending upon application. You never fully get rid of visibility of joint unless you do a build of wax and even then, you see the wax build on the casting, but you can polish that out of the casting if in a critical area. Here it's just cosmetic and just not warranted on these IMO. Best, Kelly
@@kellycoffield533 Cheers for the detailed response Kelly 👍your knowledge in lost foam casting is invaluable!
I'm sure you've been asked this 1000 times before but any reason why you aren't 3D printing the cores and lost PLA casting them?
I like the vacuum jig. Smart.
The short answer is lost foam is overall less laborious and much less expensive.
If you have used both processes this can be easily appreciated. Sure, you can 3D print a pattern, and that is much less labor than assembling machined foam pieces, but the down stream processes and costs are strongly in favor of lost foam.
Though this pattern is small for me, most would consider it to be large for lost wax (statuary art casters aside). I cast them in a 5 gallon bucket and an investment flask could not be much smaller (say 90-100lbs of investment).
Besides the cost of printable media and investment, you would need to mix, degas, invest, and then burn out ~4-5 gallons of investment. A mold that size would need to be burned out for a long time, likely days, would require a large kiln, lifting equipment to manage the preheated hot mold before pouring, and still much more labor to demold and dispose of 90lbs of investment.
The pattern and mold material cost for lost foam by contrast is literally pennies, requires no burn out, the mold can be vibratory packed and ready to pour in 5-10 minutes, and demolded by literally just dumping it on the ground……and the dry sand mold media can be reused.
My larger intake manifold castings would have 400-500lbs of investment so amplify the above by 5x. It’s a different ball game than casting small parts in 70mm-100mm flasks. This part is by far my most laborious. The larger intake manifolds are all cnc run time, much like print time. These are much more “artful”.
Best,
Kelly
@kellycoffield533 I hear you. That would indeed be a lot of investment to heat up to off gas etc like you say.
What about ceramic dip casting 3d prints? Viable?
1 good thing with 3d printing is you can define the infil, which is how solidly you want the print to be. If I set it to be 0% it will just print the outside shell of the object. Massively reducing the amount of filament used, burn out time etc.
One day I'll make a forge and get into it and try it myself. I've got too many other projects on the go already.
@@ratgreen By dipping, I assume you mean ceramic shell, and it's pretty much the same story there on cost and labor, and downstream processes. Zero infill isn't very helpful if the pattern can't be hung to support 30-40lbs or ceramic mold weight. Best, Kelly
Awesome! Hello! Please tell me what density of foam do you use? Is this a regular xps?
Yes, retail extruded polystyrene foam insulation board 1.3 - 1.5lb/ft3 density. I have an entire video on pattern materials if you search my channel. Best, Kelly
Top