Lost wax casting--advice on setting-up sprues for a mold

[Alex Jemetz]

Fellow blade-smiths, I'm preparing to cast the brass cross-guard for my Ottoman/Cossack saber project.   I'm looking for advice about how to set-up the sprues on the mold as my previous effort came out OK for my first saber but I would like to improve on the result.  I'm looking to reduce the number of voids in the mold (which I realize could be due in part to an incomplete burn-out of wax and not a high enough temperature of the mold before pouring the brass).  Below are pics of the first saber guard pre-casting and results.  All input would be appreciated as I can't find much advice on the web... 

[Gert Burkhardt]

Ha, that looks a bit like the super porous bronze dagger that I cast recently. In that case the mold was definitely not fired long and hot enough.

Leads me to the question what are you using for the mold? Investment gypsum? If so, I had good results by firing to 1350 degrees F for 3-4 hours in a kiln (after slowly heating up over 5 hours to not let it spall), then casting at 800-900 degrees internal mold temperature. Basically just what the instructions said. 

The sprue and adjacent pieces must be the thickest part in the casting that solidifies last.

 

I used gypsum molds a few times for lost wax parts and refractory cement for reusable 2-part molds as well. In both cases I pulled vacuum on the material after mixing with water and believe it helped removing most air bubbles.

 

I tried brass only once and that came out porous as well. IMO bronze is easier to cast and less noxious fumes.

[Jerrod Miller]

Things to keep in mind:

1. Every bit of molten metal surface area touching air is bad.  Design your gating to be as smooth pouring/flowing as possible.  
2. Keep your wax as smooth as possible.  Besides areas that would cause turbulence, rough surfaces and sharp corners are also areas prone to catch and trap wax  residue during burnout and cause other shell defects.  Also, it is easier to smooth out wax than metal.  
3. Follow the instructions from the manufacturer of the investment material.  
4. When you pour, you want to get the metal into the mold as fast as possible.  

[Jeroen Zuiderwijk]

If good quality castings is your primary interest, then it's better to have the feed attached to the bottom. That does two things: reduce turbulence in the metal and mixing with air, and all the gasses and other contaminants lighter then the metal can rise to the top along with the flow of the metal, rather then having to go against the flow. This is how modern castings are generally done. I generally cast from the top like you do, but that's because I want to stay close to historical methods, and learn how to deal with that. 

[Alex Jemetz]

On 10/7/2025 at 4:00 PM, Gert Burkhardt said:

Ha, that looks a bit like the super porous bronze dagger that I cast recently. In that case the mold was definitely not fired long and hot enough.

Leads me to the question what are you using for the mold? Investment gypsum? If so, I had good results by firing to 1350 degrees F for 3-4 hours in a kiln (after slowly heating up over 5 hours to not let it spall), then casting at 800-900 degrees internal mold temperature. Basically just what the instructions said. 

The sprue and adjacent pieces must be the thickest part in the casting that solidifies last.

 

I used gypsum molds a few times for lost wax parts and refractory cement for reusable 2-part molds as well. In both cases I pulled vacuum on the material after mixing with water and believe it helped removing most air bubbles.

 

I tried brass only once and that came out porous as well. IMO bronze is easier to cast and less noxious fumes.

Thank you for replying Gert, I am using Hydroperm metal casting plaster for the mold.  I'm not sure I can get the entire mold to the temperature you mention (1,350) as I don't have access to a kiln. This time I am going to use ceramic bricks to build something close to a kiln and will fire the mold for 3-4 hours.  QUESTION, do I do the wax burn-out right before I pour the brass?
I will try to make the sprues more curved to avoid sharp angles, but should I make the sprues from larger diameter wax?  Does it hurt to have many sprues?
I really appreciate your input Gert.   

[Alex Jemetz]

On 10/7/2025 at 7:42 PM, Jerrod Miller said:

Things to keep in mind:

1. Every bit of molten metal surface area touching air is bad.  Design your gating to be as smooth pouring/flowing as possible.  
2. Keep your wax as smooth as possible.  Besides areas that would cause turbulence, rough surfaces and sharp corners are also areas prone to catch and trap wax  residue during burnout and cause other shell defects.  Also, it is easier to smooth out wax than metal.  
3. Follow the instructions from the manufacturer of the investment material.  
4. When you pour, you want to get the metal into the mold as fast as possible.  

Thanks for you response Jerrod.  I do have a couple questions on sprues, should I use a larger diameter sprue?  Can you have too many sprues?

[Alex Jemetz]

On 10/9/2025 at 3:44 AM, Jeroen Zuiderwijk said:

If good quality castings is your primary interest, then it's better to have the feed attached to the bottom. That does two things: reduce turbulence in the metal and mixing with air, and all the gasses and other contaminants lighter then the metal can rise to the top along with the flow of the metal, rather then having to go against the flow. This is how modern castings are generally done. I generally cast from the top like you do, but that's because I want to stay close to historical methods, and learn how to deal with that. 

Hi Jeroen, glad to have your advice.  I'm going to try using the feed from the bottom on my next pour.  For this piece I am inclined to do a top pour but factoring all the advice I have received from you, Gert and Jerrod namely:  make the sprues more curved instead of sharp angles, burn-out the wax for a longer period of time and pre-heat the mold to a much higher temp.  I'll keep everyone updated on how it goes.

[Gert Burkhardt]

9 minutes ago, Alex Jemetz said:

....QUESTION, do I do the wax burn-out right before I pour the brass?
I will try to make the sprues more curved to avoid sharp angles, but should I make the sprues from larger diameter wax?  Does it hurt to have many sprues?....

I assume burn-out before casting would be the normal method:

- melt out the wax around 300 F, preferably let the wax run out into a tray and remove before heating further. But no matter what, a certain quantity of the wax will soak into the investment and vaporize later, i.e. stink away.

- then keep heating slowly. The investment gypsum that I used recommended to hold at 1350 degrees for (IIRC) 3 hours to totally burn out any wax remnants and dehydrate the gypsum. Lower temperature may or may not be sufficient but for my experiments it contributed to porosity and voids. Then cool back down to 800-900 and cast.

- There should be one sprue to fill the mold (the other thin wax filaments are vents to let air escape). The sprue should be the thickest (or almost thickest) part in the casting. Otherwise, it will freeze first and not allow more metal into the casting to replace the shrink when cooling, leaving sink marks. Depending on the shape that may be visible.

[Alex Jemetz]

40 minutes ago, Gert Burkhardt said:

I assume burn-out before casting would be the normal method:

- melt out the wax around 300 F, preferably let the wax run out into a tray and remove before heating further. But no matter what, a certain quantity of the wax will soak into the investment and vaporize later, i.e. stink away.

- then keep heating slowly. The investment gypsum that I used recommended to hold at 1350 degrees for (IIRC) 3 hours to totally burn out any wax remnants and dehydrate the gypsum. Lower temperature may or may not be sufficient but for my experiments it contributed to porosity and voids. Then cool back down to 800-900 and cast.

- There should be one sprue to fill the mold (the other thin wax filaments are vents to let air escape). The sprue should be the thickest (or almost thickest) part in the casting. Otherwise, it will freeze first and not allow more metal into the casting to replace the shrink when cooling, leaving sink marks. Depending on the shape that may be visible.

Thank you Gert!

[Jerrod Miller]

5 hours ago, Gert Burkhardt said:

The sprue should be the thickest (or almost thickest) part in the casting. Otherwise, it will freeze first and not allow more metal into the casting to replace the shrink when cooling, leaving sink marks.

This is not always the case.  There are times where geometry dictates that you have a very small gate (where the sprue meets the part) so that it freezes off there without any feeding in either direction.  Feeders (aka risers) can be added virtually anywhere to provide extra feed metal during solidification.  It is just pretty common practice to use the sprue/runner/gate as a feeder, so that you limit the number of contacts on the part to clean up.  

 

There are a few important things to keep in mind when determining the feeding methods for a casting.  

• Not everything needs to be fed.  Sometimes a small amount of shrink, especially in the centerline or otherwise hidden area, is perfectly acceptable.  
• Visualize how the part is going to cool, and thus solidify.  Thin sections will cool fastest, and heavy sections slower.  Sections that aren't connected but have another heat source nearby (like the lugs on your part) will cool slower than if they were isolated. More info on this below.   
• There are 2 main parts to shrinkage:
  • Volumetric liquid shrinkage happens as the metal cools but does not solidify yet.  At this stage, there is no blockage of feed paths, so you just need excess metal in the system to be able to flow in to occupy the part cavity.  
  • Feeding shrinkage happens as the metal continues to cool and solidifies.  At this stage, it is possible for dendrites forming to cut off flow of new metal from the feeder (whether this is the sprue or some other source).  

To determine how/where to add feeders to a casting (with the intent of making it shrink-free), I like to use the "largest inscribed sphere" method for quick and dirty calculations.  Basically, start at the thinnest section of your part and work your way to a heavier section.  Imagine the volume of the largest sphere you can place in the mold cavity as you go along.  You keep getting bigger spheres as you move away from the thin section.  Once you are no longer able to get a bigger sphere in that section, make a note of that location.  Repeat the process for every thin point of the part.  If all your thin points lead to a single place where the largest inscribed sphere is, that is the one place you need to put a feeder.  If there are multiple places, you may need more feeders to get a sound part.  There are other tricks, such as placing the sprue near a section of the part (but not connected) so that there is enough thermal mass in the area that it acts like a continuous section and thus can be a larger sphere.  The more technical approach for this methodology is called the casting modulus, and there are equations to determine feed lengths based on surface area and volume ratios and such.  

[Alex Jemetz]

On 10/15/2025 at 8:01 PM, Jerrod Miller said:

This is not always the case.  There are times where geometry dictates that you have a very small gate (where the sprue meets the part) so that it freezes off there without any feeding in either direction.  Feeders (aka risers) can be added virtually anywhere to provide extra feed metal during solidification.  It is just pretty common practice to use the sprue/runner/gate as a feeder, so that you limit the number of contacts on the part to clean up.  

 

There are a few important things to keep in mind when determining the feeding methods for a casting.  

• Not everything needs to be fed.  Sometimes a small amount of shrink, especially in the centerline or otherwise hidden area, is perfectly acceptable.  
• Visualize how the part is going to cool, and thus solidify.  Thin sections will cool fastest, and heavy sections slower.  Sections that aren't connected but have another heat source nearby (like the lugs on your part) will cool slower than if they were isolated. More info on this below.   
• There are 2 main parts to shrinkage:
  • Volumetric liquid shrinkage happens as the metal cools but does not solidify yet.  At this stage, there is no blockage of feed paths, so you just need excess metal in the system to be able to flow in to occupy the part cavity.  
  • Feeding shrinkage happens as the metal continues to cool and solidifies.  At this stage, it is possible for dendrites forming to cut off flow of new metal from the feeder (whether this is the sprue or some other source).  

To determine how/where to add feeders to a casting (with the intent of making it shrink-free), I like to use the "largest inscribed sphere" method for quick and dirty calculations.  Basically, start at the thinnest section of your part and work your way to a heavier section.  Imagine the volume of the largest sphere you can place in the mold cavity as you go along.  You keep getting bigger spheres as you move away from the thin section.  Once you are no longer able to get a bigger sphere in that section, make a note of that location.  Repeat the process for every thin point of the part.  If all your thin points lead to a single place where the largest inscribed sphere is, that is the one place you need to put a feeder.  If there are multiple places, you may need more feeders to get a sound part.  There are other tricks, such as placing the sprue near a section of the part (but not connected) so that there is enough thermal mass in the area that it acts like a continuous section and thus can be a larger sphere.  The more technical approach for this methodology is called the casting modulus, and there are equations to determine feed lengths based on surface area and volume ratios and such.  

This is a lot of useful information Jerrod, thanks for all the time you put into your reply.  I will reread it several times to make sure I understand all the nuances.

[Alex Jemetz]

Team, I have another question.  What is your opinion of the set-up I have in the attached pictures of my saber hilt mold?  I want to ensure any gasses have a way to escape which is the reason for all the sprues.  Let me know if I should make any changes, grateful for your advice.

 

[Jerrod Miller]

You have 1 sprue and 5 vents.  I don't think the vent coming off the end opposite your sprue is going to be doing you any favors.  If you wanted the best fill, that is where you would have the metal enter.  

 

As you have it rigged, keep in mind that the area I've indicated below is the choke point for the metal.  It will control how fast the metal can enter the mold.  Also, since the area opens up below that, the metal will be able to fan out beyond that as it falls.  

 

I assume you have this put together as a rough mock-up, and you will smooth all the connections for the red wax prior to investing.  

 

I have put together a couple super crude models to show a couple of ideas for you.  They are not necessarily to scale, just the rough concepts.  

 

In this one, the sprue (gray) should be round, but it was way quicker for me to draw it flat.  It should also enter the bottom at a tangent, so the metal swirls as it enters rather than abruptly hitting the wall square on.  The vents (red) are pretty straight forward.  Just make sure they are spaced away from the sprue enough that no metal gets into them from pouring, as they need to stay open for the gasses to get out.  

 

This is another fun approach.  Again, nothing is necessarily to scale, and it is quickly and crudely put together, but you can see the concepts for getting the metal in and the air out.  The vents and sprue (red) are hopefully pretty straight forward, though certainly not drawn to the optimal scale.  

[Alex Jemetz]

Wow!  Really appreciate the detailed answer Jerrod!  Your drawings have helped me and I'm likely to change my set-up to the second model you show where the cross-guard is filled from a horizontal position.  Thanks!  I will post pictures when I actually do the pour, hopefully this weekend.

Have a great day!

[Brian Dougherty]

@Jerrod Miller While we are talking about this (Polite way of hijacking a thread) I'm wondering about something that I have started to do out of pure instinct.  Either I'm daft for doing it, or it is a good idea.  If the latter is true, then it likely has a name in the real world outside my driveway castings, in which case I'd like to know what it is.

 

It seems like the extreme ends of parts are where they are most likely going to not fill.  I have started adding some extra sprue wax at these extremes.  This is simply a goofy notion of just pushing the edge of the part out further into a sacrificial piece.  The the sacrificial part fails to fill, then I am still good as long as the rest filled.

 

This isn't a very good example, but it is an earlier casting I was able to find a photo of where I started playing with the idea.  Those "Straps" that run up over the pivot points at the top are sacrificial.  The edges of the grid looking things are also built up with extra wax just to help the grid fill.  We millied it off afterwards.

 

Virtually everything I cast is done with vacuum assist, so that probably changes things.

 

 

 

 

 

 

 

 

 

Edited October 21, 2025 by Brian Dougherty

[Jerrod Miller]

I generally call everything like this a "flow-off".  Vents very explicitly are to get air (and other gases) out of the mold cavity, and generally lead all the way out of the mold.  Sometimes they just get near enough to the surface to let the permeability of the mold material handle the rest, but usually they go all the way out.  Gas getting trapped where it can't get out faster than the metal needs to get in can cause filling defects, so we have to give it somewhere to go.  Venting it to atmosphere is a great way to go as it ensures it all gets out and is often easy to clean up the vent contact to the part.  Sometimes, that isn't practical, though.  Sometimes we just have to move the filling defect to an area that we can either tolerate it or easily remove it (much like the shrink issue I mentioned in a previous post).  In sand casting, this is also a good trick for dealing with loose sand that may be in the mold cavity (e.g. sand that fell down the sprue).  We can add a flow-off, typically directly on the gating, specifically to have the metal wash the sand into it so it stays out of the part.  In this case, we generally call those flow-offs "sand traps".  

[Alex Jemetz]

Update:  lost wax casting of the Ottoman Kilij cross-guard.  I learned something new:  how to turn brass into copper.  My mistake was to re-use a lump of brass I had previously melted to cast the guard for the first saber I made.  This caused my mold to fail as there wasn't sufficient molten brass to fill the casting as the zinc was partially or mostly burned off in the brass lump.  I took this as a learning experience but I'm now at a crossroad:  do I carve another wax mold and try for another lost-wax casting attempt?  Or do I use this 3.5 x 6 inch block of solid brass and cut/shape the guard with brute force?  I'm leaning to another attempt at lost-wax casting but wanted to share my experience and hear any feedback.  Hope you guys have a productive weekend. 

[Gert Burkhardt]

I have cut some smaller hilt pieces with a CNC from a solid brass billet. It is surely possible but takes some machine time and, more importantly, is a very expensive proposition. E.g. a 3"x3"x6" brass block cost $380 at McMaster and you will probably end up chipping away 90% of the material.

 

Are you really set on Brass? I found bronze much less problematic to cast and if you use aluminum bronze it pretty much looks just the same in color.

[Alex Jemetz]

Thank you Gert, I am aiming to use as little advanced machine tools as possible.  I've decided to do another lost wax cast and started carving the mold last night, determined to master this method!

[Jeroen Zuiderwijk]

Depends on what you want, to have a guard or to learn lost wax casting. Learning to cast can be a full time hobby in itself, with potentially more challenges and a steeper learning curve then forging to get a low level result that is useful. Aside from lost wax carving, red casting sand is a lot easier to work with. Though in your case, you would need to use a core, which has additional challenges. And going from brass to bronze would make your life a lot easier as well, as brass melting and casting has it's own difficulties. 

[Jeroen Zuiderwijk]

P.s. if you want to do more attempts, you can also make a silicon mould, so you can cast your waxes. That way you don't have to carve a new wax every time, and you can choose an easier to carve material to make the model, like wood. Then you can make a whole bunch of waxes, and experiment with that.

[Alex Jemetz]

Thank you Jeroen, I am planning to take your advice and make a master model from wood and then make a silicon mould so I can make quick waxes.  Just waiting now for a block of time and warmer weather to burn-out the wax and pour brass right after.

[Alex Jemetz]

Update:  Mistakes were made but I'm learning from them...  Posting what I am doing so I can get any advice from our community.

1.  I finally mixed the Hydroperm in the correct proportion to water and created a mold that withstood the wax-removal and brass pour.

 

2.  Brass melt went well and I removed the dross before pouring but it turns out despite my expectation that heating the mold 7 hours at 600 F on my BBQ, I didn't remove ALL of the wax and because of this there were voids in the cross-guard.

 

My forge is built of ceramic kiln bricks so on my next attempt I will adjust the size to accommodate the Hydroperm mold.  I have a question:  would it be possible to dissolve the wax casting using mineral spirits and then allowing them to evaporate before heating the mold or is this a crazy/dangerous idea?

 

Also, would using a propane torch to heat the inside of the mold to try to get all the wax out a valid idea?

 

Here are some pictures of the process:

 

 

[Alex Jemetz]

Maybe I should have added more vents?  Below is a picture of what the wax casting, pour funnel and vents looked like:

 

[Gert Burkhardt]

I am not familiar with the Hydroperm mold material (I use regular investment gypsum powder).

IIRC, most lost wax makers recommend >950 deg F for a full burn-out, decomposition and oxydation of residual carbon.

According to my investment powder instructions, I go gradually to 1300-1350 degrees Fahrenheit, hold for 2-3 hours, then cool down to 900 degrees for pouring.

After that burnout, the mold is white again with no trace of the wax. Under 1000 degree, I still see black residue.

Edited March 22 by Gert Burkhardt