2.5 Inch Diameter Ice Ball Mold

I think they are all aluminum, with copper color anodizing

Copper I think is not food safe (at least without a, frequent, special processing)
 
Bare copper is not a good idea for food pans etc as any acidic foods will produce copper compounds.
Copper pans are usually coated with pure tin to make them food safe but must be re-coated once the tin is worn through.
The one exception I know about is a copper bowel to whisk egg whites in, they stay white and firm up fast. Excellent for meringues and souffles.
 
Of course...cooking is the simplest form of diy, no proper machinist can avoid being a proper cook!
Thanks for the advice savarin, I'll try it!
 
I am planning also to make a mold like this for use in my brother's cafe.

The idea is to use the lathe. Single cylindrical aluminum block, drill the pilot holes while still in one piece then saw in half. Use a ball turning tool to make the recess on the first part, repeat for the second. Easy machining, easy polishing!
I just need to design and build a nice ball turning tool capable of turning internal half-spheres! I have not, yet, seen such a design. It need to fit in the 'hole' it is going to turn...

(just for fun, do an online search....these molds are sold up to more than $1k......)

The attached link has version #1 of a QCTP holder to turn the radius for the ice ball. http://www.hobby-machinist.com/threads/radius.29889/ I have since modified the tool to work much better and fit on my larger lathe.

A couple of issues that I had:
- The tool has to fit inside the radius when complete. For the 2.500" ball, you need to cut in 1.250" and it needs to be dead center or you will get a irregular shaped ice ball. It is difficult to obtain the 2.500" width @ 1.250", not impossible but a pain.
- Designing the pivot point, I used a bronze bushing and clips to hold the piece that rotates. I made my on tool holder to hold a HSS insert. It is a decent set up, but over time you will get some movement.
- If you are using a small lathe, make sure you lock the saddle down, it will move. This is the reason I went to my larger lathe. I also set my cross slide and locked it in place. I set the compound at 90deg and used it to feed into the workpiece. This was the best set up for me.
- I measured everything off the center of the blanks. If you use the outside, it is more likely you will be off and have an ice ball with a ring around the center of it.
- You need to drill a small hole through the center of the sphere for air and water to escape. If not, you will can end up with a flat spot at the top and bottom. Too big a hole and you have a icicle.
- I have found that turning from outside to inside is better.
- Use plenty of lubricant, which ever you use for aluminum. I used wd-40 and it worked well.
- Speeds are critical.
- The aluminum gets cold very fast. The first ice ball will be no problem, the second it slows down significantly, the third…. well, you need to warm up the ice ball maker.

I have more info if you are interested. I also bought one of the ball nose carbide cutters as mentioned. It works great, but you need some power and a rigid lathe. I tried several ways, used the radius tool I made and finished up with emory clothe, used the tool and the ball nose, and used the ball nose by itself. No real difference as long as everything is set up.

Cheers,
Dan
 
The attached link has version #1 of a QCTP holder to turn the radius for the ice ball. http://www.hobby-machinist.com/threads/radius.29889/ I have since modified the tool to work much better and fit on my larger lathe.

A couple of issues that I had:
- The tool has to fit inside the radius when complete. For the 2.500" ball, you need to cut in 1.250" and it needs to be dead center or you will get a irregular shaped ice ball. It is difficult to obtain the 2.500" width @ 1.250", not impossible but a pain.
- Designing the pivot point, I used a bronze bushing and clips to hold the piece that rotates. I made my on tool holder to hold a HSS insert. It is a decent set up, but over time you will get some movement.
- If you are using a small lathe, make sure you lock the saddle down, it will move. This is the reason I went to my larger lathe. I also set my cross slide and locked it in place. I set the compound at 90deg and used it to feed into the workpiece. This was the best set up for me.
- I measured everything off the center of the blanks. If you use the outside, it is more likely you will be off and have an ice ball with a ring around the center of it.
- You need to drill a small hole through the center of the sphere for air and water to escape. If not, you will can end up with a flat spot at the top and bottom. Too big a hole and you have a icicle.
- I have found that turning from outside to inside is better.
- Use plenty of lubricant, which ever you use for aluminum. I used wd-40 and it worked well.
- Speeds are critical.
- The aluminum gets cold very fast. The first ice ball will be no problem, the second it slows down significantly, the third…. well, you need to warm up the ice ball maker.

I have more info if you are interested. I also bought one of the ball nose carbide cutters as mentioned. It works great, but you need some power and a rigid lathe. I tried several ways, used the radius tool I made and finished up with emory clothe, used the tool and the ball nose, and used the ball nose by itself. No real difference as long as everything is set up.

Cheers,
Dan

Hi there Dan, thanks for the helpful info!

My plans for the radius turner include a ball bearing on top (could not find a taper roller bearing small enough) and a thrust bearing on the bottom there the nut will hold the tool's shaft. And, yes, the design is such that the tool with fit up to its axis of rotation inline with the face of the blank. Of course, the tool has, still, to be constructed if I find some time...

My lathe is a small 10" so everything but the top slide will be locked. Thanks for the advice on centering, of course I would carefully adjust tool height, but had not taken into account the importance of centering on the cross-slide axis (though I imagine after the first fail I would...). Centering also seems to be crucial for repeatability, so that you can actually make the second half of the mold identical.
It seems that it would be nice to have a mark or pin-locking mechanism so that the tool can me locked in the 'straight' (as in parallel to the ways' axis) position, so that it can be centered easily in terms of the cross-slide axis.

Hope I can find the time to build the thing and test it.....

best regards!
 
Hi there Dan, thanks for the helpful info!

My plans for the radius turner include a ball bearing on top (could not find a taper roller bearing small enough) and a thrust bearing on the bottom there the nut will hold the tool's shaft. And, yes, the design is such that the tool with fit up to its axis of rotation inline with the face of the blank. Of course, the tool has, still, to be constructed if I find some time...

Hope I can find the time to build the thing and test it.....

best regards!

Good idea, the issue I had was space. As the tool enters the workpiece it quickly converges to the center in all directions- The sphere. At the edge you will end up with 2.500", but that is after you are done. Keep in mind that when you are short of the 2.500" your tool needs to fit inside. From the bottom, the sphere curves up and from the top it curves down. You will need to have a clearance radius in the tool to fit workpiece radius. Don't forget to add in clearance for the insert or lathe bit and the handle when you turn it. The handle was one of my first mods, as I rotated to 90deg it hit the workpiece. I ended up off setting the handle 45deg.

Centering the tool was/is probably the biggest pain I ran into. The idea of having a locking-pin is a good one. Building the tool was a nice challenge and I learned a great deal. I enjoyed working on the tool more than building the ice ball mold.

The other challenge is accurately measuring the depth or radius to 1.250". The diameter is easy to measure. As long as you use the center for all measurements, you will line up nicely. I used the guide rods to mate the two pieces together and turned on centers to turn the OD and make sure they mated properly. I drilled a small hole on the bottom and used a stud as a driver when turning the OD.
 
The other challenge is accurately measuring the depth or radius to 1.250". The diameter is easy to measure. As long as you use the center for all measurements, you will line up nicely. I used the guide rods to mate the two pieces together and turned on centers to turn the OD and make sure they mated properly. I drilled a small hole on the bottom and used a stud as a driver when turning the OD.

Hi drom68,

I imagine the scenario (which I guess you also followed) should be:

- turn a cylinder's OD true
- part or saw the cylinder
- use the true OD surface to center the first half, do the radius turning. repeat for the second half
- drill the small leak-hole
- probably put it together using the guides and turn the OD of the complete tool perfect.

I have omitted the description of the guide-hole drilling. Well, it is because I have not, yet, fully sorted this out. Initially I thought about drilling the part while the cylinder is still one part, but deep hole drilling is not my favorite, and, mostly, I wouldn't like to have through holes on the mold. So, drilling the two parts separately it is. Of course, alignment is the big issue. Transfer bolts will be used to transfer the holes from the first drilled part to the second. But, only the OD of the parts would be true, and I am not sure if I can perform the hole-transfer procedure properly. Perhaps a V-block where two parts will lie, with one of them clamped and the other hit by a mallet.
How did you pull this procedure through?
 
[QUOTE="thanvg, post: 302010, member:
I have omitted the description of the guide-hole drilling. Well, it is because I have not, yet, fully sorted this out. Initially I thought about drilling the part while the cylinder is still one part, but deep hole drilling is not my favorite, and, mostly, I wouldn't like to have through holes on the mold. So, drilling the two parts separately it is. Of course, alignment is the big issue. Transfer bolts will be used to transfer the holes from the first drilled part to the second. But, only the OD of the parts would be true, and I am not sure if I can perform the hole-transfer procedure properly. Perhaps a V-block where two parts will lie, with one of them clamped and the other hit by a mallet.
How did you pull this procedure through?[/QUOTE]

1. Cut each blank
2. Face both ends and turn OD ( don't worry about actual dia. just turn so it is round)
3. Find center. I did this on the lathe since I just turned it.
4. I did the guide rod holes by using a rotary table. I centered the blank under the mill, then went out 2.500". Drilled first hole and then rotated blank 180deg and drilled second hole. Repeat for other blank. The top blank you want the holes a little bigger and put a slight radius or it will be a pain to line up. This also allows for a little error as well. With the rotary table it was spot on. You can also do this without a rotary table. You will need to have center and move out the distance you want and then move back to center and reverse for the other side. I did this, not as accurate but it will still work.
5. Place in lathe and turn radius.
6. Drill small vent hole.
7. Insert guide rods, put item together and then turn on centers. This gave me a solid one piece look and kept everything in line.
8. Put what handle you want on top.
9. Clean up and put first large ice cube in and try it out. If all is well, drop ice in drink and enjoy.
 
1. Cut each blank
2. Face both ends and turn OD ( don't worry about actual dia. just turn so it is round)
3. Find center. I did this on the lathe since I just turned it.
4. I did the guide rod holes by using a rotary table. I centered the blank under the mill, then went out 2.500". Drilled first hole and then rotated blank 180deg and drilled second hole. Repeat for other blank. The top blank you want the holes a little bigger and put a slight radius or it will be a pain to line up. This also allows for a little error as well. With the rotary table it was spot on. You can also do this without a rotary table. You will need to have center and move out the distance you want and then move back to center and reverse for the other side. I did this, not as accurate but it will still work.
5. Place in lathe and turn radius.
6. Drill small vent hole.
7. Insert guide rods, put item together and then turn on centers. This gave me a solid one piece look and kept everything in line.
8. Put what handle you want on top.
9. Clean up and put first large ice cube in and try it out. If all is well, drop ice in drink and enjoy.


Thanks for the advice drom68!
The thing is, however, the more I keep thinking about the 'aligning-using-the-guides idea' the less I like it. The critical alignment on this device is to ensure that the two hemispheres are totally concentric. And, the order of operations you suggested, and also the ones I thought off contemplating on this, if using guides as reference, can't really ensure required alignment, due to the many chuck/unchuck operations.

What I think would be ideal, is to construct a male/female mating surface on the two blanks, by turning appropriate recesses on the face of the blanks. The recesses with be turned on the same set of operations as the radius turning, without disturbing the chucked blank. This way, recess will be dead center with the hemisphere. All you need to do is turn the recess of the second blank so that you have a nice sliding fit, which can be obtained by trial and error since the first blank will, already, finished.
I was about to add "and then use whatever method is suitable for the drilling the guide holes, e.g. hole transferring would be a breeze having the recess already", but, not sure that guides are required after all...Have to try and see.

Disclaimer: of course, I never forget that we are talking about ice molds, and absolute levels of accuracy are not required. But, we all know, it is not so much about the destination as it is about the journey...Working to required tolerances is, of course, sign of experience and proper engineering, there is no need to build a mallet with micrometer accuracy, however for me as an amateur such tasks are useful training, mostly mentally. Could I make it with such levels of accuracy if it was required? Well, I'll try....
 
Thanvg, I was in the same mindset as you are when I was making mine. I have learned a little more on when to be accurate and when to be good enough. This project is great training for that. There are a thousand solutions, it is up to you to decide on which one you want to use. I do caution against making a mallet with a micrometer approach when working on the mold. The radius tool you make is where you want to be as accurate as you can. The mold is where you want to loosen up and be within a tolerance.

As far as turning recesses, that is one of a thousand ways. It is not needed, but if you feel comfortable with it- I say do it. I think that is what makes this a great project, you can try several ways and see what the outcome of it is. Can be a little costly as aluminum can be pricey. I had excess and was able to try several ways and learned a great deal.
 
Back
Top