DIY ice ball maker

[Junkbot]

So I'm sure people have seen the original Taisin model.

But the price was fairly ludicrous ($1000+). Then other companies started making knock offs with different materials that were a bit cheaper.

Now I've come across DIYers that have made pretty much the same thing for even cheaper.

Sold on Etsy.

My question is, has anyone here tried to lathe one of these themselves? I'm sure the material is a bit expensive, but the concept is so simple. The DIY post above goes into some detail on the process, and it looks a bit more involved.

[tanstaafl2]

Would be a fun novelty to have. I certainly don't have the skill or desire to make one myself but they are starting to approach a price point where I might really consider buying one. And I am guessing at least some bars would as well.

But still a bit much for me. Get it down under $200 and I could give it some serious thought!

[slkinsey]

Those things are really only useful if you're starting from clear block ice. Otherwise, just use a spherical ice mold.

[Junkbot]

Those things are really only useful if you're starting from clear block ice. Otherwise, just use a spherical ice mold.

I agree. I'm following Camper's method of clear ice formation through directional cooling (stick a cooler of water into the freezer), and have made wonderfully clear ice. Was hoping to get more fancy with the presentation than chunks of crystal clear ice. I know Camper has experimented with 'molds' in the cooler with success, and as cool as perfect rectangles of clear ice are, I'm still trying to figure out a viable method of making ice spheres.

[tanstaafl2]

Those things are really only useful if you're starting from clear block ice. Otherwise, just use a spherical ice mold.

Is there a good mold you would recommend? The few I have seen seem to get mixed reviews at best. Amazon for example offers these which seem to get a few decent reviews (I even had them on my Christmas list but Santa apparently thought better of it...).

I may still buy them just to satisfy my own curiosity and see how they do!

They also have these which look like they might do a pretty good job but are far more than I would be interested in paying for as an experiment!

[slkinsey]

I have several of these from the MoMa store. Work like a charm.

[tanstaafl2]

I have several of these from the MoMa store. Work like a charm.

Thanks! I will check it out.

[HowardLi]

Is this just for presentation?

Otherwise, there's no benefit to spherical ice.

[slkinsey]

It depends, I think, on whether or not the liquid is chilled prior to introduction to the ice.

[HowardLi]

It depends, I think, on whether or not the liquid is chilled prior to introduction to the ice.

If you're talking about how much ice gets melted, yes. But that doesn't depend on the shape or surface area of the ice, only the temperature differential.

[Junkbot]

It depends, I think, on whether or not the liquid is chilled prior to introduction to the ice.

If you're talking about how much ice gets melted, yes. But that doesn't depend on the shape or surface area of the ice, only the temperature differential.

The purpose of spherical ice is to minimize the surface area of ice/water contact. By minimizing the surface area, less ice melts in a given time compared to ice with a comparatively larger surface area. This is most beneficial for drinks you want to keep cold, but not get too diluted. Another experiment to try is to get the same mass of cubed and crushed ice, put them in separate cups of water, and observe which melts first.

Also, ice balls are very nice for presentation

[jrshaul]

I know a machinist who can produce molds like that for under $200 - just call up a local shop and they should be able to do it no trouble. I might try it myself.

[qrn]

Unless you have a Metal lathe its probably a non starter(I do ,but not going to try any)

Bud

[dcarch]

It depends, I think, on whether or not the liquid is chilled prior to introduction to the ice.

If you're talking about how much ice gets melted, yes. But that doesn't depend on the shape or surface area of the ice, only the temperature differential.

--------

Also, ice balls are very nice for presentation

I think that's the only reason. Otherwise a chilled glass ball dilutes your drink the least.

dcarch

[HowardLi]

It depends, I think, on whether or not the liquid is chilled prior to introduction to the ice.

If you're talking about how much ice gets melted, yes. But that doesn't depend on the shape or surface area of the ice, only the temperature differential.

By minimizing the surface area, less ice melts in a given time compared to ice with a comparatively larger surface area.

How do you figure?

[ScoopKW]

By minimizing the surface area, less ice melts in a given time compared to ice with a comparatively larger surface area.

How do you figure?

A sphere holds the most volume with the smallest surface area. Cubes are going to slough off more water than a sphere.

But I think the whole sphere thing is a waste unless using hard, clear, relatively oxygen-free ice. And while I have a lathe, it's not for metal working. Surely someone here is friends with a machinist?

[Junkbot]

It depends, I think, on whether or not the liquid is chilled prior to introduction to the ice.

If you're talking about how much ice gets melted, yes. But that doesn't depend on the shape or surface area of the ice, only the temperature differential.

By minimizing the surface area, less ice melts in a given time compared to ice with a comparatively larger surface area.

How do you figure?

I believe melting ice is a function of surface area (given all else is equal) since the ice-water interface is where the most heat transfer occurs. Crushed ice will cool a drink considerably faster since it has a much higher area to transfer heat compared to cubed ice. However, that means that crushed ice will melt much faster as well, resulting in drinks that come back to room temperature in a relatively short period of time. A cube of ice will cool a drink much slower and most likely to a warmer temperature compared to the crushed ice.

Overall, the same amount of energy will be transferred from the ice to the water as they both will eventually come back to room temperature, but the time it takes for that to happen will be different.

But I think the whole sphere thing is a waste unless using hard, clear, relatively oxygen-free ice.

This can be easily done by freezing water in a cooler. The directional cooling (freezing from the top down) will effectively 'push' all dissolved gases and impurities down, resulting in very clear ice. Camper English has a blog that experimented with this idea a ton (www.alcademics.com).

Edited January 11, 2012 by Junkbot (log)

[dcarch]

The thinking should also take into consideration that only part of the ice is submerged, regardless of the shape.

By definition, the total BTUs(latent heat) in a given weight of ice counts.

How fast will the ice cool your drink? depends also on if you stir your drink or not.

Spherical ice does not give you the option of quantity of ice like crashed ice.

dcarch

[HowardLi]

The surface area of the ice is one of the factors that affect cooling rate, yes. But we are not interested in cooling rate, are we? No, only how much water gets melted for an equivalent amount of cooling.

But if you think about it, is it possible that one type of ice can be better than another? Assume all the ice is at the same temperature. You have two cups of water, one with cubed ice and one with a large ice sphere. The cubed ice glass will cool much faster. That much is obvious. Then, how do we determine how much ice melts, when the glasses are at equilibrium? Simple - the ice melts in proportion to the added heat energy from the liquid that is cooling. But, the (relative) heat energy is the same in both glasses! So what is in the cubed ice glass that allows for more ice to melted than the large sphere ice?

If you can figure that out, you may be able to win a Nobel prize...

Edited January 12, 2012 by HowardLi (log)

[bmwrtmike]

LINK Would say a 2" stainless steel sphere stored in the freezer work better? No Ice to melt, no chance of picking up odors from the freezer, and alot less money to have a few in the freezer.

[Junkbot]

The surface area of the ice is one of the factors that affect cooling rate, yes. But we are not interested in cooling rate, are we? No, only how much water gets melted for an equivalent amount of cooling.

But if you think about it, is it possible that one type of ice can be better than another? Assume all the ice is at the same temperature. You have two cups of water, one with cubed ice and one with a large ice sphere. The cubed ice glass will cool much faster. That much is obvious. Then, how do we determine how much ice melts, when the glasses are at equilibrium? Simple - the ice melts in proportion to the added heat energy from the liquid that is cooling. But, the (relative) heat energy is the same in both glasses! So what is in the cubed ice glass that allows for more ice to melted than the large sphere ice?

You'd be correct that the heat energy exchange between the liquid and ice is the same in both glasses (how much water gets melted for an equivalent amount of cooling) if the glasses were in a closed system, which they are not. You're assuming that both drinks will have the same equilibrium temperature, and thus no energy loss to the outside environment. However, the drink will not have the same equilibrium temperature. Because of the greater surface area of the cubed ice, the drink will be overall colder since it is drawing energy out of the liquid at a faster rate (faster than the rate at which the room/air can heat up the drink). It may not be much, but there will be a temperature difference in the drink with the spherical ice since the rate at which it cools the drink will be less than the rate at which the cubed ice cools its drink. Hence, if the room/air is putting heat into the system at a constant rate, the drink with the smaller cooling rate (spherical ice) has to have a greater temperature.

Another way to think about this is with CPU heatsinks. You'll notice that every design is finned to maximize surface area, which increases the rate at which heat can be dissipated, thus lowering the overall temperature of the CPU.

Ultimately though, simply do the following observational experiment: put crushed ice in one cup of water, and the same mass of cubed ice in the other. Which one melts first? Granted the crushed ice will make the drink colder, but it will not last as long.

The most used application I've seen for the ice balls is for whiskeys, particularly scotches. Most people don't really want their drink to be super cold, just cooler than room temperature (which is why most people don't put scotch freezer). Furthermore, they don't want to water down the drink too much, but a little water is desirable. I'll be the first to admit that the rate at which spherical vs cubed ice melts won't be that great, but the idea is scientifically sound. And again, the presentation is over the top, which I think contributes to the overall experience.

LINK Would say a 2" stainless steel sphere stored in the freezer work better? No Ice to melt, no chance of picking up odors from the freezer, and alot less money to have a few in the freezer.

True, but I'm not sure how well they'd perform at cooling the drink. Steel has a specific heat capacity of ~0.5 J/gK compared to ice, which is around 2 J/gK. So you'd have to have a a lot of steel to cool the drink down and maintain that temperature. They make something like what you're suggesting: whiskey stones (http://www.thinkgeek.com/homeoffice/kitchen/ba37/), which are made of soapstone (~1 J/gK). Puts a whole new twist on whiskey on the rocks *rimshot*.

[dcarch]

I wonder if there is choking hazard with spherical ice.

To increase thermal capacity of a steel sphere, the inside of the sphere can be filled with phase-change gel.

dcarch

[slkinsey]

LINK Would say a 2" stainless steel sphere stored in the freezer work better? No Ice to melt, no chance of picking up odors from the freezer, and alot less money to have a few in the freezer.

True, but I'm not sure how well they'd perform at cooling the drink. Steel has a specific heat capacity of ~0.5 J/gK compared to ice, which is around 2 J/gK. So you'd have to have a a lot of steel to cool the drink down and maintain that temperature. They make something like what you're suggesting: whiskey stones (http://www.thinkgeek.com/homeoffice/kitchen/ba37/), which are made of soapstone (~1 J/gK). Puts a whole new twist on whiskey on the rocks *rimshot*.

More to the point, the reason ice is so good for cooling a drink is precisely because it melts. The heat of fusion for ice is 334 J/g. In fact, virtually all of the cooling action of ice in a cocktail results from melting. Ice and stones and whatnot don't melt, and therefore their cooling effect is dwarfed by that of ice.

[HowardLi]

The surface area of the ice is one of the factors that affect cooling rate, yes. But we are not interested in cooling rate, are we? No, only how much water gets melted for an equivalent amount of cooling.

But if you think about it, is it possible that one type of ice can be better than another? Assume all the ice is at the same temperature. You have two cups of water, one with cubed ice and one with a large ice sphere. The cubed ice glass will cool much faster. That much is obvious. Then, how do we determine how much ice melts, when the glasses are at equilibrium? Simple - the ice melts in proportion to the added heat energy from the liquid that is cooling. But, the (relative) heat energy is the same in both glasses! So what is in the cubed ice glass that allows for more ice to melted than the large sphere ice?

You'd be correct that the heat energy exchange between the liquid and ice is the same in both glasses (how much water gets melted for an equivalent amount of cooling) if the glasses were in a closed system, which they are not. You're assuming that both drinks will have the same equilibrium temperature, and thus no energy loss to the outside environment. However, the drink will not have the same equilibrium temperature. Because of the greater surface area of the cubed ice, the drink will be overall colder since it is drawing energy out of the liquid at a faster rate (faster than the rate at which the room/air can heat up the drink). It may not be much, but there will be a temperature difference in the drink with the spherical ice since the rate at which it cools the drink will be less than the rate at which the cubed ice cools its drink. Hence, if the room/air is putting heat into the system at a constant rate, the drink with the smaller cooling rate (spherical ice) has to have a greater temperature.

At equilibrium, the ice will also be at 0 degrees (or whatever temperature the liquid is at due to alcohol depression). Because both glasses have the same amount of heat energy in the ice, whether the ice starts at -20C or -8C, the same amount of ice would melt until the ice and water were at the same temperature. Therefore, the only melting after equilibrium would be due to the heat gain from the air of the room, which, for all intents and purposes, would be the same for both glasses.

EDIT: I just realized something. If the drink is colder than 0 degrees

Edited January 13, 2012 by HowardLi (log)

[Junkbot]

At equilibrium, the ice will also be at 0 degrees (or whatever temperature the liquid is at due to alcohol depression). Because both glasses have the same amount of heat energy in the ice, whether the ice starts at -20C or -8C, the same amount of ice would melt until the ice and water were at the same temperature. Therefore, the only melting after equilibrium would be due to the heat gain from the air of the room, which, for all intents and purposes, would be the same for both glasses.

EDIT: I just realized something. If the drink is colder than 0 degrees

How do you explain crushed ice melting faster than cubed ice based on the above?