time to melt 1 ton ice 3

[johnhjo]

Hi,

I would like to know how much time is gonna take to melt 1 ton of ice in different kind of outside temperature. Is there any formula/reference/table I could learn from?. Assuming under normal pressuree ( 1 atm ) and no other heat source. Thank you.

 

[JStephen]

You can figure the amount of energy required to melt the ice readily enough, but figuring the rate at which that happens would get rather involved. It would depend on temperature, wind speed, exposure to the sun, shape of the ice and provision for drainage, etc. Back in the old days, they'd harvest ice, stack it in warehouses surrounded by sawdust, and it would last through the summer (that didn't work so well in Texas, though).

 

[ddkm]

To solve this, you would need to know the total heat source Q. In your case, this "heat source" is the environment of the block of ice.

With a known Q, then it can be represented as follows:

Heat Required to melt a mass of ice is:
Heat = mL

where m = mass of ice (in this case, 1 tonne or 1000kg)
L = latent heat of fusion

But Q = Heat/t where t = time taken

therefore:

t = Heat/Q


Example:
Block of ice at atmospheric condition
m = 1000kg
L = 333.7 kJ/kg (at 1 atm)
Q = Q (in kW)
Therefore, the time taken:
t = 1000kg x 333.7kJ/kg / Q kW
= 333,700/Q seconds



---engineering your life---

 

[ddkm]

Sorry, I should have used brackets, which would be clearer:

Let me try again. For the bottom section:

Therefore, the time taken:
t = (1000kg) x (333.7kJ/kg) / (Q kW)
= 333,700/Q seconds



---engineering your life---

 

[tombmech]

Actually, I thought this was a trick question. One ton of a melting block of ice provides 12,000 Btuh (BTU's per hour) over a 24 hour period. This was the standard that created the "TON" definition for refrigeration.

 

[sailoday28]

The original question was time to melt. See the response of JStephen (Mechanical)

 

[johnhjo]

Well, the idea is to learn how efficient chiller is with ice thermal storage. I would like to know (if possible) formula from strach. I hear a rumor saying ice thermal storage is efficient in concept but not in reality even ice is made during off peak hour.

 

[MintJulep]

ddkm's derivation is correct. However, note that Q is not a constant. Q is a function of the surface area available for heat transfer. As the ice melts, the area changes.

 

[IRstuff]

I didn't see anything that incorporated the specific heat part of the process. My father's refrigerator is busted and the freezer compartment runs at a constant -20ºF. You need to warm that block of ice up to the melting point before the latent heat of fusion is applicable.

TTFN

 

[25362]

Even well below 0.01 deg C, ice has a vapor pressure. Just look at a phase diagram: below 4.58 torr (triple point) ice may sublime into vapor without turning first into liquid water. The same as with carbon dioxide on Mars; or, more down to Earth, with naphthalene moth balls.

In short, given the right conditions, sublimation by absorbing heat from the surroundings can take place at constant temperature and pressure.

 

[Latexman]

This would be so easy and inexpensive to get real data on! Do you have a scale with a weight recorder vs. time? You'll have to ensure the water runs off, but the ice doesn't. A support with a "holey" top and sloped bottom. Make up 4-5 blocks of ice. One is 1 in[sup]3[/sup], one is 10 in[sup]3[/sup], one is 100 in[sup]3[/sup], and one is 1000 in[sup]3[/sup] in size. Get the idea?

Good luck,
Latexman

 

[FredRosse]

Ice thermal storage can be used to increase the nominal capacity of a refrigeration system which has daily load variations, such as a typical building air conditioning system. During the day, when air conditioning loads are high, the refrigeration machine plus the ice stored thermal energy system provide cooling. At night, the loads are reduced, and the refrigeration machine can make new ice for use the next day.

Ice chips in a water tank form a very good source of chilled water at about 32F (0C). One ton of ice melted in a 24 hour period is equal to 12,000 BTU per hour for 24 hours, the source for the common definition of a "Ton" of refrigeration. Melted ice water is then circulated to the refrigeration load directly, or to a heat exchanger, taking heat away from a conventional chilled water loop, typically about 56F (13C) in, 44F (7C) out.

If the tank is insulated, the efficiency can be very good, with virtually no significant losses to the environment.

Now for the bad news: In a normal chilled water system, the chillers work their thermal refrigeration cycle between two temperatures, and the larger the temperature difference, the less efficient the chiller cycle.

A typical chilled water system might have a refrigeration cycle that works work between 95F (35C), the ambient, and 37F (3C). This cycle cools the chilled water to 44F (7C), as some temperature differential is needed for the heat exchanger. The refrigeration cycle rejects heat to the environs at 95F (35C), and is cooled with 85F cooling tower water.

To make the ice however, the cold temperature of the refrigerating cycle must be at a temperature considerably lower than the ice temperature, perhaps 20F (-7C). The refrigeration machine must work against a larger temperature differential, and therefore is less efficient.

The ice making machinery consumes additional power to shave the ice from the surfaces upon which it forms, to transport the ice to the ice tank, and to stir the ice tank to maintain good heat transfer from the ice pieces to the water in the ice tank.

 

[tombmech]

Note: FredRosse got the jump on some of this before I was able to post it. I agree with most of his comments. However, there are other forms of ice storage than the ice-slurry scenario he describes. Regardless, the increased temperature differential with ice gives it a technical advantage in heat transfer effectiveness. That comes with significant operational disadvantages, though - so chilled water thermal storage is just as competitive, long term.

Well, the idea is to learn how efficient chiller is with ice thermal storage. I would like to know (if possible) formula from strach. I hear a rumor saying ice thermal storage is efficient in concept but not in reality even ice is made during off peak hour.

John,
If that was actually your intent, you asked the wrong question. Efficiencies have only an ancillary contribution to thermal storage. What you are really doing is time-shifting peak demand - the same as time-shifting a television program with a VCR. Thermal storage allows one to "record" ton-hours of cooling. This is much more than an equivalent substitute of the chillers used to "charge" the storage, though. The tank responds dynamically to load peaks, and may operate as if it were many more chillers than the numbers used to charge it.

As referenced in my "trick" question post above, one ton of ice possesses 24 hrs x 12,000 Btu/hr heat capacity. That capacity exists - regardless of the time used to melt it. That time is entirely dependent on a variable load. Moreover, the price in efficiency was already paid in freezing it.

Whatever heat transfer efficiency penalty was in effect when freezing the ice, becomes a heat transfer benefit in melting it.

In thermal storage systems, efficiency improvements are ancillary effects that occur by optimizing the run load of the chillers, and the condenser water temperature leverage. The run loads can be optimized to the most efficient loading of the chillers, since the load is entirely dictated by you in "charging" the storage.

Condenser water system advantages present themselves by more "leverage" with greater temperature gradients - e.g., cooling at night with reduced cooling tower temperatures. Chillers must be carefully selected to take advantage of this quality however - or you may find yourself adding heat to the condenser side of the process just to maintain head pressures. High-pressure chillers, or chillers with variable-speed drives, are more tolerant to lower head pressures, and take advantage of lower condenser water temperatures.

Those two factors have the largest effect on efficiencies with thermal storage. However, those efficiencies are still a net penalty compared to direct cooling.

The decision about ice versus water (as in a chilled water storage tank) is mostly an operational one. The ice process will be marginally more efficient in the heat transfer modes due to a higher temperature gradient. (As stated earlier, the efficiency of melting and re-freezing is contained within the process, though.) However, ice production itself contains many physical obstacles that penalize the process. On the other hand, water storage is a volume issue, and the temperature gradient must be consistent through the whole system - including the coils in the using devices. So, it has disadvantages, too.

If you have the space, though, I would vote for chilled water storage in almost every case - not ice.

 

[johnhjo]

Thanks for the post...

I thought the same as Latexman at first, but such experiment yield different result while room temperature change.

Tombmech, chilled water storage never cross my mind.... Does this technology could be applied on any chiller or only specific chiller can do?

 

[tombmech]

John,
That is the advantage of chilled water storage as opposed to ice - it doesn't take any special type of chiller. It does help if a wider deltaT is used - typically up to 20 degrees F. However, that's a "system" deltaT and is mostly in the tank and the coils. The chillers' deltaT's can be a little less, and the range is increased on the high end, not the low end. So, the chillers are still producing 42-45 deg.F. supply water - well within normal chiller technology.

What that means is that with the tank, you can take it or leave it - the chilled water system does not require it to operate. It only requires it if the peak loads exceed the capacity of the running chillers, and then you only need it for capacity during peak periods. Many installations use the tank all the time, though. On less than peak days, some systems can "coast" entirely with the thermal storage, and not run any chillers during the peak periods.

Chillers used in ice systems, on the other hand, cannot directly feed the load - ever. There are newer designs where the ice storage can be valved out of the circuit, and still operate the cooling system. However, even then, there is some method of recirculation and blending to bring the temperature back up to a usable mid-40's F.

The disadvantage with chilled water storage is space and volume. The higher deltaT's are desired to minimized this somewhat, but you still need a pretty good volume to pay for itself. It's not just volume, either - the tank must not exceed a certain aspect ratio between diameter and height. For a general rule-of-thumb, you want a tank with a diameter 2-2.5 times the height for successful stratification. Make it too tall to save footprint space, and you will have difficulty stratifying.

This is somewhat counter-intuitive until you consider that the interior skin of the tank has a proportionally higher effect on disturbing the water in taller tanks. Larger diameters allow the water to settle out more. Tank design is centered on stratifying the water as much as possible, creating a "battery" of charged water with a maximum temperature gradient from top to bottom.

Sorry for digressing so much - it's another favorite subject.

 

[tombmech]

Please let me correct a couple of statements:

For ice storage, you must typically run the water through some sort of heat exchanger mechanism to deliver any cooling to the end-use devices without the ice storage in the circuit. Although admittedly, my expertise at ice storage systems is limited. Quite honestly, I have rejected them at each opportunity in favor of the much simpler chilled water storage.

Also, there are chilled water storage systems, that if carefully optimized with ideally scheduled charging times, will actually end up with a net improvement in efficiency.

It's been my experience that contrary to conventional wisdom, the greatest savings advantage is in reducing the first cost of intitial equipment (saving numbers of chillers, pumps, and cooling towers). This has a tremendous cascading effect for the life of the system.

One system I put in was a 10,000 ton chiller plant using 2ooo ton centrifugals. Building a 2 1/2 million gallon storage tank allowed us to install only 3 chillers, not 5, for a net savings in first cost of several million$. In operation during peak summer periods, and even with a ridiculously cheap electric rate (<5¢/Kwh), savings of up to $8000 per hour during the summer afternoons was not unusual. (Note: these were demand charges with a real-time rate structure. KWH savings from the reduction of 5 chillers to 3 chillers adds even more savings!)

In round numbers, this added up to almost $500K over a typical summer. Over the minimum 20-year life of the plant, that's $10 million in savings. Add to that the fact that the tank was cheaper than the two extra chillers, and the whole exercise became a no-brainer. This was not an unusually dynamic load, either - a 60/40 split between active chillers and thermal storage is not uncommon.

 

[Latexman]

johnhjo,

Then record the room temperature vs. time too.


Good luck,
Latexman

 

[johnhjo]

Thanks again Tombmech,

No need to be sorry for, it is interesting subject to be discuss. I havent heard such a system using water storage here before (in my 2 years of hvac experience). The closest thing my company had done was installing water tank mixing normal water and chill water in order to maintain 15 deg Celcius water suply for industrial purpose ... By the way, do you have any reference about a book / home page discusing about chilled water storage design in depth?.

 

[tombmech]

Chicago Bridge and Iron, and Natgun tank systems can probably give you the quickest, down-to-earth explanations. There have also been many articles written through the years in the ASHRAE Journal. A Facilities Engineer from Texas Instruments wrote many of the articles. I can't remember his name or look it up right now, but those are good places to start.

The mistake many Engineers make is trying to equate thermal storage with a fixed rate of cooling capacity - and calculate the payback with that assumption. Depending on the load profile, though - thermal storage can act in a transient fashion as many more chillers. The whole point is to level the load with a fixed capacity of chillers, then use the tank for the transient peaks. ("Transient" in this context usually means a 4-5 hour period on a summer afternoon.)

 

[tombmech]

http://www.natgun.com/

CB&I is a huge process tank mfr (good experience), but this link will take you straight to their chilled water storage tanks:

http://www.cbi.com/services/strata-therm-thermal-energy-storage.aspx