Purdue E-Pubs

Li, Daqing

Shivashankar, Suresh

Purdue University 
Purdue e-Pubs 
International Refrigeration and Air Conditioning 
Conference School of Mechanical Engineering 
2024 
Dynamic Simulation of An Automatic Commercial Ice Maker 
Daqing Li 
daqing.li@copeland.com 
Suresh Shivashankar 
suresh.shivashankar@copeland.com 
Follow this and additional works at: https://docs.lib.purdue.edu/iracc 
Li, Daqing and Shivashankar, Suresh, "Dynamic Simulation of An Automatic Commercial Ice Maker" 
(2024). International Refrigeration and Air Conditioning Conference. Paper 2499. 
https://docs.lib.purdue.edu/iracc/2499 
This document has been made available through Purdue e-Pubs, a service of the Purdue University Libraries. 
Please contact epubs@purdue.edu for additional information. 
Complete proceedings may be acquired in print and on CD-ROM directly from the Ray W. Herrick Laboratories at 
https://engineering.purdue.edu/Herrick/Events/orderlit.html 
Dynamic Simulation of An Automatic Commercial Ice Maker 
Li, Daqing*; Shivashankar, Suresh 
Copeland Company, Cold Chain, 
Sidney, Ohio, US 
Daqing.Li@copeland.com; Suresh.Shivashankar@copeland.com 
ABSTRACT 
2101, Page 1 
A dynamic model of automatic commercial ice maker was established on GT-SUITE platform to study the 
performance of a R404A commercial ice maker and verified with test results. The model was adapted to a R134a 
modification of this ice maker and compared its performance with test data too. Finite state machine tool provided 
by the GT-SUITE platform is employed to simulate the operation mode switching of automatic commercial ice 
maker. 
1. INTRODUCTION 
The transient nature of ice machine operation that continuously cycles between ice formation and harvest modes 
makes steady-state simulation not suitable for system performance prediction. Some efforts have been made to 
implement a quasi-steady-state approach to model the dynamic behavior of an ice machine. The model implemented 
can have good agreement between model predictions and test data with heavy tuning of the model parameters. But 
this approach leads to the model only can deliver satisfying prediction performance when limiting to the specific 
refrigerant and machine configuration. A dynamic model of automatic commercial ice maker was established on 
GT-SUITE platform to study the performance of a R404A commercial ice maker and verified with test results. The 
model was adapted to a R134a modification of this ice maker and compared its performance with test data too. 
Finite state machine tool provided by the GT-SUITE platform is employed to simulate the operation mode switching 
of automatic commercial ice maker. 
2. SYSTEM DESCRIPTION 
The schematic of an automatic commercial ice maker is shown in Figure 1. Its refrigeration circuit consists of a 
compressor, a fin-tube condenser with a condenser fan, an evaporator coil attached to ice grid bin, a suction-liquid 
line heat exchanger, an expansion valve, and a hot gas valve. Its water circuit consists of a circulation pump, a water 
tank, a water filling valve, a drain valve, and a water distributor. The simulation model is focused on the evaporator 
freezing and harvesting process. During freezing phase, water is fed from top of evaporator water feeder to fill up 
the ice-cube grid and is being frozen into solid ice by the low temperature evaporating refrigerant flows through the 
evaporator coil tube that attached to the bottom plate of the evaporator ice-cube grids. Hot gas valve is closed during 
this period and expansion valve is opened to let liquid refrigerant from condenser being fed into evaporator to 
absorb heat generated from water frozen into ice. When ice inside the ice-cube grids reaches certain thickness, a 
micro-switch will be triggered to start harvesting process. Hot gas valve will be opened to let high temperature 
refrigerant gas from compressor discharge pipe being fed into evaporator coil to melt ice on the bottom plate into 
water and ice-cubes can be freed from ice-cube grids and drops into collecting bin below the evaporator. Expansion 
valve will be closed off during this period to stop feeding liquid refrigerant into evaporator. After ice-cubes being 
20th International Refrigeration and Air Conditioning Conference at Purdue, July 15 - 18, 2024 
2101, Page 2 
harvested, freezing phase will be restarted and hot gas valve will be turned off and expansion valve will be opened. 
Another cycle of ice-making will start again. Operation of automatic commercial ice maker is inherently a dynamic 
process cycling between freezing and harvesting phases with refrigerant flow being controlled by hot gas valve and 
expansion valve. 
Condenser 
:c 
)) Condenser Fan 
(( 
)} 
Hot Gas Valve (( 
g ) 
~ 
L !Evaporator Grid )t __ _, ..,,,) 1' ' _l_ -Ill'• .... .... 
- ~ - , ..... Expansion 
l) 
Valve C 
~ ----- .,. - ·-
II 
t ompressor 
I ~ Heat I 1~ - :I_ ,. ---,.~;;:y-l-c=._ -i-,1.1l-___ _Jl ) Exchanger 
Circulation ~!Drain ~ ~f Water Fill 
Pmnp 
Figure 1. Automatic Commercial Ice Maker Schematic 
2.1 R404A ACIM and Modeling Approach 
A R404A ice machine is picked to be modeled with following major characteristics: 
Compressor: a ten coefficients compressor map was provided 
Condenser: a fin-tube condenser with detailed information on configuration and flow circuit was given 
and air flow rate as well as fan watts 
Evaporator: a lump parameter approach will be employed to model icetray and a straight tube with 
total length of the coil will be modeled for in-tube refrigerant flow and heat transfer 
Expansion valve: a PID controlled electrical expansion valve will be modeled to maintain target 
superheat at evaporator exit. 
Connection tubes: all connection tubes length, pipe diameter and wall thickness are given. 
Liquid line and suction line are bundled together as suction-liquid line heat exchanger. 
3. GT-SUITE ACIM SIMULATION MODEL 
3.1 GT-SUITE Model of R404A Ice Machine 
A R404A ice machine GT-SUITE model is setup as shown in Figure 2. A finite state manager is used to control the 
sequential operation of the ice machine to cycle through freezing cycle, harvesting cycle and refill state 
automatically as real machine operation. Condenser is modeled with a detailed fin-tube heat exchanger that has a 
customized flow circuits arrangement. Evaporator is modeled as a straight tube that its wall temperature is 
determined from heat transfer of connecting ice tray bottom plate which is connected to the thermal mass of water 
inside icetray. A CV4.5HPQ EEV   is modeled as expansion device with a PID controller to maintain target 
superheat. Liquid line and suction line heat exchanger is also modeled. 
20th International Refrigeration and Air Conditioning Conference at Purdue, July 15 - 18, 2024 
Ambitnt-1 
0 
P11ssu11 
Monitor 
PhTsOlagram 
Refrlg-1 
Event: imtr-1 
n P-ie- ---~1>t--- ---1 •". 
Fi rd., 
Fil r-1 
,, 
t---o.'-i..-- ~ ,._- ----1 . ::::::;.• ,__--,---~ 
Finit1St1t1 
Manager-1 
Fi rd., 
Fil •2 
Figure 2. GT-SUITE R404A Ice Machine Model Schematic 
3.2 Finite State Manager Implementation 
2101, Page 3 
GT-SUITE provides a finite state manager tool to help implementing sequential operation based on specified 
conditions. Three states are defined for ACIM operation: ice-making, harvesting, and refilling with this tool and 
state transitioning sequence as well as conditions are defined as shown in Figure 3. 
,1 _....,,., ,I sto ... lTr- ,I o,puu ,I - ,I lnitWlullon ,I AdY-ed 
...... 
SIIII DewioOOM klMI~ !UfV'tlO,,,, .., ... 
Dun'lltll olCCNIN- OJ63 I 
rwnr.. o1•11 c- 01011!1l7 0.11.1, 
Nm 0-1"9t ·= ·= , __ 
1•12 
Tll«malC~ 9' .. 9' 
"""'' 
., 
.,_, 
9' 
"'""' 9' 
........ 
Tram,,oono.atotlOnl k1N1U'lf0orlt lfat"111!1r'QComolr.l 
n-,,.._,concll!lclM T._IC~ •31.5 r w1:ar •32.5 
SllrtSDW 
.,.,_ 
Figure 3. Finite State Manger for ACIM Operation Sequence Modeling 
20th International Refrigeration and Air Conditioning Conference at Purdue, July 15 - 18, 2024 
2101, Page 4 
3.3 Evaporator Icetray Model 
The evaporator icetray model is shown in Figure 4. A straight pipe element with total length of 378" (9.6m) of 
original coil length is used to model refrigerant flow and heat transfer inside the tube and an 18" (0.457m) 
discretization length is used to capture the phase change process of refrigerant inside the pipe . A thermal node is 
employed to model the bottom icetray plate that directly in-contact with the evaporator coil tube. A thermal block is 
employed to model the bottom part of icetray that will be melted into water during harvesting. A thermal mass is 
employed to model the water mass that will be frozen into ice cube. The thermal mass will be in contact with a 
temperature node at feeding water temperature and being assigned a very high convection heat transfer coefficient to 
simulate refilling of feeding water in icetray. 
Figure 4. Evaporator Icetray Model 
4. RESULTS AND DISCUSSION 
An ACIM was tested per ASHRAE 29-205, and ARI 810-2016 standards specification and test results are used to 
compared with GT-SUITE dynamic simulation results. A typical operation condition has ambient air temperature at 
90°F (32.2°C) and feeding water temperature at 70°F (21.1 °C). Water inside icetray temperature at center and 
bottom are plotted during whole 3600 seconds running time as shown in Figure 5. There are two completed freezing 
and harvesting cycles in 3600 seconds operation time and the third harvesting cycle is started but not completed yet. 
PMTomperalU18s 
70 
Therma1Mass pal1 lceTray_iee-1 
65 
00 
55 
50 
!!;: 45 
I! 
I ~ - Center - bottom 
! 35 
3J 
25 
20 
15 
10 
0 1Wl 1!m 2IJXI 25CO 
Tme(s( 
Figure 5. Evaporator Icetray Temperature 
20th International Refrigeration and Air Conditioning Conference at Purdue, July 15 - 18, 2024 
Evap 
SUCtion Uno flom 
Evaporator Col Outlet to 
comiressor-10 be 
lnSulaed 
Oil 
Equalzation 
Line 
0 
318" 
M1/2MC-07 
TXV 
318" 
¼" 
Uquld1D &Kllon Heat 
Exchanger 
318" 
Hot Gas Bypass 
Valve 
¼'" Process Port 
Equalzatlon 
Line 
¼" 
¼" 
¼" 
2101, Page 5 
The comparison between simulation results and test results for R404A ACIM is listed in Table 1. 
Table 1: Comparison between Simulation Results and Test Results for R404A Ice Machine 
Metric 90/70 °F (32 .2/21.1 °C) (No Turning) 90/70 °F 132.2/21.1 °C) (Turning) 
Item Test Model Delta Test Model Delta 
Total Cycle Time (min.) 20.5 20.2 -1.5% 20.5 20.5 0% 
Ice per 24 hrs. (kg) 169.8 172.6 1.6% 169.8 169.8 0% 
Energy per 45.4 kg Ice (kWh) 5.99 4.95 -17.4% 5.99 5.84 -2.5% 
Energy per 24 hrs. (kWh) 22.43 18.86 -15 .9% 22.43 21.85 -2.6% 
Only tuning done is adjusting condenser heat transfer factor to 0.5 to bring condensing temperature closer between 
test and simulation data. The difference in total cycle time and ice made per 24 hrs. are within 2% even without 
tuning and dropped to 0% with tuning. But energy input difference between model and test data are more than 15% 
without tuning and are within 2.6% after tuning. 
5. R134a ICE MACHINE MODIFICATION AND MODEL 
5.1 Modification of R404A Ice Machine into R134a Ice Machine 
The schematic of modification of R404A ice machine into R134a ice machine is shown in Figure 6. Two 
RRT73CIE compressors are installed to replace the original R404A compressor. Optimal refrigerant charge was 
determined with test. Test methods and instrumentations are still following the ASHRAE 29-205 and ARI 810-2016 
standards specification. With the given information, a R134a ice machine model is setup in GT-SUITE. Two parallel 
connected compressors RRT73CIE are used. All other parts and parameters are kept the same as R404A Ice 
Machine 
Figure 6. Schematic of R134a Ice Machine 
20th International Refrigeration and Air Conditioning Conference at Purdue, July 15 - 18, 2024 
5.2 GT-SUITE Model of R134a Ice Machine 
Tr .ate-1 
~ 
Ice ay_ ... . ,
PhTsDiagram 
R1frl1"1 
Figure 7. GT-SUITE R134a Ice Machine Model Schematic 
•• 
60 
40 
35 
30 
25 
20 
,. 
10 
500 1000 
Port Te""eratures 
Tl'lennalMaSS pan: ICeTray_lCe-1 
1500 2000 2500 3000 3500 
- Center 
boHDm 
Figure 8. R134a Ice Machine Evaporator Icetray Temperature 
2101, Page 6 
Water inside ice tray temperature at center and bottom are plotted during whole 3600 seconds running time as 
shown in Figure 8. There are three completed freezing and harvesting cycles in 3600 seconds operation time and the 
fourth freezing cycle is started but not completed yet. 
20th International Refrigeration and Air Conditioning Conference at Purdue, July 15 - 18, 2024 
2101, Page 7 
5.2 Comparison of Simulation Results of R134a Ice Machine 
The comparison between simulation results and test results for R134a ACIM is listed in Table 2. 
Table 2: Comparison between Simulation Results and Test Results for R134a Ice Machine 
Metric
92/70 °F (33.3/21.1 °C) 
(Condenser Turning) 
92/70 °F (33.3/21.1 °C) 
(Superheat Turning) 
Item Test Model Delta Test Model Delta 
Total Cycle Time (min.) 18.05 16.25 -10.0% 18.05 17.63 -2.3% 
Ice Made per 24 hrs. (kg) 191.43 214.24 11.9% 191.43 197.47 3.2% 
Energy per 45.4 kg Ice (kWh) 8.78 6.34 -27.8% 8.78 6.67 -24.0%* 
Energy per 24 hrs. (kWh) 37.05 29.96 -19.1% 37.05 29.04 -21.6%* 
Even with room controlled at 90°F (32.2°C) DB temperature, the condenser inlet air temperature was measured 
around 92°F (33 .3°C). The condenser turning refers to the same tuning being applied during R404A model results 
comparison. The extra tuning applied on R134a model is adjusting the superheat control target to 12°F (6.7°C). The 
difference in total cycle time and ice made per 24 hrs. are within 4%. But energy input difference between model 
and test data are more than 20%. By inspection the simulation data, it was found out that the operation condition 
during freezing cycle is out of the envelope of the compressor used as shown in Figure 9. Thus, lead to an 
underestimation of compressor power compared to test data, even suction and discharge pressure are very close. 
Operating Envelope: RRT73C1E-CAA (HFC-134a) 
• Rating Point 
150 
140 
130 • 
~ 
._ 120 
c:i. 
E 110 i! 
"C 100 C 
0 
(.) 
90 
80 
70 
60 
-20 -10 0 10 20 30 40 50 60 
Evap. Temp. (°F) 
Rating Ref. No.: 07 -1128 
Figure 9. Operating Envelope of RRT73C1E-CAA Compresssor (HFC-134a) 
5. CONCLUSIONS 
GT-SUITE is a powerful dynamic simulation platform that provides extensive toolsets for building an automatic 
commercial ice maker dynamic model. Beyond heat exchanger model and flow management model, finite state 
manager and other control toolsets included with GT-SUITE helps to implement a time sequential operation based 
on real machine operation logic easily and realistically. Lumped parameter approach of thermal mass, thermal block 
or thermal node toolsets provided by GT-SUITE can be employed to build an accurate model that generates good 
prediction results agree well with test data. When implement 10 coefficients polynomial compressor model, the 
prediction results may have a huge discrepancy with test data when operating outside the operation envelope . The 
model will be further validated with test results at other rating conditions and will be used to explore other 
compression technologies such as scroll compressors and other green refrigerants for automatic commercial ice 
maker. 
20th International Refrigeration and Air Conditioning Conference at Purdue, July 15 - 18, 2024 


English

Dynamic Simulation of An Automatic Commercial Ice Maker

Hu, Yifeng

Shen, Bo

Gluesenkamp, Kyle R.

Yana Motta, Samuel F.

Krishnamoorthy, Sreenidhi

Shirey, Don

Experimental Investigation on Heating Performance of A Cold Climate Thermoelectric-Assisted Heat Pump

Jounay, Maelle

Cauret, Odile

Teuillieres, Cedric

Tran, Cong Toan

Methods for Real-time Assessment of Refrigerant Charge in Residential Heat Pumps: Experimental Evaluation in Climatic Chambers

Asztalos, Katherine J.

Ameen, Muhsin

Waikar, Ameya

Rowinski, David

Evaluation of Flow, Heat Transfer, and Phase Change Characteristics in Microchannel Condensers using Computational Fluid Dynamic (CFD) Simulations

Kim, Jeebeom

Lee, Hoseong

Computational Study of Thermal Performance Enhancement in High-energy Density Data Center through Immersive Liquid Cooling

Robinson, Matthew Nicholas

Goedeke, Scott

Maxwell, Taylor

Pellicone, Devin

Blatt, Morton

Beaini, Sara

Development and Evaluation of a Loop Thermosiphon Heat Exchanger as Passive Replacement to Secondary Loop for Low GWP Heat Pumps

Robin, Mark L.

Atmospheric Chemistry and Fate of Selected Refrigerants

Wu, Songhao

Balaraman, Anand

Fix, Andrew

Oh, Jinwoo

Braun, James E.

Warsinger, David M.

Energy-Saving Potential of a Dual-Module Humidity Pump for Residential and Commercial Air Conditioning

Cheekatamarla, Praveen

Sharma, Vishaldeep

Fricke, Brian

Yana Motta, Samuel

Mitigation of Safety and Environmental Challenges Posed by Low and Ultra-low GWP Refrigerants

Yang, Bin

Chretien, Ludovic

Performance Investigation of A Residential Air Source Heat Pump With Partial Variable Speed Electronics Driving a PSC Rotary Compressor As Single Stage, Two Stage And Variable Speed Compressor

Purdue University West Lafayette

Not a member yet