Refrigeration demands for meat processing A.K. Fleming

Demandes de froid pour la transformation de la viande

L'industrie d'exportation de la viande en Nouvelle- Z#lande est la plus forte utilisatrice de froid, i l faut

donc savoir avec quelle efficacit~ elle utilise I'~nergie,

On d#crit le materiel de grandes usines de transformation de la viande et I'on analyse thermodynamiquement le fonctionnement de rinstallation. On a obtenu ainsi le nombre de fa~ons permettant des ameliorations substantielles du rendement et des ~conomies d'#nergie.

The meat exporting industry in New Zealand is the largest user of refrigeration, and thus it is necessary to know how efficient it is in the use of energy. The equipment of a large meat works is described,

A recent survey of refrigerated facilities in New Zealand 1 established that the frozen-meat export industry was byfar the largest user of refrigeration. The industry has a gross cold storage volume of 1.4 mil l ion cubic metres, two-thirds of the national total, for meat production of about one mil l ion tonnes annually. It is important that the demands for meat refrigeration should be well documented, to permit rational design of equipment and economies in usage.

This paper summarizes the results of a thermo dynamic evaluation of a large meat works 2. The fol lowing recent developments made this survey part icularly opportune:

More intensive chi l l ing and freezing of meat is being introduced along with greater use of air condi t ioning and lower cold storage temperatures, all of which require additional refrigeration capacity.

Increased energy demands brought about by these changes caused concern and prompted a search for economies, which are now particularly relevant due to rapid energy cost increases.

The objectives of this work were to determine specific refrigeration demands for various meat processing operations, to predict plant performance (refrigeration produced, energy consumed, condensing temperature and coefficient of performance) for given meat product ions and ambient temperatures, to find the true cost of producing refrigeration and to assess possible economies in refrigeration demand and production.

Exper imenta l The meat works surveyed had a daily design throughput of 1 000 cattle, 10 000 lambs and 800 pigs, and a total annual meat production of 33 000

The author is at the Meat Industry Research Institute of New Zealand (INC), PO Box 617, Hamilton, NewZealand. Paper received 27 February 1978.

and the operation of the plant has been analysed thermodynamically. This indicated a number of ways by which substantial improvements in efficiency, and saving of energy, can be attained.

tonnes. The ammonia refrigeration system was large and complicated, having grown with the plant over some 60 years. Two-stage compression was used, with three rotary-vane compressors (total displacement 1.83 m 3 s -~) for Iow temperature refrigeration and eight high-stage compressors (six reciprocating, two screw) with a displacement of 3.1 2 m 3 s -1 . All were electrically driven, with a total connected load of 2 230 kW.

The nominal refrigerating capacity was 3.7 MW at a condensing temperature of 32°C. Ten shell-and- tube condensers were used, with condenser water cooled through crossf lowand counterf low cooling towers and two spray ponds.

All the evaporators were operated in a flooded condit ion using pump-circulated ammonia from five l iquid-vapour separators servicing chillers, air condit ioned areas, freezers (-27°C and -37°C systems) and cold stores.

Extensive thermodynamic surveys were made over one-week periods in winter and summer, corresponding to minimum and maximum refrigeration loads. Information was obtained from plant records for the intervening period. Turbine meters were installed at twelve locations to measure liquid ammonia flow. These meters proved reliable for up to a year, although failures occurred if the liquid ammonia supplies became exhausted and high velocity gas caused over-revving of the impeller. As a check on the ammonia mass flow, and to establish compressor efficiencies, the gas flows were measured in compressor suction lines using a pitot-static tube and an inclined manometer 3. Other instrumentation installed for the survey included a data logger (to monitor temperatures), kWh meters, event recorders and pressure transducers.

Results

Mass balances between metered gas and liquid agreed to within an average of 6%. The demand for

Volume 1 Number 4 November 1978 0 1 4 0 - 7 0 0 7 / 7 8 / 0 4 0 2 1 7 - 0 5 SO2.00 © 1978 IPC Business Press Ltd. and IIR

2 1 7

Plont evaporating temperalures Chilling and

-10

-20

Refr~aerotion effect 1 h average

£ 4

r tompressor power

0 I I ] I I 1 2 3 4 5

Doys

Fig 1 Compressor power demand, total refrigeration effect, and smoothed evaporating temperatures during peak load conditions in summer Fig. 1 Demande d'energ/e du compresseur, effet fr/gor/hque tota/ et temp#ratures d'#vaporat/on arrond/es au eours des condit ions de charge de po/nte est/va/es

2

4

4 8 tiiiilittttttt

12 16 20 Time of doy. h

Fig. 2 Component refrigeration loads on an hourly basis for maximum daily load in summer.•air conditioning (high temperature Ioad),ra chilling (high temperature Ioad).•cold storage (low temperature load), E] freezing (low temperature load) Fig. 2 Composants des besoms de fro/d sur /a base d'une heure pour la charge journahere max/male en et#. • cond~t/onnement d'alr (charge a temperature elevee),~refr/gerat/on (charge temperature e/evee),•entreposage fr/gorff/que (charge & basse temperature), [] cong#lation (charge b basse temperature)

16 • Tolol produciion

ToloI r ; ~ n g e r ~

Y ~ Frozen meet •

I I I I

Weekly meet product ion. 100 Tonne

Fig. 3 Correlations between weekly meat and offal production and refrigeration produced Fig. 3 Corr#lations entre la product ion hebdomadalre de v/ande et d'abats et le fro/d produ/t

12

8 z

_s ~4

refr igeration was characterized by high peak loads, and in summer, when the refrigeration demand exceeded the compressor capacity, evaporating pressures and temperatures rose (Fig. 1 ).

In Fig. 2, the highest daily load recorded (306 G J) has been broken down to illustrate the component demands. The changing ratio of low to high temperature refrigeration is of interest. During working hours, high temperature loads (air condi t ioning, chil l ing) predominate, while the low temperature loads (freezing, cold storage) peak at night. This changing demand ratio is characterist ic of a meat works and requires flexibil i ty in compressor operating condit ions. A detailed analysis of the individual loads is given in Table 1.

Data from thirteen weeks, taken at random from a year's operation, were used to supplement the survey results. Accurate knowledge of compressor characterist ics enabled the refr igeration demand to be calculated from the plant log book, and this was correlated with meat product ion (Fig. 3). Total meat product ion averaged 635 tonne per week, of which 73% was frozen. For the purposes of this analysis, chi l led product ion was taken as all beef and veal (58% of total production), pigs (8%), and o~qe-third of the sheep and lambs (25%). Frozen product ion was taken as all export meat plus all offal products (9% of total production). Average specif ic refr igeration consumpt ions were:

Chil led p roduc t i on - high temperature refr igeration (-13°C) 740 kJ kg -1

Frozen p r o d u c t i o n - I o w t e m p e r a t u r e refr igeration (-32°C)1 230 kJ kg -1

Total p roduc t i on - total refr igeration (-25°C average) 1 340 kJ kg -1

C o m p r e s s o r s Evaluations of individual plant items were included in the survey, but on ly the results for compressor tests are given here, In Fig. 4, both the volumetr ic and the overall eff iciencies are shown, the latter being defined as the ideal

tO0 41

"7 --o

80- r

:5::

60- Q--

~ 4o

~e 2 0 ]

/

0 Looding lO0°& = Type Compound Displacemeqt,O" 47 m3 S -1

c..,a

-.t ~ 0 3

ob m ~

O -,,!--4 --LO rOO~

::~::il ! .... 75°Io 50"/o

Screw 0.52

r~

.....=

:::::!

iii il iiiii] !:!3

:2::I "i00O/o 5o% 10o%

I 100%

:i:~:

100% 8 Cylinder vee-b~ock Rotory vane

0-34 1"02 0.40

Fig. 4 Measured efficiencies of compressors. The pressure ratio is g=ven at the top of each histogram bar.l-lvolumetric efficiency,• overall efficiency (including motor)

Fig. 4 Rendements mesures des compresseurs. L e taux de compression est donn# en haut de chaque barre du d/agramme. [ ] rendement vo/umetrique, [ ] rendement total (y compris ce/u/ du moteur)

218 International Journal of Refrigeration

100

power required for isentropic compression divided by the actual input power to the motor.

The poor part-load performance of screw compressors has been of some concern. Performance claims by some manufacturers have not been substantiated in practice and it is disappointing to find large screw compressors (c. 1 000 kW) replacing older more efficient reciprocating compressors on the basis of supposedly efficient operation, and operating down to 10% of their capacity. Screw compressors perform satisfactorily at full load, with overall efficiencies of about 55-60% at reasonable pressure ratios (4:1); but when fully unloaded, they

80

60

40

20

0 20 40 60 80 100 % Full load capacity

Fig. 5 Part-load performance of oil- injected screw compressors

Fig. 5 Performance a charge partiel/e des compresseurs b ws reJect/on d'huffe

2'0

o 1"5

5_ E 8 1.0 u

0"5 IJA

• 0 o •

f ~ compressors •

I I I I I ] o 2 4 6

Meet production, 1000 tonne

Fig. 6 Correlations between monthly meat product ion and electricity consumption for the whole works and the ammonia compressors

Fig. 6 Correlations entre/a product ion mensue//e de wande et /a consommat~on d#/ectmcite pour/ 'ensemble des usines et des compresseurs ~ atom•mac

Condensing temperature, °C

20 22 24 26 28 30 32 34 36 ! I i I i I ~ I I

" ~ High lerrc~ralure (-13°C) \ ~ "~ r i ge ra t i on re f r i ge ra t i on \

4 ~ o o \ 2 '

! ~- 3 " " ' , ~ ver age ( - 25~£ ) ~ ,

_ + 2

2 + x

t I I ~ + T ' - I 9 10 11 12 13

Condensing pressure, 100 kPo

Fig. 7 Correlation between condensing pressure and COP (weekly averages). The dashed lines ( 1 ', 2', 3') refer to the respective ideal coefficients of performance for the ammonia cycle

Fig. 7 Correlations entre la pression de condensation et /e coefflc~ent de performance (moyennes hebdomadalres). Les hgnes en pomti l le ( I ~ 2 ~ 3 ") correspondent aux coefhclents de performance/d#aux respectffs pour /e cycle a atom•mac

30

25

20

m _ _ i i i i " . . . . I 2 3 4

Refrigeralion effect , MW

Fig. 8 The relationship between refrigeration effect, ambient wet bulb temperatures and condensing temperatures/COP (shown in brackets)

Fig. 8 Relation entre I'effet fmgorifique, les temperatures ambiantes de bu/be humide et /es temperatures de condensation/coeff icient de performance (entre parentheses)

Volume 1 Number 4 November 1978 219

Table 1. Average daily components of refrigeration load under peak summer conditions

Tableau 1. Composants journaliers moyens des besoins de froid dans des conditions estivales de pointe

Air conditioning Ammonia - 14°C to secondary refrigerant (glycol) -5°C

Cutting room. 10°C,47 men, 1 500 m a

Cutting room. 10°C,124 men, 1 500 m 3

Cannery, 5-15°C, 5 000 m 3

Other. 4 500 m 3

Chilling Ammonia -14°C

Beef, 2 600 m 3

Other

Freezing Boneless beef, 3 tunnel freezers

Offal, -37°C ammonia

Plate frozen offal, -31°C

Lamb carcasses, -31°C ammonia. 5 200 m a

Cold storage Ammonia -31°C

Bulk-stow lamb carcasses, 20 000 m 3

Palletized beef cartons, 7 500 m 3

Entire works

Load, Production, Load kJ refrigera- Recirculation GJ tonne factor* tion per kg ratio

meat

14 29 480 0.28

23 91 250

23 18 0.49 1280

18 - - - -

41 166 250 0.38

9 37 240

43 107 400 0.41

7 16 440

1 4 250 0.27

37 46 800

NA

2.7

8.7

44 NA NA 1.0 0.30

10 NA NA 2.2

270 - - 0.63 - - - -

* Load factor = (24-hour average) / (hourly peak) NA = Not applicable

may consume 50% or more of thei r ful l load power, as shown in Fig. 5.

Electricity Consumpt/on Up to 70% of the e lect r ic i ty used by the works was associated wi th the ref r igerat ion system. Half of this was used by the compressors, the remainder by fans (especia l ly in freezers and cool ing towers), pumps (ammonia, g lycol , condenser, water), l ights and heaters. Most pumps and fans ran cont inuously , i r respect ive of meat product ion, and consequent ly their power consumpt ion was comparable wi th that of the compressors even though the connected load was lower. The l ines in Fig. 6 have been extrapolated to give power demands at zero meat product ion. These intercepts agreed with the e lect r ic i ty usage recorded dur ing a three-week c losure of the works.

Condensing temperature and coefhcJent of performance (COP)The coef f ic ient of per formance was found to be a funct ion of the condens ing

pressure, which was in turn dependent on the ambien t wet bulb temperature and the total re f r igerat ion effect developed by the plant (Figs 7 and 8). These data were obta inab le because the weekly average evaporat ing temperatures were constant to + 3°C and because the condenser water was rec i rculated through cool ing towers.

The annual average COP for the whole plant was est imated to be 2.3, at a mean evaporat ing temperature of -25°C. The ideal COPwas 3.9, g iv ing an overal l e f f ic iency of 59%.

UtEizabon of refrigeration Fig. 9 shows the est imated annual d is t r ibu t ion of the re f r igerat ion demands. 'Losses' in ch i l l ing and freezing were heat inputs f rom sources other than meat: air blast fans, humid i t y reheat control , air f i l t ra t ion etc.

Cost of producing refrigeration Fig. 1 0 shows the cost of producing ref r igerat ion at the eng ine- room

220 In ternat ional Journal of Refr igerat ion

Fig. 9 P r o p o r t i o n s of c o m p o n e n t loads on an annua l basis

Fig. 9 Proportions des charges partlelles sur une base comrnerclale

for New Zealand condit ions. For electr icity at 1.2 cents per kWh, the relative cost components were: annual capital charges on plant (20 year life, 1 0% discount rate) 26%, labour 36%, electr ic i ty30%, repairs, maintenance, insurance 8%. The figures are for a two-stage plant producing 34.6 TJ of refr igerat ion a t -35°C annually, and 31.7 TJ at-1 5°C.

Conclusions

Since only 25% of the total refr igeration produced was used to cool meat, there appear to be considerable opportuni t ies to effect economies in refr igerat ion and consequent electr ici ty demands.

Freezing produced the highest demand on the system and heat input from the fans accounted for nearly half of this heat input load. The fan power in the three carton tunnels freezers alone was 260 kW, and these were left on almost cont inuously. Fans could be turned off when not required and a rational evaluation of the merits of high air-blast speeds is warranted.

In a i r -condi t ioned areas, reduct ion of heat loads could be achieved by minimizing air infi ltration, using local extract ion fans for hot humid air, and avoiding spil lage of waste hot water. A low sensible to total heat ratio (0.45) in the meat cutt ing rooms indicated high heat loads from these sources.

In cold stores, losses from open doors accounted for 70% (400 kW) of the load in older, mult i-storied blocks, compared to only 20% (23 kW) of the load in a modern store with mechanical ly operated

u

4 [, Electricity cost,

0 I [ I L 0 -40 -30 -20 -10 0 1

Evaporating temper~ure, °C

Fig 10 Re f r i ge ra t i on cos t a s a f u n c t i o n o f e v a p o r a t i n g t e m p e r a t u r e and e lec t r i c i t y cos t

Fig. 10 Co#t du froid en fonction de la temperature d'#vaporation et du coCJt de I~lectrlcit~

doors. Automat ic door closers, air curtains or air locks all offer potential savings here.

There could be a reduction of the temperature range over which the refr igeration cycle operates, by ensuring adequate water f low at condensers and cool ing towers, keeping the condenser tubes clean and purging air from the system, and by avoiding excessively high mean temperature di f ferences at evaporators. (The survey average was 1 6°C).

Better control by plant operators could be introduced.

An index of performance (as in Fig. 6) or a COP could be established and one could then strive to improve it.

References

1 Fleming, A.K. Bu// l lR56(1976) 1, 19-27 2 Fleming, A.K. Survey of an ammonia refrigeration plant at

a meat works. Meat Ind Res Inst of NewZealand 432 (1975)

3 Fleming, A.K. Flow measurements in industrial refrigeration systems. Annexe 2 Bull IIR (1976) 81-85 Washington

Dr A. Keith Fleming is Assistant Director of the Meat Industry Research Institute of New Zealand (inc.). He joined the Institute

in 1963 after graduating with a B.E (Chem.) and was involved with various aspects of meat and by- products processing before specializing in refrigeration. During 1968-1971 he studied for a Dr Ing. degree

.at Trondheim, Norway and was appointed to his present position in 1977. He is a member of the New Zealand Institution of Engineers, and is a member of Commission B1 and a vice-president of Commission D2 of the IIR.

Volume 1 Number 4 November 1978 2 2 1