Home   |    International   |    Contatti   |    Info   |   

New 1D simulation developments to meet current and next car generation engine cooling challenges

fcaFig. 1 - Most of terrestrial vehicle engine cooling system components and test
facility technologies were available at the beginning of 20th century, but modern
vehicle capabilities require better parts and methods (image authors: Atelier
Mempli, Fong )

Current car generation offers increased comfort and capabilities, but customer expectations about fuel economy and environmental concerns drive automotive industry to improve over and over. During the last decade thermal engineers have been focusing their attention on engine cooling systems, introducing new parts, new functions and devoting more resources to proving ground or real world validation sessions. As a consequence, vehicle level thermal, hydraulic and aerodynamic simulations have to evolve alongside improved testing.
This paper explores a possible approach to model current and nearly future car engine cooling systems based upon Flowmaster code, transient runs but basic vehicle and engine usage estimation. Final goal is getting a quick but reliable method for system sizing, alternate components checks, assessment according different environmental conditions or modified driving patterns and vehicle usage.

Altri case studies sullo stesso argomento

Fluidodinamica 1D
Altri case studies sullo stesso argomento

Calendario 2018
Corsi di formazione sulle tecnologie software

Formazione personalizzata
Percorsi formativi personalizzati e traning-on-the-job: contattaci!

Workshop e seminari
Visita la pagina degli eventi: potrai partecipare alle iniziative EnginSoft ed essere informato sui più importanti appuntamenti europei

Newsletter EnginSoft
Leggi l'ultimo numero ed Abbonati alla rivista!


Chiedi all'esperto!
Articolo completo


Fig. 2 - A Flowmaster 1D thermohydrauilic simulation model
built by KISS approach to investigate dynamic engine
cooling system performances includes radiator module layout

Old dogs, new tricks
Internal combustion engines generate heat to deliver mechanical power but excess heat has to be removed to prevent failures. Nowadays, nearly all passenger cars are equipped with liquid based cooling systems, whose basic design is simple and efficient yet one century old: a water and antifreeze mixture removes heat from the engine and delivers it to a “radiator” that discharges it into the environment. Keeping the powertrain cool is the primary function of car cooling systems and driving reliable cars was fair enough for our grand-grand fathers and mothers: but the larger the car market, the more demanding the customers so in order to increase passenger thermal comfort both a coolant pump and a thermostatic valve sensitive to coolant temperature were introduced. The pump delivers hot coolant whatever temperature or engine load or revs, the valve blocks cold coolant flow to the radiator, transforming the engine cooling system into an heat source to keep warm passengers cabin or defrost windscreen: a real safety improvement in cold weathers.
The secondary function of engine cooling system as a thermal power sink raised interest of car makers when more challenging vehicle pollutant emission factors were imposed. If you can prevent coolant removing heat from the a cold engine or the gear box for a while after cranking up you can make engine metal and engine oil warm up faster and achieve a better fuel economy by reduced mechanical losses. In the last decade supporting vehicle fuel economy improvement became the tertiary function of engine cooling system, but it came with a price: it has to be assessed running dynamic low to mid vehicle speed and engine loads, too far from severe driving conditions required to size radiators and fans.


Fig. 3 - All these vehicle share the same baseline engine,
b ut their lap times and thermal results on track would be
very different. Fia Formula F4 courtesy of Tatuus Race Car

A larger number of check and release sessions to grant both vehicle reliability and efficiency could impact new vehicle project timing and costs, but increased availability of affordable service facilities and gentlemen driving club circuits makes possible parallel testing by hot cell, proving ground and public roads runs, getting the best of each one.
In a single day you can cruise in the countryside and driving downtown in rush hours experiencing a real customer vehicle usage to calibrate engine cooling system software, then tow uphill or running power laps on a track to assess engine cooling system hardware. Hot cells offer tunable environment conditions and mimic different road, vehicle and engine set up in a single location and are not restricted by weather and climate.
But in order to achieve the correct balance of test methods engine cooling performance simulations have to be improved too, extending their capabilities from system sizing and alternate components/subsystem checks to dynamic vehicle actions analysis – to prevent issues, provide guidance and allowing test result interpretation.


Fig. 4 - A collection of small, single purpose steady state
simulation models assess reliability of main model features
and heat maps reducing tuning loops and saving time

What engine cooling simulation?
If passenger car engine cooling system simulations have to be more dynamic it means that they can be considered as a subcase of transient vehicle thermal management checks. Therefore Flowmaster 1D CFD environment is appropriate to set up new capabilities and verify model building issues, being worldwide used by the automotive industry and proven effective for hydraulic and heat transfer studies. Compared to simulations for thermal management 1D CFD for engine cooling deals with higher flows and heat load, while vehicle speed profiles have to be estimated rather than imposed : compared with sizing simulations it’s much more time consuming and labor intensive. In order to meet both reliability, compliance with project schedule and costs the “Keep It Simple, Stupid” or KISS approach is effective, so extended capabilities engine cooling 1D CFD models were built according the following guidelines:

  • cooling airflow branch and each coolant branch sub-networks reduce all non-component related pressure losses to a single, overall discrete loss element;
  • fan(s) and valve(s) driven by Flowmaster controllers, not by equations;
  • simple lumped mass engine and thermal bridge loop to deliver heat rejections;
  • Re-Nu heat map for air cooled heat exchangers extended for zero flows;
  • vehicle speed, engine revs and heat rejection computed according vehicle and engine data outside Flowmaster environment, then included as time based inputs;
  • exact engine cooling module layout to get component interactions.


Fig. 5 - Release hot cell thermal tests results
compared with simulation results

What for engine cooling simulation?
Considering functions of automotive engine cooling systems it seems reasonable investigating new simulation models for hot cell and track tests first. They are used to release engine cooling systems and set up targets and vehicle specifications by investigation of similar or similarly powered vehicles. Compared to real road testing both hot cells and tracks check a smaller range of driving conditions, but these ones are more challenging being unrestricted by traffic rules: additionally, they reduce safety issues, prevent security and press leaks about new models and allow quick repairs or tuning by their own facilities.
Usually it’s not difficult set up boundary conditions for 1D thermal models simulating hot cell testing but 1D thermal models simulating track runs need more efforts. If onboard recorded data are unavailable boundary condition should be set according expensive and time consuming full scale vehicle dynamics studies, but KISS approach and bibliography show a simplified yet effective approach under the following guidelines:

  • vehicle is reduced to a slimy, lumped mass stick to the ground by tire grip only;
  • vehicle does not accelerate or brake running bends;
  • vehicle cornering speed is set according maximum tire grip and bend radius;
  • vehicle accelerates and brakes running straights;
  • vehicle acceleration is limited by tire grip, even if engine traction would be higher;
  • vehicle brakes to match target cornering speed, using maximum tire grip.
  • Block based scripts achieve a 4% to 12% maximum error estimating lap times by these simple rules, provided gear shift schedule, vehicle drag and engine power are available.


Fig. 6 - Predicted and recorded engine coolant temperature
during a track test

Test case, simulation build up and tuning
An European market oriented, mid size car equipped with a high power output gasoline engine an a classical cooling system is used as a reference to develop 1D CFD models for hot cells and track test studies. Engine coolant components were set up considering nominal performances and heat rejection maps was nominal too, but extended to cover zero power and zero engine speed conditions to prevent solver numerical issues. Aside main simulation model a certain number of simpler auxiliary models were build and used to check out reliability of heat exchanger maps or provide “equivalent” pressure drops for coolant and cooling airflow branches. Then main model steady state, cold flow rate estimations results were compared with test bench and CFD flow results and despite a fair to good agreement was achieved appropriate corrections were implemented: as usually happens the largest deviations were fond for low vehicle speed cooling airflow estimations.

Once assessed flow reliability predictions every 1D CFD engine cooling model set up for transient studies should be thermally tuned and the simpler steady state hot cell tests results, if available, are well suited for this purpose. These tests produce a slow temperature evolution inside the engine cooling system, allowing refinement of fluid volumes and other non-fluid related thermal masses imposed inside components: additionally, steady state hot cell test results they can be used to double check engine nominal heat rejection map.
A test performed turning off and then turning on passenger cabin AC system, possibly including various vehicle speed, serves a similar purpose assessing reliability of predicted AC system heat rejection to cooling airflow upwind main radiator . This parameter is very important to achieve a good simulation precision for low vehicle speed cooling tests.


Fig. 7 - Some example of “what…if” comparison model runs: track
simulations are performed modifying vehicle, transmission, tire or running
a different track at all

Comparison of simulation and test results
After tuning 1D CFD engine cooling model was used to estimate temperatures running both hot cell and track tests oriented to final vehicle release. Predicted and recorded hot cell test results are shown in figure 5, being very close each other for the simpler test on the left. The more complex, three stage hot cell test on the right shows a fair enough prediction, the largest difference with experimental data does not exceed 6%. Simulation detects all events occurred, including thermostatic valve opening, transition between test stages and the fan cycling from full power to half power occurred during the final, most challenging stage.

Similar results are achieved running track simulations, as show in figure 6. This 15 minute long test is simulated using as boundary conditions reported vehicle speed and engine revs, calculated heat rejections and model calculated coolant and cooling airflows. Results are again fair enough, largest difference are found during event transitions and final hot soak.

Predictive mode for new vehicles
1D CFD upgraded models can anticipate hot cell test results for a new vehicle because thermal network takes care of different engine cooling components, while effects of different specifications such as gear ratios improving fuel economy or increased towing capacity to meet customer expectations are accounted for as new boundary conditions. According preliminary evaluations track test results can be anticipated too, as long as modified engine, vehicle, track parameters are handled by block based vehicle performance simulations. Figure 7 shows comparison of coolant temperature prediction according:
(a) tests performed with AC system engaged and disengaged;
(b) tests performed running an European and a North American track;
(c) alternate and standard gear upshift schedule;
(d) nominal and reduce tire grip due to dust or rain on the track.


Fig. 8 - Introduction of grade effects extends engine
cooling system dynamic simulations

Summary. But not the end
For an increase in complexity, building time and labor 1D CFD Flowmaster full transient models improve evaluations of engine cooling system dynamic performances.
However, usage so far suggests a certain number of improvements and developments:

  • investigations of all classical engine cooling module layouts;
  • investigations of all dual loop engine cooling systems variants;
  • higher level control strategies for electrical coolant pumps and valves;
  • better control strategies for active grill shutter devices;
  • feasibility studies to include calculation of vehicle drag power and engine or transmission heat rejection inside Flowmaster models;
  • elevation profiles for vehicle performance evaluation to take care of tracks such as Imola, Spa and Nordschleife and extend dynamic analysis to mountain road testing.

Articolo pubblicato sulla Newsletter EnginSoft Anno 11 n°4

N. Cauda, M. Colantoni, G. Gotta
FCA Research and Development, Engine Systems

EnginSoft SpA | © All rights reserved | VAT nb IT00599320223
Questo sito utilizza cookie tecnici e cookie analytics, che raccolgono dati in forma aggregata | Informativa completa sulla privacy e sui cookie