Resolving NEV testing challenges with a high degree of accuracy, efficiency, and agility

In automotive, with its ever-increasing pace of innovation and electrification, time is of the essence!

Time-to-market and rationalization in the development of automotive powertrains results in an ever-increasing focus on efficient product validation in terms of cost, lead time and environmental footprint. Additionally, more and more e-drives will be part of, or be the only, power unit in the drivetrain architecture of passenger cars, rendering new specific NVH aspects as gear whine and electromagnetic whine from e-machines and inverters increasingly important towards total product performance and end-customer satisfaction. Highly controlled conditions and elimination of masking sounds therefore are critical for the successful testing of EV and hybrid powertrains.

“Highly efficient and affordable electric propulsion systems are essential for the mass adoption of electric vehicles at any level of electrification, including a battery electric vehicle, plug-in hybrid vehicle or fuel cell vehicle. Modularity, while preserving compactness, is of the essence to combine customization towards specific OEM needs with scale.” Punch Powertrain CTO Gaspar Gascon explained.

 
This article describes 3 approaches to increase agility of the testing process, which allow higher efficiency, more reliable results, shorter time to market – while at the same time saving expert’s time, energy and decreasing harmful exhaust gasses. Notably: universal ICE emulation, the NVH chassis dynamometer (NCD) and NVH Driveline Test System (NDTS). The latter 2 offer complimentary solutions to a test track, allowing to zoom in on a specific – to be investigated – aspect of NVH, on either vehicle or transmission level with high efficiency and speed. After the aspect under investigation is resolved, validation can happen in the other direction.

Universal ICE emulation, allowing infinite simulations

The challenge: When performing durability or functional tests on a transmission system level, whether a conventional or a hybridized one, at some point in time will require the availability of the internal combustion engine (ICE) which has to be matched with the transmission. With ICEs in the development phase this results in a few risks and disadvantages: availability of prototype engines can cause delays in lead time or deviating technical specifications from the final engine, maintenance, or even technical problems during durability testing with prototype engines results in longer downtimes. In addition, bigger spreads on dimensional and performance parameters might increase the system test measurement inaccuracy and in addition the fuel consumed increases the environmental footprint.
The solution: At Punch Powertrain a highly agile and modular methodology and test setup has been developed to overcome these challenges. A super low-inertia servomotor (rotor inertia close to 0.1 kg.m²) is used as an input motor to the transmission, which is closed-loop controlled with an advanced ICE simulator model. The model can be either based on a measured engine characteristic or on a scaled generic ICE model, tailored to the needs of the OEM. Even the typical torque oscillations, inherent to ICEs can be simulated up to 600Hz, with an 8kHz closed-loop control, resulting in highly accurate torque measurements. This requires a super low inertia and highly accurate torque measurement, and the set-up is equipped with a data acquisition system, capable of processing the real-time data.
Some of the highly specific features that can be included in the simulation are air and ignition path first order dynamics, engine mount dynamics, inertia compensation emulation for the difference in inertia between ICE and input motor on the test rig, speed lift up requests, configurable starter motor behavior for start-stop functionality and engine stalling behavior.

The result:  Gaining accuracy, speed and sustainability
This progressive set-up brings myriad of advantages to the development process. Firstly, it allows perfect repeatability with no variation between different engines. Secondly, the typical time-to-market can be reduced by several weeks or several months as there is no need for a prototype engine to start the system level tests for the transmission. The motor development can be planned separately from the transmission development. The third important benefit is sustainability, with an energy consumption that is reduced by as much as factor 4, compared to testing with ICEs. In the described set-up the input power of the electromotor is almost completely recovered through the load motors, connected to the output driveshafts of the transmission and acting as the road load, being used in generator mode, and consequently only consuming the internal transmission losses. Additionally, it removes the need for a fuel supply system in the test cell or the need to dispose of exhaust gasses, dissipate waste heat from the ICE or deal with the risks of fuel leaks and vapor.

NVH Chassis Dynamometer (NCD)

The challenge: Subjective noise evaluation (SNE) test on the test track is influenced by background noise and road surface quality, rendering it difficult to make good acoustic measurements and to exactly quantify issues. Moreover, weather conditions and test track availability may hamper additional tests to solve the issue quickly. To tackle these challenges, a NCD can be used. While making use of this installation, further features can be considered to offer an even higher degree of of accuracy, replicability, and reliability – as existing set-up – described below.
The solution: To assess the NVH performance of the entire powertrain inside the vehicle, a 270kW, 75” chassis dynamometer for 2WD vehicles is installed in a hemi-anechoic environment. With vehicle speeds up to 270km/h, a vehicle mass between 700 kg and 3700 kg (2500kg per axle) a wide range of vehicles and tests can be covered with the setup. The test vehicles are equipped with multiple arrays of vibration sensors and microphones inside the engine bay and cabin to measure the acoustic performance, signature vehicle measurements, transfer path analysis, acoustic camera measurements and Operational Deflection Shape (ODS) analysis in all conditions. The chassis dynamometer is currently equipped for FWD vehicles; however, it has the flexibility for AWD installation with minimal efforts.

To eliminate external noise pollution a box-in-box principle is applied, where the full concrete chamber of >140m² is fully decoupled from the building and its surroundings by metal springs, connecting it to a concrete foundation block, with a mass of 150ton and complying with ISO3744:2010. This allows to bring the background noise level in the room below 20dB(A) with a cut-off frequency below 80Hz (ISO10534).
A remote-controlled vehicle cooling fan, with wind speeds up to 100 km/h is present to realistically simulates the engine cool down during road driving. The set-up of the ventilation system is carefully tuned, to minimize the operation sound impact on measuring results to insignificant proportions. A drive robot (providing pedal input) is applied, allowing very precise input, eliminating operator fatigue or slight inaccuracies.
The result: This combination of highly controlled, highly realistic environment, in combination with full control over the input, allows for a very precise definition of testing conditions, as well as their perfect replicability. This allows for quick and accurate iterations to solve specific NVH-related issues.

NVH Driveline Test System (NDTS)

The challenge: detailed analysis and adjustments are further required on the transmission level. This step is highly important in the development of hybrid and fully electric drives: allowing the elimination of masking sounds, while precisely controlling the inputs of any driving situation simulation. To tackle this challenge a unique set-up was developed in close collaboration with AVL, including proprietary processes and software. Throughout the development of the NDTS special attention went to highly relevant parameters, such as: reaching a rotational inertia as low as an ICE for the input motor to be able to simulate the highly dynamic behavior and torque oscillations up to 700Nm and 300 Hz of the ICE;  as well as a hemi-anechoic chamber setup to reach ISO 3744:2010 standards, resulting in – for example – a background noise level inside the test room of less than  20dB(A) and a cut-off frequency at 80Hz (ISO10534).
The solution: with the help of above-mentioned emulation software, an electric input motor (700 Nm, 10 000 rpm and 400 kW) that simulates the ICE is located inside the hemi-anechoic test cell. The 2 output motors (3232 Nm, 3000 rpm and 220 kW each), which represent the wheel loads, are positioned outside the test cell, connected with custom built carbon fiber shafts of >2,8m in length to the output shafts of the transmission.
This setup can be used with conventional and hybridized transmissions for signature NVH measurements, sound power measurements, torsional vibration measurements, endurance degradation measurements etc. from an early development phase onwards, allowing quick anticipation and design iterations to improve the NVH performance of the transmission. To run tests with the e-machines of MHEV or PHEV applications, a sophisticated DC source is available which allows the use of battery models to simulate parameters as state of charge (SoC), battery degradation, … in a very repeatable and safe way and can be finetuned to specific customer applications or technologies for battery packs. With a little re-build, the set-up can be expanded to EV testing.
The results: taking the transmission out of the vehicle and placing it on the highly controlled NDTS environment, enables replication of any scenario encountered in the vehicle (test track or chassis dyno) on a transmission level, further increasing flexibility and agility. In addition to NCD readings, in this set-up any masking noises from the car are eliminated and the same inputs (throttle, brake, gear lever position) can be used to perfectly replicate any driver action.  Solution iteration speed is greatly improved as build-over times are reduced from 4-8 hours on a vehicle to only 1-2 hours on the NDTS. Finally, as the driver inputs are parametrized, it is possible to perform test scenarios more accurately than with a test driver (e.g., take-off with throttle at a constant 30 %).

At Punch Powertrain a 1100 square meters state-of-the-art new Test Center has been developed and installed in 2019 in line with its ambition to deliver this extensive expertise in testing as an added value and support for our customers and the move towards e-mobility. In this process, the company forges partnerships with testing experts, as well as welcoming customers and partners for the use of its facilities.
At Punch Powertrain we have intensively invested over the course of past years to launch pioneering approaches to bring agile and scalable solutions to the market, in everything we do: our products, our industrialization process and our testing. Specifically, today, we will focus on our testing practices, which lift the bar of automotive benchmarks. “Many innovations are born from our passion to deliver the best result we can envisage, helping the customer in solving their challenge.” Punch Powertrain CTO Gaspar Gascon concluded.
Geert Jans –  Corporate Test Center Leader, Kristof Marcelis – Testing Hardware & Software Development Leader, Pepijn Peeters – Test Rig Engineer all from Punch Powertrain contributed to this presentation.