Seeking Validation at Navistar

November 1, 2014

“It’s a dance of a lot of different activities moving together, and depending on the scope of the changes we’ll use a variety of different means to validate the durability,” said Navistar’s Craig Birkett.

As Chief Engineer, Vehicle Engineering at Navistar Inc., Craig Birkett (SAE member, 1983) oversees the entire validation process for Navistar vehicles, including testing and simulation. Having spent many years working on structural, acoustical, and vibration-related issues at both Navistar and Daimler Trucks North America, Birkett has accumulated significant CAE experience in the commercial vehicle industry and recently chaired the “Durability and Life Testing” session at the SAE 2014 Commercial Vehicle Engineering Congress. SAE Off-Highway Engineering Assistant Editor Matthew Monaghan recently spoke with Birkett about the validation process at Navistar; following is an excerpt of their conversation.

How does your role of chief engineer of vehicle engineering fit into the validation process at Navistar?

We have different vehicle engineering chiefs that focus on different product lines. My focus right now is on cost and weight reduction. I look over multiple programs, and we look at how the vehicle will be validated to our requirements in the time frames that we have for developing it, so that at the start of production we have a vehicle that’s adequately validated for its durability, performance, and interior acoustics. With that, we combine simulation, then component testing, fuel testing, and finally proving ground testing for the vehicles when we get our production-level parts on it.

How do you see the results of CAE simulation comparing with physical testing?

There is a close cooperation usually between CAE and testing. Because of the large number of combinations of vehicle that each OEM has, it’s not practical to test physically every single one, so CAE does an initial sorting often and then selected high-volume or worst-case vehicles, there would usually be some testing to confirm the CAE and to level set their predictions, so that we can make judgments or do evaluations with more confidence with CAE on the configurations that can’t be fully tested.

How are you able to integrate customer requests into vehicle builds?

In most cases, we only build trucks when they’ve been ordered by a customer, so it’s almost as if each vehicle is somewhat of a custom build. There are a large variety of different engine ratings to be tuned for different customer needs. These vehicles are tools to do a job, so as such it’s extremely important to customize and give the customer the right tool for his job. There’s a continual variation, where we might have an order for five or 10 trucks that we’ll route through our custom engineering group to a higher-volume product like our highway tractors, which might have several tens of thousands of production a year.

Can you briefly describe a typical validation process?

The life cycle of a program would go first from doing some simulations on the new proposed designs; the role of simulation is as much a design process as a validation process. The CAE team works closely with the designers doing a lot of iterations to come up with a good initial design. At that point, we might have a mule vehicle or more fully developed vehicle that’s more close to our final design, and we’ll do some testing. In that testing, part of it will be a dual process. One will be to validate that iteration, but also we will correlate some of the CAE work and identify if we miss any issues. Those tests might also be used to set up some shaker tests, where because of uncertainty in the material property, a shaker will give us some additional information. At that point, we might do a quick iteration on the design and then put our final validation parts onto a full vehicle and run a full-vehicle test. It’s a dance of a lot of different activities moving together, and depending on the scope of the changes we’ll use a variety of different means to validate the durability.

What types of failures are you looking for during testing?

In a shaker test, we’ll monitor the test closely at the end of each shift and look for cracks. If the test engineer finds any, we immediately will take pictures and get the design engineers and decide whether to continue with the failure or replace the part right away with something else. We’re looking for cracks and complete failures and then we usually will allow some time for some iterations in case there are failures to try out new designs. We can also look at the more in-depth studies on a shaker by adding some strain gauges or load cells and track right away what’s happening with changes in our design, how it reduces the strain or some of the g loads that are acting on a part.

How do supplier-designed parts fit in the validation process?

A big part of our validation is reviewing their validation process and making sure it covers all of our full-vehicle issues and that their testing or CAE adequately addressed that. Suppliers are a big part of our validation work, and we’ll often send a test engineer to their test facility to understand their test better or consult with them on that. We give suppliers requirements of what the part needs to do to meet our full-vehicle requirements and then rely on them to ensure also that their parts do satisfy those requirements.