Industry Automotive
Automotive

3D printed intake manifolds made in Windform XT for VM Motori successfully performed the dyno tests

3D printed intake manifold functional prototypes were manufactured by CRP Technology for engine manufacturer VM Motori to improve EGR system. The components, in Windform XT, were used for dyno tests with excellent results.

3D printed intake manifolds made in Windform XT for VM Motori successfully performed the dyno tests
In Short

Application

3D printed intake manifold for testing turbo diesel engine

Solution

Selective Laser Sintering, Windform XT Carbon fiber filled composite material

Challenge

Manufacture 3D printed, functional intake manifold prototypes for dyno tests, thermal and mechanical stress resistant, shortening the standard lead-time

Features of the application

Field tested required
Functional component
High strength
Lightweight
Performance like conventional process-made parts
Resistant to continuous stress accelerations
Shock resistant
Stiffness
Withstand high strain mechanical stresses

Customer's requests

Highly customizable
Maintaining high-end user quality standards
Minimized production speed
No tooling cost
Reduced time to market
Remarkable accuracy

Services supplied

Complete engineering assistance
Suggestions for overcoming critical application consultancy
Technology and material selection
Testing

3D printed intake manifolds made in Windform XT for VM Motori successfully performed the dyno tests

How 3D printed intake manifold increases the engine performances

3D printed intake manifold enhances the performance of engine.
The modern Turbo Diesel engines have increased their performances compared with “old” Diesel engines. Unfortunately, new severe requirements regarding their emission have to be faced.

The use of EGR (Exhaust Gas Recirculation) can help engine manufacturers to keep under control the emissions to accomplish to the Euro 4 requirements.

In order to improve EGR system VM Motori S.p.A. (Cento, FE – ITALY) decided to test different alternative intake manifolds. Instead of taking a long projecting and engineering time, VM engineers’ good knowledge of Additive Manufacturing techniques allowed to spend only a few weeks to reach the best performance of the manifold and the engine.

Intake manifold, CAD design
Intake manifold, CAD design

The study was performed using 3 different models, 3D printed by CRP Technology in Windform XT, a Carbon fiber filled material from the Windform range of composites for Laser Sintering systems. Windform XT is now replaced by Windform XT 2.0.

Each prototype was prepared with a typical lead-time of 2/3 days. This kind of “fast-production” was very useful for VM engineers, shortening 10 times or more the standard lead-time for this kind of product.

The 1st model was called “V1”, and it was used for dyno tests without any issues.

Then the VM engineers worked on other two releases, “V2” and “V3”, to optimize the partitioning of EGR on each cylinder, reaching the goal of a maximum of 4% of difference among the 4 cylinders (as a target for Euro 4 emission level).

Intake manifold
Intake manifold

Usually “standard” lead-time for a sand cast aluminium intake manifold is about 2 months. In this case they made 3 different manifolds spending only 6/9 working days. So the engineering phase that usually takes several months, lasted less than 2 weeks.

In the following table we can see some interesting data to give an idea about thermal and mechanical stress, applied to the manifold, during normal engine use (real life engine operating points where the EGR is activated so the intake temperature is increased by mixing fresh air and hot exhaust gases).

Mode
Air Mass kg/h
EGR Rate %
T. Manifold °C
T H2O out °C
T H2O in °C
Average:
67,37
83,56
81,29
1
21
57,00%
65,80
79,20
78,70
2
56
36,00%
60,70
82,90
80,90
3
71
44,00%
60,00
81,70
80,80
4
87
22,00%
44,40
82,80
80,30
5
94
34,00%
50,40
82,60
81,30
6
105
20,00%
49,70
83,50
80,80
7
105
27,00%
71,00
84,00
81,10
8
115
26,00%
47,20
82,60
81,30
9
48
49,00%
73,50
82,40
80,90
10
123
21,70%
64,50
84,20
81,10
11
95
27,50%
60,30
83,60
81,10
12
71
34,50%
73,80
83,30
80,40
13
148
23,00%
85,40
85,40
82,00
14
176
27,00%
72,70
84,80
82,30
15
280
19,00%
91,70
87,10
83,80
16
275
20,00%
106,80
86,90
83,80
Mode
Air Mass kg/h
EGR Rate %
T. Manifold °C
Pressure Intake Manifold BAR
Max. Output 110 KW (150 Cv) @ 4000 Rpm
550
0,00%
55,00
1,50

In the tables we can see also some values about H2O temperature, because there’s coolant fluid flowing in a duct beside the manifold, increasing its temperature (because of reduced heat exchange trough the wall) but without affecting its functionality at all.

VM engineers have done several tests, without any kind of problem. The manifold has supported also some tests lasting 2,5 hours, with the engine at the maximum output, and at the end, it was still perfect, ready to be used again.

For the production of the new Turbo Diesel Common Rail 2000 cc engine, designers decided to use aluminium manifolds and when they tested the first engines, they observed exactly the same results of Windform XT manifolds.

It demonstrates how reliable and useful can be a study done in an incredible short time, thanks to Windform XT properties.

Sidebar: the new Chevrolet Captiva will be powered by this engine

From Chevrolet official web-site: “From the manufacturers of the very first SUV back in 1935, Captiva is the latest in a long line of authentic sport utility vehicles offering great value for money in a stylish and practical package. Designed specifically for the European market, it will be available with front or all-wheel drive, five or seven seats and with a 2.4 litre petrol or leading-edge 2.0 litre diesel engine and will appeal to a wide range of customers. Captiva will be on sale in Europe from spring 2006.”

Engine/powertrain:
RA420DH6
Type / 4 cyl/in line, common rail diesel
16 Valves
Displacement / 2,000 cc
Max. output / 110 kW/150 PS at 4,000 rpm
Euro 4