Zero Twist Feeder by IRO

Get our


By Brecht Vermeulen, Project Manager at Vintecc, Creator of Smarter Machines

Our competences

  • Matlab/Simulink Modeling
  • Digital Twin
  • Rapid Prototyping

By introducing Model based design and codegeneration, the path from idea, to proof of concept, to market introduction became significantly shorter.

Brecht Vermeulen

2min read

IRO offers a comprehensive range of weft feeders and auxilliary equipment for Rapier, Projectile, Air and Water jet weaving machines. The company is located in Sweden and is part of the Vandewiele group.

IRO Zero Twist Feeder

The Zero Twist Feeder is an active controlled system to feed flat tapes and yarns without twist to a weaving loom.


When weaving with a classic feeder, the yarn is being pulled from the bobbin in the axial direction of the bobbin and buffered on an actuated axial spool to cope with the discontinuous yarn demand of the loom at high speeds.

This way of feeding creates a twist in the yarn fed to the weaving loom. When weaving flat yarns, eg. flat plastic tapes, this results in lower qualtity textile as the twists in the yarn can be seen in the textile as small gaps.

The yarn must be stretched at all times whilst weaving. If the tension would be too low, this might result in yarn loops in the fabric. If the tension would be too high, the yarn could break. So preferably, the yarn tension remains constant, despite the high yarn accelerations during each infeed.


IRO has developed a system where the bobbin is actively unrolled in the radial direction.
An active controlled bufferarm deals with the difference between the bobbin unrolling a constant speed and the repetitive but variating yarn demand of the weaving machine per loom cycle. The buffering and release of yarn from the buffer should happen so that the yarn tension remains as constant and low as possible during the whole infeeding cycle.
A low inertia spring tensioned sensor arm measures the yarn tension and feedbacks it to the control system actuating the bufferarm.

Model based approach

Vintecc was involved in the development of the advanced control algorithms necessary. The speeds at which the loom is running are too high for a classic feedback system. Also the control system should be flexible and smart enough to cope with:

  • Different types of yarn
  • Different types of bobbins
  • and different loom types with different widths running at various speeds.

For this development we used the model based approach.

First a plantmodel / digital twin was made of the Zero Twist Feeder and the yarn demand by the loom in Simulink, allowing to start the development of the control software in simulation. The control model itself consists of an automatic calibration routine to cope with the variablility (of yarns, bobbins and looms) and a learning control strategy to improve keeping the yarn tension constant whilst weaving.

Of the created control model, code was generated and deployed to a controller target for testing on a prototype of the Zero Twist Feeder. This proved the concept of Zero Twist Feeding showing a significant increase in textile quality.

User friendly interface on a tablet

Next to this, Vintecc also developed an HMI to control the system, change some parameters and visualize important signals, so that textile engineers at IRO could try out the system and suggest improvements.

The relevant machine information is retrieved from an OPC-UA server and visualized on a web page.

This includes important diagnostic information and visualization of some key information in graphs, such as the yarn tension. Several machine settings can be set from the HMI to configure the Zero-Twist Feeder.

The Zero-twist feeder is delivered to the customer with a small tablet. For userfriendliness, the HMI was also deployed as an Android app, which shows the same HMI full screen after connecting to the local wifi hotspot of the Zero-twist feeder.

On a password protected service page, service technicians and engineers can access Capture.

Capture is logging all important sensor values and the machine state at a high rate locally on the IPC, with a retention policy of one month. If a customer experiences problems, this helps to investigate their issue and give usefull feedback.

Development speed increase

The use of model based approach had several advantages.

First of all the creation of a digital twin allowed to develop the control software faster without risking damage to the actual setup.
IRO also appreciated that the model was realistic and could be talked about in engineering units.
This made the discussions with non-software minded engineers easier.

Secondly, the codegeneration of the control model allowed for fast iterations of software.
This speeded up the development significantly.
Only 1,5 year after IRO built the first prototype and Vintecc got involved for software development, the working machine was presented at the ITMA2019 fair in Barcelona.


Vintecc assisted IRO with the software development of the Zero Twist Feeder.
By introducing model based design and codegeneration, the path from idea, to proof of concept, to market introduction became significantly shorter.
Within the Van De Wiele group, the workflow used for the development of the Zero Twist Feeder has caught the attention of several interested divisions.

After the market introduction, it was decided to kick off the production and sell the first units to customers.  IRO and Vintecc further continue to collaborate to improve the Zero Twist Feeder regarding performance and features.


By introducing Model based design and codegeneration, the path from idea, to proof of concept, to market introduction became significantly shorter.

Brecht Vermeulen

2min read

Our competences

  • Matlab/Simulink Modeling
  • Digital Twin
  • Rapid Prototyping

Do you have a similar project and you can use our expertise in Digital Twin and rapid prototyping?

Other cases you might find interesting

Our solutions

Software programming / Sensor technologies / Automatic code generation

Process simulation / Virtual commissioning / Automated testing

Data collection & exploration / Reporting & alerting / IoT & cloud

Machine learning / Data driven insights / Vision systems