Inspection Technology
Real-time inspection is strongly desired across the industry. Electroimpact offers real-time inspection to ensure ply-fill.
Electroimpact Inspection UI
Boeing-certified and in use every day in the Composite Wing Center, Electroimpact’s Inspection System will show data in many forms.
The ply-approval criteria is tailored to the specific customer’s requirements. The system can be setup to use a statistical model based
on the data collected, a goal post model, or both. In our facility we use both a statistical model and a goal post model following the following 2 rules:
- |μ±3σ|≤ .3"
- |Err_max |≤ .5"
μ-mean
σ-standard deviation
Err_max - Maximum Error from Nomina
Ply-Fill
Electroimpact AFP cells verify ply fill (ply boundary, lap/gap, tow error) by measuring tow end placement and evaluating layup surface profile.
RIPITx
RIPITx completes in-process inspection of tow end placement, laps/gaps, FOD and more to guarantee ply fill and enable automatic ply buy-off. Uninterrupted layup is a critical component to the AFP4.0 philosophy, and RIPITx enables complete part builds with no operator intervention or manual inspection.
Ply Boundary – RIPIT (Real-time In-Process Inspection Technology)
This proprietary inspection system identifies the location of tow ends. It evaluates system inputs and many feedback mechanisms on the AFP head to estimate tow end placement in real time. The system identifies the location of tow ends 100% of the time. The accuracy is characterized as follows:
|μ +/- 2σ| < 0.10"
The base version of RIPIT costs no extra money.
For an additional cost, high accuracy 3rd party sensors can be added to the system to improve the standard deviation of measurements. We currently offer a 3rd Party profilometer system. This improves tow end inspection accuracy to:
|μ +/- 2σ| < 0.04"
Lap/Gap
3rd Party profilometer system:
Similar to the sensor used in tow end measurement, this sensor is used for lap/gap measurement. It offers the following capability:
Gaps:
| Mean error |
less than 0.005" |
| Standard deviation |
less than 0.005" |
| Repeatability standard deviation |
less than 0.003" |
| 0.007" < Detection size < Tow width / 2 |
Laps:
| Mean error |
less than 0.010" |
| Standard deviation |
less than 0.010" |
| Repeatability standard deviation |
less than 0.003" |
| 0.015" < Detection size < Tow width / 2 |
NOTE on spar type parts. Due to the sharp corner geometry of the box shaped spar, there is a blind spot for lap gap measurement in the spar corner region. This is potentially impossible to overcome because the measuring device trails the rotation center. As such it ends up pointing in space as the machine traverses the 45- and 90-degree ply orientation in the spar corner region.
Tow Present and Twisted Tow Detection
The profilometer system can detect twisted, missing, broken and folded tows with high accuracy. It reports location of the defect, which can be sent to the laser projector system to assist the operator with locating the defect on the part. The twisted tow and missing tow detection is not delivered with the profilometer system by default, as it requires the profilometers to be spaced more tightly; it can be purchased as an option.
FOD
Eliminate FOD – Keep the Head Clean
This is our preferred strategy and is working out very well in our internal AFP4.0 tests. The PEK, combined with planned maintenance, will reduce FOD to a very small amount. Buildup of resin and fuzz doesn’t occur with the materials we’ve tested.
Detect FOD – 3rd Party Profilometer
We are currently developing a FOD detection system that uses a 3rd party laser profilometer. This is at TRL 5-6. Initial results show reliable FOD detection of features greater than the following dimensions: .1" x .1" x .05" tall.
Detect Poly Backer
For AFP material systems that use poly backer, we have a poly backer detection system that looks to see if any poly backer is attached to the tow as it enters the final redirect. This ensures that the poly backer is not getting fed onto the part.
Process Error Detection
Drive Roller Wraps
We detect when material wraps around the drive roller by monitoring torque disturbances on the drive motor. When a wrap is detected the motor stops motion to limit the wrap severity.
Backing Film Detection
We offer an option to add sensors on the head to detect if backing film travels beyond the backer take-up spool. Upon detection the payout is stopped before the backing film ever makes it onto the part.
Compaction Roller Wraps
With the adoption of the VSS laser heater, compaction roller wraps have essentially been eliminated as a source of error. Even so, we do have a camera system for detecting compaction roller wraps that typically stops the machine before two revolutions, and often before even one. This significantly reduces damage to the part and limits the amount of rework required.
Optical Laser Template (OLT)
Electroimpact offers Aligned Vision laser projectors. These can be fixed-mounted on lighting truss posts or mounted on a moving AFP machine to provide enough range to cover the entire work zone.
Electroimpact offers full featured laser projector integration. Our software is capable of displaying:
- Ply boundaries |μ +/- 2σ| < 0.05"
- Individual tows or courses
- Tow ends
- Programmed overlaps and gaps
- Tooling locations
- Course direction
- Defects
The user interface includes a full 3-D model of the current ply to allow for easy operator identification of errors or areas of concern. The CGTech programming system outputs the laser files for each ply automatically—no need to program them separately.
The OLT identifying a specific tow within a course.
EI4.0 Data Analytics
EI4.0 is Electroimpact’s data analytics software package. It analyzes logs generated during operation of the AFP machine and presents users with insights distilled from those logs. The up-to-date information provided by EI4.0 is vital for mangers, maintenance staff, part programmers and machine operators, all of whom need to be able to make informed decisions quickly based on real data.
For the Manager
In order to make informed management decisions, it’s necessary to have a real picture of how the cell is being used and where problems are arising. EI4.0 captures all aspects of the production process and allows managers to see how statistics like machine reliability, floor-to-floor rate, and equipment utilization vary over time. Performance can also be compared between shifts.
As an example, in the following pair of pie charts, which represent average time spent on various tasks over a large number of builds, Shift 2 tends to spend about half as much time inspecting the part using the laser projector as Shift 1 does. Perhaps there is knowledge among the workers in Shift 2 that needs to be communicated to Shift 1 to help them work more quickly.
For the Machine Operator
Building an AFP part involves many repetitive sub-processes, as the part needs to be laid up layer by layer. The EI4.0 part build dashboard prevents the machine operation team from getting lost in the weeds by providing production benchmarks based on past performance. The dashboard compares progress through the current build with historical averages and can be programmed to give the operators production targets for each shift.
For the Maintenance Technician
EI4.0 allows the user to drill down through detailed historical information on the performance of individual parts of the AFP machine. This drill-down capability is useful for maintenance workers who are interested in identifying points of failure and fixing them to keep the machine working at capacity.
As an example, if the maintenance technician wants to diagnose head performance, he can graph Mean Strips Between Failure by head over a set period of time, say the last two months.
Head 1 has significantly fewer Mean Strips Between Failure.
To investigate, the technician can drill down further to see frequency of tow error by head lane.
Lane 2 seems to experience significantly more errors than other lanes, and lanes 6 and 9 are almost tied for second place. Maybe it’s time to replace some hardware to bring the tow error rate down.
For the Part Programmer
EI4.0 can identify individual courses which tend to produce a greater number of tow errors. Based on this information, the part programmer can investigate the NC program in order to determine why those certain courses tend to cause layup problems.
Increasing Reliability and Machine Utilization
It is Electroimpact’s goal to help all our customers achieve stellar machine performance and reliability, and to provide customers with data that helps them to keep the machine laying tow as much of the time as is possible. EI4.0 is a keystone in this effort, providing the data necessary for continual process improvement.
