Editor | On 18, Dec 1997
Squeak and “Buzz” Project
TRIZ Case Study
About the Authors
Michael Lynch has over 20 years of automotive experience and has been leading the implementation of TRIZ (Theory of Inventive Problem Solving) methodology at Ford Motor Company for the last four years. Mr. Lynch developed and is currently leading the Inventive Problem Solving Team that utilize both TRIZ and SIT (Structured Inventive Thinking) to solve intractable company problems at Ford’s North American Quality Center. Mike has been awarded Ford Motor Company’s highest technical honor (Henry Ford Technology Award) and has received patents in both the United States and Europe. Mr. Lynch holds a Bachelor of Science degree in Electrical Engineering from Lawrence Technological University and a Master of Science in Engineering Management from Wayne State University.
Benjamin Saltsman has applied the TRIZ methodology over the last four years in the consumer products and automotive industries. Over the last year and a half, Ben has worked for Ford Motor Company solving intractable company problems at Ford’s North American Quality Center. During this period, over 45 invention disclosures have been submitted. Mr. Saltsman also has two Defensive Publications to his credit along with one patent pending. Ben holds a Master of Science degree in Materials Science from Stevens Institute of Technology and a Master of Science Degree in Metallurgical Engineering from Moscow Automechanical Institute.
Colin Young has been involved in the SIT and TRIZ methodologies at Ford Motor Company for over two years. During the last year and a half, Colin has applied these methodologies while solving intractable company problems at Ford’s North American Quality Center. This has resulted in the submission of over 45 invention disclosures. Mr. Young has received one United States Patent and has another patent pending. Colin also has one Defensive Publication to his credit. Mr. Young holds a Bachelor of Science degree in Mechanical Engineering from University of California at Davis and is currently completing a Master of Science in Mechanical Engineering at Purdue University.
Windshield and backlight molding squeak problems have been in existence for several years. The issue became more complicated about 10 years ago when a change from metal molding to plastic was undertaken. Over the last five years several teams tried to resolve the situation and a number of design alternatives were explored. Unfortunately, no robust and cost effective solution was found.
A temporary solution has been developed and recently implemented by the Vehicle Program Team that resolves this concern, however, additional materials, assembly labor and variable cost were required. In addition, this solution only addresses the flutter (buzz) problem and has no effect on molding squeak.
Since conventional problem solving techniques have not provided an acceptable solution for the squeak and flutter (buzz) issues, John Megdan and Paul Ashburn sponsored this special project that utilized an inventive problem solving methodology, referred to as TRIZ (See Section 6 for more details). The goal of the project was two-fold:
- Identify technical root cause of squeak and flutter (buzz) phenomena.
- Develop inventive solutions that eliminate both conditions without compromise.
|Michael Lynch, Engineering Supervisor|
|Benjamin Saltsman, Product Design Engineer|
|Colin Young, Product Design Engineer|
|William Gulker, Product Design Engineer|
|Michael Martin, Product Design Engineer|
|Ellen Brown, Product Design Engineer|
In this phase we work closely with the Subject Matter Experts (SME’s) to review any prior test data and methods. The next step in this phase is to review prior solution attempts to learn from past experience and ensure that there will be no duplication of prior effort. In addition, we interview available SME’s and review manufacturing processes to acquire additional information that may not have been documented. In parallel with the above, we collect benchmarking information and conduct a patent search to establish the “state of the art.” At this point, we develop a technical description of the problem and formulate the contradiction(s).
In this phase, it imperative that the team members adopt a structured thought process that breaks psychological inertia. It’s been our experience that outside consulting firms overwhelm the participants with every aspect of TRIZ and fail to spend adequate time explaining the fundamental thought process of the methodology. Ed Sickafus (SRL Physics Department Manager) shared this concern and has developed an “in-house” Structured Inventive Thinking (SIT) course that resolves this issue. It is recommended that all team members participate in this training.
By this time in the project, the problem has been properly defined and the contradiction(s) formulated. This typically results in the development of a fair number of conceptual solutions.
In this phase, the team works with the TRIZ software to ensure coverage of the entire solution space. This allows the team to look across engineering disciplines and identify potential solutions to the problem that may have been applied in other industries.
The team engaged in an in-depth study of the technical background information to develop a thorough understanding of the fundamental physics associated with the problem.
Over the last five years a number of alternatives have been tried and a substantial amount of experimental work has been completed. A small team of engineers analyzed this information and structured it in the manner consistent with the TRIZ methodology.
In addition, the team met with key individuals who have worked on this issue over the last several years to develop a thorough understanding of the problem. The same list of questions was prepared and discussed with each of the Subject Matter Experts (SME’s). Most of this information is summarized in the Roadmap A, page *.
An outside company – Defiance STS, Troy, Michigan – was contracted to perform the majority of the squeak testing. In our recent review of the test equipment setup, several major faults were identified. Important test parameters such as humidity, quality and handling of the test samples were not controlled, resulting in poor test repeatability. A proposal to update the test equipment and procedure will be implemented prior to verification testing to ensure that the test and equipment are capable of producing valid results.
Based on the interviews with the SME’s, it was determined that squeak is caused by a phenomenon highly dependent on environmental conditions. Temperature appears to have the most significant effect; competitive benchmarking has demonstrated that squeak becomes more pronounced under cold conditions (i.e.-20oC). A sound level of 50 dB or greater is considered objectionable.
4.2 Flutter (Buzz)
Since this issue was relatively new, limited test data was available for our review. Flutter or “Buzz” typically occurs while driving in a cross wind at a vehicle speed in excess of 55 mph. These conditions can be simulated and tested in a wind tunnel that allows for rotation of the vehicle in the test chamber and has the capability of controlling critical environmental conditions.
The group classified prior solution attempts in a form of a roadmap. The purpose of the Roadmap was to capture all attempts made during the last several years to resolve the Windshield/Backlight molding squeak and flutter (buzz) problem. The following roadmap summarizes this information. A summarized description of each level can be found on pages * through *.
4.3.1 Roadmap A
IMPORTANT: This roadmap illustrates the situation prior to the beginning of this project.
Windshield / Backlight squeak and rattle type problems can be classified in two major categories:
- Flutter (Buzz)
Let’s consider the Squeak part of the Roadmap first:
22.214.171.124. Layer A
Squeak is known to be caused by a Stick-Slip mechanism.
126.96.36.199. Layer B
Traditionally stick-slip phenomenon in the moldings have been dealt in one the three possible ways: promote motion, prohibit motion and in-the-rabbit design.
Stick-Slip phenomenon can be approached from either the position of promoting motion (slip) or prohibiting motion (stick) of the molding lip at the contact point with the sheet metal. In-the-rabbit design is a fundamentally different approach since a molding lip rests in the rabbit, however, this design adversely affects styling and will not be considered as a valid alternative in this analysis.
188.8.131.52. Layer C
Most of the effort in the past several years has been concentrated in the area of promoting motion. The directions considered were: variation of the contact point geometry, lubrication and material selection.
A hinge design appears to be the only conceptual approach to prohibiting the motion.
184.108.40.206. Layer D
Various contact point geometry alternatives have been studied. Macro level (stipple, pointed end, round end) or micro level (RIM – reaction injection molding – provide rough surface; extruded provides smoother surface)
Lubrication appears to be the most obvious and, therefore, the most considered area. Prior attempts varied from expensive lubricants such as Krytox® to Attchison Colloid, which are currently used by major molding suppliers. Also included in this category is flocking. The weakness of lubrication is the robustness of application and associated expense.
Molding material variation was another area of study and experimentation. Soft materials seem to perform better from the squeak point of view, but they are more susceptible to flutter.
Hinge design was initially implemented in the form of a two-piece molding design. Later an attempt was made to come up with a single piece molding, which relied upon flexing of the entire lip to accommodate for the motion. Either design is prominent to flutter.
Now let’s consider the Flutter/Buzz part of the Roadmap.
220.127.116.11. Layer A
It is known that Flutter/Buzz phenomenon is caused by Low Stability of the molding lip. This effect is typically observed at ambient to warm temperature range, high speed (above 55 mph) driving condition and a strong cross wind.
18.104.22.168. Layer B
Increase stability is an obvious approach to resolve low stability problem.
Air pressure build-up under the lip was suspected to cause instability, which resulted in several solutions.
Additionally, engineers have recognized the temperature effect concern (plastics get softer at higher temperature and stiffer at lower temperatures), but we did not find any attempts to temper this effect.
22.214.171.124. Layer C
Increasing stability of the lip manifested itself in increasing stiffness.
Foam blocker is the primary conceptual approach to reduce the pressure build-up under the lip. Additionally, attempts have been made to vent (reduce) the pressure in the cavity.
126.96.36.199. Layer D
Increasing stiffness is suggested to be implemented by selecting stiffer material and making the lip thicker (however, this will to make the squeak worse).
Foam blockers are used as a permanent solution where it is injected under the lip and foamed in place or as a temporary solution where a 2 inch piece at the bottom of the A-pillar was recently changed to a foam strip under the entire length of the A-pillar. Needless to say this solution adds extra cost, and its long term durability is questionable.
Two directions have been explored in the venting arena: venting to the roof and holes in the molding. Both directions had appearance issues associated with them.
4.3.2 Patent search
A brief patent analysis was performed to determine ‘the-state-of-the-art’ in fixed glass molding technology. Patents over the period of 1990 through 1996 were reviewed in detail. The statistics of the patent analysis is shown below:
|296/93 Windshield weather-strips||
|296/192 Windshield cowl||
Thirty-nine relevant patents were considered. The following is a summary of our analysis:
4.4. Mechanisms of squeak
|Squeak is caused by stick and slip phenomenon occurring at the contact point of the molding and the painted body sheet metal.
As illustrated by the schematic on the left, the energy released during the transition from static friction to dynamic is transferred to audible sound.
|Flutter is a high speed phenomenon. It is primarily caused by a Bernoulli effect.
When high speed air passes above the wing-shaped molding, the pressure above the molding drops, resulting in a lift force. This effect is reinforced by somewhat pressurized air inside the cavity formed between the A-pillar and the molding.
This results in a cyclic release of pressure heard as a “buzz.”
|The flutter (buzz) is caused by unstable variations of pressure that occur under certain, high-speed flow conditions. The air enters the covered channel formed by the space between the A-pillar, the windshield, the urethane bead and the windshield molding. The pressure caused by this air builds until it is sufficient to lift the molding lip. This lifting of the lip releases the pressure, and soon the lip closes against the sheet metal and the process repeats itself.
In addition, other phenomena, such as Karman vortex shedding, may also contribute to the flutter (buzz).
5.1. Roadmap B
IMPORTANT: This roadmap illustrates the situation after completion of this project.
NOTE: The solutions developed during this project are shown in gray boxes.
Before we describe the solution concepts, it is important to specify the restrictions under which these concepts were developed. Should these restrictions change in the future, the solution set could be modified and/or expanded.
The following restrictions were imposed:
- The molding is of extrusion type. This type of molding is primarily used for windshields and backlights.
- The styling can not be changed. This means that the molding is required to have a lip concealing the cavity formed between the A-pillar and the glass.
- No drastic changes to the molding design and/or manufacturing process are permitted.
5.3.1. Interim solution
The Vehicle Program Team developed and implemented an interim fix for the flutter (buzz) issue. A foam blocker is installed in the rabbit (cavity between the molding and the body sheet metal). This blocker prevents pressure build-up, and, thus, reduces the lift force on the molding lip. Additionally, the molding cross section was made thicker and the material changed to higher durometer. Since stiffer materials are known to promote squeak, slip coat is applied to the under side of the molding to reduce the probability of squeak.
As stated in the introduction, this team’s goal was to develop solutions for both squeak and flutter(buzz) without increasing manufacturing/assembly complexity and labor or material cost.
In this section we will review newly developed solution concepts. Please refer to the Roadmap B on the previous page and note that the following solution concepts address both squeak and flutter.
5.3.2. Composite material
As a result of the structured approach undertaken by the team, a new conceptual direction on how the squeak issue can be approached was revealed (Layer B, Prevent sound wave propagation box in the center of the Roadmap). As discussed above (Mechanisms of squeak, Page *), squeak is caused by stick and slip phenomenon. This results in a formation of a wave traveling along the molding lip. This wave can be related to the audible sound wave. It is suggested to use composite structure (Layer C) of the lip to make the lip stiffer in the tangential direction and, thus, prevent sound wave propagation. Several possibilities can be explored.
An important feature of the following concept is that one can choose PVC with low durometer (soft), which is known to be less prone to squeak, and achieve required stiffness by reinforcing the structure.
The concept of composite structure is a true TRIZ solution. In early 1950’s the founder of the TRIZ methodology, Henry Altshuler, developed the Table of Contradictions. The use of composite materials was specified as one of the inventive principles applied to resolve the contradiction of increasing strength while decreasing weight.
After viewing the high speed video film of squeak, the team concluded that the early stages of squeak are characterized by formation of little ‘bubbles’ along the edge of the lip. As the vibration input increases, the frequency of these ‘bubbles’ increases as well, this results in motion across the entire lip edge.
|A single filament (wire) along the edge of the lip will interfere with the above described mechanism and, therefore, eliminate squeak. The fiber can be co-extruded with the PVC molding.
Reinforcement techniques are used in various industries, including automotive. For example, rubber tires are reinforced by metal cord.
188.8.131.52. Multiple parallel filaments
Multiple parallel filaments is an extension of the previous concept. These filaments could be used if a single wire does not provide enough reinforcement.
|From the TRIZ point of view this is a natural step along the evolution line. There are eight laws of evolution of technological systems in the TRIZ methodology. One of them, the Law of Increased Dynamism and Controllability contains a line that states: “technical systems evolve from a monosystem to bi-system to poly-system.”|
184.108.40.206. Fillers (fibers, globules, etc.)
This concept describes a different composite structure. Separate randomly oriented short fibers or fibers oriented in a preferred (extrusion) direction can be pre-mixed into molten PVC and subsequently co-extruded.
|The benefit of such design is that the matrix material can be soft (which is good for squeak) and the stability required for flutter resistance can be provided by the reinforcement.
From the TRIZ point of view, an ‘infinite’ number of elements (small fibers) can be thought of as the ultimate stage of a poly-system.
Mesh will further stiffen of the molding lip, while preserving softness of the outside surface. Again, mesh can be co-extruded with the PVC.
|Mesh reinforcement is used in many areas, for example, seal manufacturing and concrete reinforcement.|
220.127.116.11. Metal foil
The metal foil can extend half to two thirds of the lip length, so that the very edge of the lip remains very flexible to accommodate for build variation. (Although the edge will be flexible and can potentially flutter, the flutter range will be beyond the maximum vehicle speed). However, it is important to keep in mind that the molding can not be made extremely stiff because it may interfere with the glass installation (may push the glass out of the body opening while urethane adhesive cures).
|From the TRIZ point of view this concept makes the best use of the available resources. Metal foil is already used in the molding to ensure tight connection of the molding to the glass. This concept suggests extending it into the lip.|
5.3.3. Surface modification
It is common knowledge in the auto industry that lubrication will reduce friction. Among the most efficient lubricants used to reduce friction between the molding lip and the body sheet metal is a fluorinated lubricant Krytox® (developed for NASA in late 1950’s). Unfortunatly, this lubricant is also the most expensive.
A new way of promoting motion of the molding lip is by modifying the surface. This concept is shown on the left of the Roadmap B. This method incorporates lubrication into the molding manufacturing process.
From the TRIZ point of view, this can be thought of as combining two separate objects (molding and lubricant) into one (molding with modified surface). Also this complies with another TRIZ law of transition from macro-level to a micro-level.
18.104.22.168. Plasma conversion
Plasma conversion is a relatively new process used for high density polyethylene (HDPE) and thermo-plastic olefins (TPO). This technique is used to create a Teflon surface layer on HDPE and to activate the TPO surface. Teflon – fully fluorinated HDPE – has a low coefficient of friction and will perform better than conventional slip coat due to the fact that it’s coefficient of friction is significantly less temperature dependent.
22.214.171.124. Fluorine treatment
Similarly fluorine treatment is used for HDPE. In this process fluorine gas is introduced in the chamber with the plastic part and chemical reaction on the surface of the material takes place. This results in a formation of a Teflon-like layer on the PVC surface. Teflon has an extremely low coefficient of friction. This process is currently used by Ford ACD on fuel tanks to prevent permeation of the gas vapors through the plastic shell.
Alternatively, a thin layer with low coefficient of friction (HDPE or Teflon powder) can be co-extruded. This type of process is used for hoses, seals, O-rings, etc.
5.3.4. Material variation
Several materials (PVC, Alcryn, EPDM) have been considered in the past for the fixed glass molding. Another group of widely used plastics in the automotive industry are thermo-plastic olefins (TPO).
126.96.36.199. Thermo-plastic olefin
TPO’s are currently used for bumper facia. After some required treatment TPO can be painted. Use of this material for the fixed glass molding will allow Ford to have color-keyed molding. This opens new opportunities for styling.
5.3.5. Rolling lip
The area of prohibiting motion at the edge of the molding lip has not been explored sufficiently in the past. The underlying idea is that the motion occurring between the edge of the molding and the body sheet metal (approximately 0.1 mm) can be absorbed by the flexible structure of the molding. However, the contradiction to keep in mind is that the molding cannot be soft due to a potential flutter problem.
The two designs, described below, have a stiff lip (anti-flutter measure) and accommodate for relative motion with specific lip edge layout.
Both of these concepts also provide an edge that is more stable against bending in the plane of the sheet metal, which will eliminate the condition for initiation of slip and squeak.
Wheels are extensively used in various industries to translate sliding friction into rolling. In the two concepts described below the wheel is used in a somewhat different way.
188.8.131.52. Wheel design
The concept of the wheel at the edge of the lip will allow the edge to oscillate about the neutral point.
184.108.40.206. Tubular lip edge
A tubular edge at the end of the lip will buckle like a match box compensating for the motion.
Such construction will allow the molding to absorb relative motion of the glass with respect to the sheet metal by aligning the soft layer towards the edge of the lip. Stability against flutter will be provided by the stiff outside layer.
If you recall from the Problem Definition, flutter is more prevalent at high temperature and squeak increases at low temperature. To further decrease flutter and squeak, the material pair can be selected so that the molding will be pressed to the sheet metal stronger under warm (hot) temperature conditions and weaker under cool (cold) temperature conditions. Should such pair of plastics be hard to find, the metal foil discussed in the Paragraph 220.127.116.11 can be replaced with a bi-metallic foil with the desirable temperature induced behavior.
The concept of the bi-material is a natural step along the evolution path, according to the TRIZ methodology, as systems evolve from the mono-system to bi-system to poly-system.
5.3.7. Attach end point
Another way of prohibiting motion of the edge of the lip is to attach the end point. The following three approaches can be considered.
Gluing of the lip to the sheet metal appears to be most straightforward. The benefit is apparent: if the edge of the lip is glued down, then no squeak or flutter can occur. However, at least two concerns should be considered:
- Application of liquid glue at the assembly plant is highly objectionable.
- The glue applied cannot be water soluble.
There are several possibilities that address these objections:
- A non-marking glue (i.e. used on Post-It® notes) is applied at the molding manufacturing facility.
- Solid glue (i.e. used on paper envelopes) is applied at the molding manufacturing facility. The glue is ‘activated’ by a catalyst after the molding is installed.
18.104.22.168. Sticking material
Often times rubber seals stick to the part they are sealing. Typically it is considered undesirable. Conversely, in the case of the molding, this effect can be desirable. If edge of the molding sticks to the sheet metal, this will eliminate the conditions for squeak and will provide improved protection against flutter. This concept will work particularly well with a bi-material molding.
22.214.171.124. Magnetic molding
If magnetic particles are mixed into the PVC molding, it will have magnetic properties and attract to the steel body sheet metal. Additionally, the magnetic particles can be aligned in a preferential direction during extrusion to improve magnetic properties of the molding.
Replacement of a mechanical interaction with a non-mechanical (magnetic, for example) is another inventive principle used in the TRIZ methodology. Magnetic moldings are currently used for refrigerator doors.
5.3.8. Downward force
A novel direction for solution was uncovered for flutter/buzz issue. As discussed in the Paragraph Mechanisms of flutter, Page *, high speed air above the molding is believed to be one of the major causes of the flutter issue. This solution translates a harmful effect into a useful effect. This principle is often used in the TRIZ methodology to obtain innovative solutions.
126.96.36.199. Anti-wing shape
It is suggested to use an anti-wing shape for the molding so that the very same air which causes the molding to flutter will impose a positive downward force and prevent it from fluttering. This concept has been applied on a number of systems across many industries from windshield wipers to race cars.
6.1. TRIZ and SIT
The innovation methodologies TRIZ and SIT were used to structure this problem and generate solution concepts. TRIZ is a Russian acronym for Theory of Inventive Problem Solving and is the result of over forty-five years of research and development that began in the former Soviet Union. Through this research, it was recognized that the same fundamental problem or contradiction had been addressed by a number of inventions, but in different areas of technology. In addition, it was discovered that the same fundamental inventive principles and subsequent solutions were used over and over again, often separated by many years. The research scientists sought to extract, compile and organize this knowledge and reasoned that if the latter inventor had knowledge of the earlier solutions, their task would have been straightforward.
TRIZ is a structured methodology for innovation, a methodical way of examining inventive problems, exploring the solution space and developing conceptual solutions. It is a means of reaching across engineering disciplines to solve problems by using solutions from other areas of technology and industry. TRIZ generalizes worldwide experience in invention, systemitizes successful methods of solving technological problems and reveals regularities in the evolution of technological systems. This methodology is gaining popularity in the industrial community as companies expand their engineering competency from a corporate “lessons learned” knowledge base to a cross-industry technology base.
SIT (Structured Inventive Thinking) is an extension of TRIZ, conceived in Israel by TRIZ practitioners with the goal of making the TRIZ methodology more accessible and easier to learn. The Ford Research Laboratory Physics Department is continuing to develop the SIT methodology and has developed a training course for Ford engineers.
We have used a mixture of TRIZ and SIT methods on this problem. Both methodologies share the following qualities and provide systematic, disciplined ways to:
- Translate a general problem, such as molding squeak, into a specific technical objective, such as stiffen molding in direction X, while softening it in direction Y.
- Solve the specific technical objective by using general inventive principles that are contained within the methodologies. These principles are drawn from many diverse engineering disciplines and provide engineers with solutions that are outside their areas of expertise.
The advantages of these methodologies are that they focus the solution efforts on the technical root cause or causes of the problem, and then provide the tools to invent design solutions. The result is a variety of innovative solution concepts of which several are useful, practical solutions to the problem.
General Problem: The molding lip must not squeak when subjected to small scale random oscillations against the painted body sheet metal and, additionally, the lip must not buzz when presented with any high speed air flow situation likely to occur during usage.
Experts on the squeak problem already knew that squeak was caused by the molding lip edge transitioning between sticking and slipping against the sheet metal. Many attempts had been made to keep the lip edge in a permanent slip mode, hence slip coat, Krytox, etc., but few engineers had investigated methods of keeping the edge in the stick condition. Since this branch of the solution tree was less developed, we focused considerable effort here (see Roadmap B).
Existing test data had shown that the amount of relative motion occurring during slip was surprisingly small. It was on the order of 0.1 mm. Thus, it appeared that if the molding lip were soft enough, it should be able to accommodate this minute motion without slipping. Test data showed that softer molding lips were less prone to squeak than stiffer ones.
The experts on the buzz problem pointed out that as a molding lip becomes stiffer, buzz problems are reduced or eliminated. The solution requirements of the two problems seemed to conflict: The molding lip must be stiff to prevent buzz and it must be soft to prevent squeak. Once the problem was properly formulated, the TRIZ software yielded the problem statement: Make the lip stiff in the lift-off direction to prevent buzz, but soft in the tangential direction to prevent squeak. This is a statement of a specific technical objective. The next step was to create designs that satisfied this condition.
One innovative principle is ‘use rolling to counteract friction’ by analogy to ball bearings. We considered how the lip edge could be made to roll against the body. Rounded lip edges had been tried before, however, the rounded edge was not provided with a hinge to pivot on. We produced a rounded edge with a hinge. Tank tracks represent another type of rolling contact, where a loop shaped track is used to enlarge the contact area. A different design was conceived that made use of that analogy (see Rolling Contact Concepts on page *).
By providing a lip in rolling contact with the body, it should be possible to make the lip stiffer, to prevent buzz, and at the same time make it softer in the tangential direction, to prevent squeak. Rolling contact allows relative motion without slipping.
We would like to thank the following individuals who kindly provided support throughout the project, shared with us their knowledge of the subject and assisted us with gathering information:
|John||Megdan||Engineering Support Manager|
|Support & Technical Assistance:|
|Mike||Freeman||Product Design Engineer|
|Jenny||Remington||Product Design Engineer|
|Jon||Skelly||Product Design Engineer|
Additionally we would like to thank engineers from Defiance STS for sharing their knowledge and expertise.