Doctor Hart's evidence
167 In connection with the issue of obviousness, Dr Hart's evidence is potentially of some importance. He is highly qualified, both academically and by practical experience, in areas which are presently relevant. He holds the degrees of Bachelor of Engineering with Honours, Master of Engineering Science, and Doctor of Philosophy in Monash University, is presently the principal of a company providing mechanical and electrical engineering consultancy and has been closely involved with heavy vehicle dynamics, mechanics and braking. He has provided four different reports to the solicitors for BPW. In the first report, Dr Hart answered 15 questions which had been posed by the solicitors. He was told in his letter of retainer that:
The priority date for the Patent is 19 August 2003. When you are considering what comprises the common general knowledge and prior art base you need to do so by reference to this date and to ignore any information that is after that date.
The terms "prior art base" and "common general knowledge" were then defined.
168 Question 1 enquired as to the way in which a vehicle's suspension system works by reference to a typical vehicle suspension available in Australia prior to 19 August 2003. Question 2 enquired as to other systems available prior to that date. Question 3 asked as to characteristics, measurements or criteria commonly used as indicators of suspension performance, again by reference to components of the typical system. Questions 4 and 5 directly concerned the Patent. Dr Hart was provided with those parts of the Patent headed "Field of Invention" and "Background Art". The questions addressed those parts. I shall return to those questions. In Question 6 Dr Hart was asked about the regulatory regime in existence in Australia at and prior to 19 August 2003 in relation to suspension performance of haulage or freight trucks, including NRTC Guidelines. Question 7 dealt with road-friendly suspensions, enquiring as to tests, rules, guidelines or other parameters and methods of testing. Question 8 was as to the capacity to measure damping ratio, oscillation frequency and/or road-friendliness of a suspension system whilst the vehicle was in use, and as to whether such measurement had been done at or prior to 19 August 2003. Question 9 again dealt with the Patent, seeking Dr Hart's view as to the problem identified by the Patent and the solution. His reply was :
The problem or problems that are identified in the "Field of Invention" and "Background Art" statements that are desirable/necessary to solve are:
• Development of a test that can be conducted quickly without dismantling any parts;
• Development of a test that is accurate and takes into account vehicle weight, suspension geometry, multi-axle suspension characteristics and shock absorber characteristics;
• Following on from the point immediately above, development of pass/fail criteria for the method that is developed.
169 Question 10 was as to common general knowledge as at 19 August 2003, although that expression was not used. I shall return to his answer to that question at a later stage. Question 11 similarly dealt with common general knowledge. Question 12 dealt with available tests of suspension performance in 2003. Question 13 was as to organizations or companies which tested suspensions for road-friendliness as at August 2003 in Australia. Question 14 asked as to commercially available systems for measuring suspension performance in Australia at or prior to 19 August 2003. Question 15 was as to the way in which any such system worked, and whether it required removal of the vehicle from service.
170 In a second report described as the "Stage 1A Report" he considered three further questions. Question 16 was as to whether electronic braking systems and electronic weighing systems were available in Australia prior to 10 August 2003 and, if so, how they worked in relation to suspension components. Question 17 was as to whether any of those systems monitored suspension performance as part of their operation and, if so, how they operated. In Question 18 Dr Hart was asked to attempt to solve the problem posed in the background art. I shall set out below the problem as posed for his consideration. In a third report, described as the "Stage 1B Report" Dr Hart was asked four questions, designed to clarify aspects of the terminology used in Report 1A, and two questions designed to clarify his methods of solving the problem. In the fourth report, described as the Stage 1C Report he was asked four further questions, all designed to explain or expand upon his proposed solutions.
171 Much of Dr Hart's evidence concerning suspension systems generally confirmed the evidence of other witnesses and is largely non-controversial. To the extent necessary I have summarized that overall body of evidence in earlier parts of this judgment. His evidence otherwise goes primarily to the questions of common general knowledge and inventiveness.
172 Question 10 enquired as to sources of knowledge and information which a person in Australia, interested in suspension performance monitoring, was likely to have ascertained, understood and regarded as relevant prior to 19 August 2003. Dr Hart identified various organizations and publications. By question 11 he was asked how, if he had wished to keep himself informed of relevant developments in suspension performance monitoring in August 2003, he would have done so. He replied:
Contact the Australian office of the [Society of Automotive Engineers] and ask for the name of an Australian expert and follow up. Seek information about upcoming technical meetings.
Visit the then NRTC website. Read reports being done for the NRTC by consultants. Go to review meetings organized by the NRTC. Speak with NRTC officers and ask for reference to Australian experts.
The NRTC reference would probably have led to Dr Peter Sweatman at Roaduser International in Melbourne. The references would probably also have led to the Australian Road Research Board in Vermont near Melbourne Victoria. Dr Sweatman formerly worked at the ARRB.
Contact the Australian Road Research Board and determine whether they were working in this domain.
173 In Question 12 he was asked about suspension performance testing in 2003. The only tests of which he was aware were tests to determine compliance with road-friendly suspensions, in particular VSB-11. In answer to Question 13 he identified organizations or companies then involved in testing suspensions for road-friendliness. In Question 14 he was asked if there were any commercially available systems in Australia, at or prior to 19 August 2003, which monitored suspension performance as part of their operations. He was not aware of any such instruments other than roller shakers. These devices were not mounted on the vehicle under examination. He identified some electronic braking systems as having been available prior to 10 August 2003. They did not monitor vehicle suspension. He said, concerning electronic weighing systems:
On-board weighing systems calculate the vehicle weight based on signals from air-bag pressure or load-cells. As a vehicle load is increased, the pressure in the air-bags of an air-spring suspension (if fitted) increases proportionate to the load on the suspension. If all the air-bags are sensed then a reasonably accurate estimate of the weight on the axle group can be made. Onboard weighing systems must be calibrated after set-up and periodically thereafter. That is, a known load is placed onto the truck at the usual position and the system is set to give an accurate reading.
A load estimate can also be obtained using load cells that are placed under the feet (or otherwise on the mounting plate) of a fifth-wheel coupling. This type of coupling is used on a prime-mover to pull a semi-trailer. The coupling supports the weight of one end of the trailer. If a measure of deflection or stress is available, then the imposed weight on the coupling can be estimated.
It is also possible to estimate the weight on a spring suspension by monitoring the deflection of the spring using a load cell or strain gauge. However, I do not believe that such a system was commercially available in Australia for road trucks in August 2003.
I am also not aware of load-cell based monitoring of fifth wheel couplings being commercially available in Australia in August 2003.
I recall that weighing systems that monitor air-bag pressure (Air-Weighing Systems AWS) were available in Australia in August 2003. They were available from a number of suppliers. …
174 In Question 17 he was asked whether any of these systems monitored suspension performance as part of their operation. He replied:
Both air-weighing systems (AWS) and Electronically Controlled Braking Systems (EBS) monitor suspension air-bag pressure. They do not monitor suspension performance. These systems specifically reject (or cannot follow) any high frequency pressure variations that arise from road bumps. They respond to the quiescent pressure level.
175 As I have observed, in question 18 Dr Hart was asked to attempt to solve particular problems allegedly posed in the acknowledged background art, namely:
It is therefore desirable to provide for an accurate test for testing the performance of suspension components on a vehicle to determine whether the suspension is functioning properly and within legally accessible limits while the vehicle is in use and without requiring the removal of the vehicle from service. In addition, it is desirable to provide for a test procedure for testing vehicle suspension components that does not require subjective interpretation by a technician.
176 Dr Hart said:
A suspension on a heavy truck must perform three essential functions. It must transfer the load weight from the truck onto the axle. Secondly, it must alter the height of the chassis rails in response to changes in the road surface height or orientation. Thirdly it must isolate the suspended parts from road bumps to an acceptable degree.
…
Hence, the essential characteristic of a suspension on a heavy road truck is to reduce the level of road bumps that are experienced by the suspended parts. The road bumps are experienced at a much higher frequency than are changes in road height/orientation. Therefore, the "quality" of the isolation is mainly concerned with the higher-frequency part of the suspension response. I will call the third function the bump-isolation performance.
A fourth functional aspect might be added to the list of suspension requirements. This is that the suspension response does not cause unacceptable road damage. This aspect is of particular concern to road-agencies. It is relevant only to heavy laden trucks. Road damage is related to but somewhat different from bump-isolation performance.
…
A suspension that has a degraded or broken component is unlikely to have an acceptable bump-isolation or road friendly performance. Therefore, a suspension that fails the road friendly suspension test may be defective. I will adopt the "road-friendliness" requirements as stated in VSB11 … as the definition of objective suspension performance. Other definitions could be developed.
177 Dr Hart then identified four ways in which the in-service performance of a suspension could be checked. The first method was probably not viable in August 2003, the relevant technology not being available. It involved measurement of the distance between the axle and the chassis rails at each suspension location. This information would provide a measure of the natural frequency and damping ratio. Solutions 2, 3 and 4 were as follows:
Solution 2:
Measure the air-pressure in each air-bag of an air-spring suspension. Determine the overshoot and damping performance from the air-pressure fluctuations. From these measurements infer the performance of the individual suspension.
A pressure sensor capable of responding to higher-frequency pressure changes (say 10 HZ) is needed. Electronic analysis of the pressure signals is then conducted. The suspension performance might be inferred from the pressure variations.
Method 2 relies upon measurement of the pressure variations in each of the air-bags. It is only applicable to suspensions that contain an air-spring. The technology needed to implement this method was available in August 2003.
Solution 3:
Measure the wheel speed on each wheel. Determine the variation of wheel speed during each tyre revolution. Look for tyre slip events that indicate that the tyre has lost contact with the pavement. This may indicate excessive suspension overshoot.
178 There was more detail concerning this method but clearly, it did not involve determination of forces acting on the suspension by weighing. However it would perform the same task as that prescribed in the Patent and would be on board the vehicle.
179 Solution 4 was:
Sense when a clash occurs between the suspension and the bump stop. Excessive clash on one suspension probably indicates excessive overshoot. A suspension that experiences substantially more clashes than other suspensions in the group may be defective.
180 Again, there is further detail. Again, the proposed solution does not involve weighing. Again, it would fulfil the same purpose as that identified in the Patent and would be mounted on board the vehicle.
181 The term "overshoot" was discussed in Report 1B. In effect, it is the extent to which the suspension moves away from its final position before settling in that position. The term "bump stop" is also explained in Report 1B. A suspension usually has two bump stops. They limit the travel of the suspension in each direction.
182 In the Stage 1B Report Dr Hart provided further details of the steps for each method. As to method 2 he provided two alternative approaches which appear at pp 15-18 of his report as follows:
183 The first paragraph on p 16 is misplaced. It relates to one of Dr Hart's other proposed solutions. The first paragraph should read:
The air bag pressure sensor is an (analog) electronic device that produces an instantaneous value for the air pressure in the air bag referenced to atmospheric pressure.
184 Dr Hart was also asked to identify the equipment required for each of his proposed methods. He responded at pp 8 and 9 of the Stage 1C Report as follows:
Method 2 Air-bag pressure method
There are four functional elements needed for this method:
Sensor: A pressure sensor (also called a transducer) is required for each of the air-bags in each suspension that is monitored. In the usual design, air pressure variations cause expansion or movement of a chamber that in turn causes an electrical resistance to change its resistance value. Thereby a voltage variation in an electric circuit occurs.
Pressure sensors that give an electrical output are commonly used on trucks to measure the air pressure in the brake system. In this application the sensor must be able to respond to the rapid pressure fluctuations that may occur in a suspension air-bag. That is, a sensor with an adequate frequency response is needed.
ECM: This is an electronic unit that runs a program. It electrically monitors inputs and thereby learns about the signal level from the sensors. The ECM would receive and analyse the signal(s) from the sensor(s).
For this method the signal is a continuously-varying voltage level. Analysis of the signal is needed to determine the overshoot, damping ratio and the natural frequency. That is, the ECM would probably determine the second-order mechanical system parameters of associated bumps for each monitored suspension.
Alternatively the ECM might determine the frequency spectrum of each suspension over a preset time period. If one spectrum is different to others, a faulty suspension might be indicated.
The ECM has outputs that cause processed information to be transferred to a display or to a computer for further analysis.
The ECM could be a custom-designed electronic circuit or a PLC electronic card. Further details are provided in the answer to Question 29.
If the suspension monitoring system is not permanently installed the ECM might be a lap top computer with an input card.
Operator Interface / Display: This allows the report from the ECM to be seen. Operator inputs such as a reset or request for the history of one of the suspensions might be entered via the display unit and transmitted to the ECM.
The Display might have a liquid crystal display unit or an LED display unit and some switches or lights. Such technology was readily available in August 2003.
Base computer: The data might be transferred from the ECM to an external computer via an output data port. Thereby records for the truck (and potentially for the fleet of trucks) might be stored and made available to the workshop and other interested people. Data transfer could occur via a serial computer port or via a CAN (controlled area network) connection.
185 All of the components were commercially available in 2003.
186 Dr Hart was cross-examined only briefly. At ts 1067 ll 30-45 he agreed that interesting information would be derived from the measurement of forces experienced between the axle and the chassis and recorded over time. He agreed that it might be a useful way of assessing suspension degradation. He agreed that in his report he had identified the need for a "clean bump" in order that his methods might work. He suggested that such a bump might be recognized from a response in the data which was typical of a "broad sharp bump", such as hitting a speed hump. I understood Dr Hart to have meant that a clean bump would generate a response in the suspension which would be identifiable as coming from such a bump.
187 A substantial part of the cross-examination concerned the question of whether Dr Hart's second solution included weighing. Commencing at ts 1072 l 43 the following passage occurs:
Mr Franklin: Now, there is no weighing in this flow chart method you - - -?---No, I haven't shown it, but clearly the weight is - would need to be known.
The weight of a truck would need to be known?---Well, yes.
Yes. But actually when you are taking the pressure sensed in the airbags you don't have to, for example, use the sort of air pressure transducer that one would find in an airway system to convert that into the weight for the purpose of the system. What you need is to sense the pressure and record - sample the air pressure and record those air pressure figures?---Air pressure is being measured and various other parameters are being inferred from the air pressure measurement.
Yes.
HIS HONOUR: What other parameters are being inferred?---Well, the weight, for example, your Honour, on the - that the airbag is carrying would be important.
MR FRANKLIN: Well - - -
HIS HONOUR: So you do infer the weight from the air pressure?---Yes, that's true. Yes.
Mr Franklin: But the actual parameters that you are determining from your sampled air pressure in this flow chart method are defined. You estimate the natural frequency, the overshoot and the damping ratio. Now, for the purpose of doing that you don't need to determine the weight associated with an air pressure signal, do you?---The weight is relevant because the natural frequency, for example, will change with weight. So you should know what the weight is so you can then judge whether the natural frequency is an acceptable number.
Right, yes. I understand that. You should know what the weight of the - carried by the suspension is before you start the test?---During the test, yes.
His Honour: Did you say "during the test"?---During the test.
Mr Franklin: And that - but that would - during the test you would want to know the weight carried by the suspension and that is a weight determined prior to commencing your test. So you know what the weight carried by the suspension is if it is "X" tonnes?---Well, you are assuming that it is the same as when you - before you started the test. It may not be. The truck might be leaning over, in which case the weight on a particular suspension might be different from that on the other side.
What I am suggesting to you, Dr Hart, that in the flow chart that you have set out here what you are doing is sampling air pressure in each of the four airbags, comparing that air pressure and, as we have discussed, looking for the odd man out. This system that you have described here would not, for example, I would suggest to you - could not, for example, be described as an electronic weighing system?---I have assumed that the weight was known.
Yes. And your system of sampling the air pressure in all the four airbags wouldn't properly be described as an electronic weighing system, would it?---I would want to be sure that I knew the weight in each of these suspension airbags. So weird [sic] is an inherent measurement in this process.
Where in this flow chart do you refer to weight?---I don't.
If weight was an important characteristic that you needed to determine for the purpose of this, shouldn't you have mentioned it?---Yes, I should have. I should have said something about weight.
Now you don't need the weight to determine the damping ratio, do you?---No. But you need the weight to interpret that number against any pass/fail criteria.
Well don't you in your system - this system here - all you need is your second order type responses in respect of each airbag, and that will be an air pressure that has been sensed installed?---Yes.
And you look for second order type responses, discard all other data, and then what you do ultimately, having - sorry. Then you look for second order type responses that are consistent with a clean bum [sic], and that is the next block, a simple sharp bum [sic]. And then from those second order responses which are the stored air pressure second order responses, you estimate three things: natural frequency, overshoot and damping ratio?---Yes.
And then you add those parameters to the history file and then ultimately you analyse that and you look for the odd man out?---Yes, but the looking for the odd man out assumes that the weight is equal on each airbag. So it is relevant to know what the weight is. Trucks - trailers, if they lean over, your Honour, if the camber of the road causes them to tilt you tend to get more weight on one side than on the other. So you should take that into account.
You are not in that - you want to know as part of your - the background to estimate your natural frequency, overshoot and damping you will want to know that the weight is equal on all of the airbags. Now, assume that - - -
His Honour: Or that it is not equal, I suppose.
Mr Franklin: Or that it is not equal?---Yes, it is known.
But assume that it now tells you that the weight is equal on all airbags or satisfactorily close to mean that you could - - -? --- Yes.
- - - discard weight variations such as cornering from the consideration?---Yes.
You don't need - you don't then need any weight measurements to make the determinations, do you, to determine the parameters?
His Honour: Well you have got it, haven't you? You have been told they are equal. You haven't - you are just assuming that they are equal is what you are saying. Is that what you are saying?---Yes, your Honour.
Well, you do have the information, don't you? You know they are equal. How can you find that out without weighing?
Mr Franklin: Well, it - - -
His Honour: I really don't understand what you are trying to do here, but anyway, go on.
Mr Franklin: You use - on what you have just told his Honour, you would want to know the weight on the airbags - - - ?---Yes.
- - - before you make the determination?---Yes.
Now having established the weight on the airbags, you don't actually use the kilogram figure, for example, to determine damping ratio, do you? It is not part of the determination of damping ratio?---That is true, however damping ratio changes as the weight changes. So the interpretation of the number you get should be made with respect to the weight that is carried.
188 There was a question concerning the term "overshoot", and then Mr Franklin asked:
And you don't use the weight information for the purpose of overshoot, do you?---Overshoot - pressure overshoot can be measured without weight. It is - I am not really understanding the question sorry.
Well, you say - I think you have answered the question?---Yes.
That you can ascertain the pressure overshoot without the weight.
HIS HONOUR: Well, what is overshoot? What do we mean by that?---It is the extreme - the peak value of the pressure with respect to the average value. So there is an oscillation in the pressure and the overshoot is the first peak value.
I see.
MR FRANKLIN: And you don't use the weight information to determine the oscillation frequency, do you?---No.
189 As I understand Dr Hart's evidence he would sense changes in pressure in the air bags in order to calculate, from such changes, the extent of overshoot, damping performance and natural frequency. Tramanco's preferred embodiment involves measurement of changes in air pressure, followed by a conversion of the measurement into weight. At least in the preferred embodiment, fluctuation in weight is used to determine oscillation frequency, damping ratio and impact loading of the vehicle. The conversion of pressure changes to weight changes may, or may not make it easier to calculate oscillation frequency and damping effect. However such conversion would facilitate determination of impact loading. Dr Hart seems not to have identified any reason to calculate that figure as part of his task. However he considered that knowledge of the load on each vehicle wheel at the time of the pressure measurement was necessary. It is, I suppose, arguable that Dr Hart did not recognize the possibility of using the same information to calculate weight on the axle or on the road surface, assuming that such weight could be accurately measured whilst the vehicle was moving, a problem referred to in the evidence. The Patent seems to assume, rather than provide a solution to that problem.