Measurement Report

Speed Record Attempt Made by Richard Jenkins in the Yacht Greenbird on March 26, 2009

By Bob Dill 4/5/2009

Summary: Based on the data from the primary GPS I am confident that Richard achieved a three second average speed of 126.2 mph at 18:05 UTC (11:05 PDT) on March 26, 2009. The secondary measurements strongly support the primary one. No deviations from the NALSA Requirements for Speed Record Attempts were observed. The rest of this report reviews how compliance with the individual NALSA Speed Record Regulations was achieved.

1. Wind Power: It has been determined by Bob Schumacher, myself and others by analysis of Greenbird’s design and systems that it meets the requirement that it be powered by wind power only.

2. Stored Energy and Yacht Control: No stored energy was used in the operation of the yacht. All controls were directly operated by Richard.

3. Starting the yacht: The yacht easily started by itself in the high wind of 3/26.

4. Course flatness: Ivanpah, like most dry lakes has an elevation gradient of well under one foot per mile. This is evident from observing the lake when it is partially flooded. My own long distance measurements indicate its curvature is considerably less than the curvature of the earth (about 9”/mile). I am confident that the course Richard sailed is well within the one-meter elevation requirement.

5. Liability Statement: Richard understood and agreed to the conditions of the regulation #5 liability requirement.

6. A measurement plan was submitted to and approved by NALSA in January. It was used for this trial. A copy is being emailed is attached to the same email this report is. Measurement details covered in this regulation are discussed in detail below.

7. Observers: Bob Schumacher, Dennis Bassano, Chris Augustin and Phil Augustin served as official observers. Bob is a NALSA board member and Dennis is the president. Approximately 100 additional people were present. See the Spurious Data section below for more details. A scan of Dennis’ hand written observers report is attached to the same email this report is. Chris Augustin’s email is a second observers report will be forwarded separately.

8. Regulation 8 requires report on measurement methods, calibration, commentary on spurious data, deviations from the measurement plan and the data itself. All of this is discussed below.

9. One mph Speed Record Increment: The speed Richard achieved is 9.5 mph faster than the previous record: well in excess of the one mph increase required by the NALSA Regulations.

Sailing Conditions: The wind direction was northwest at speeds at 30 to 40 mph with a peak gust of 47 mph recorded by Dennis. This allowed Richard to make good use of the long axis of the lake. He chose to stop sailing for maximum speed after clearly achieving a new record in order to minimize the risk of breaking the boat. Runs made after 11:05 were mostly for the purpose of getting better video footage. On one of those runs he came within a few mph of eclipsing the old record again.

Measurement Method Discussion: GPSs were used for all measurements. Since the NALSA Regulations were last updated nine years ago the capabilities of sub-$200 GPSs have improved dramatically. In my opinion the best of these is the Lycosys GT 31. It logs NMEA data at one hz onto an SD card: no more worrying about the batteries dying before transferring the data. It calculates velocity by doppler calculation: the preferred method for velocity. With a SiRF Star 3 High Sensitivity chip it has excellent satellite reception and 24+ hr battery life.

The GT 31s do not appear to have significant low speed filtering which allows static measurements. Based on static measurements I have estimated a velocity measurement uncertainty at 95% confidence of 0.1 mph: five times better than the NALSA 0.5 mph requirement for accuracy. These static measurements serve as a calibration check. (If you are interested in the details please contact me.)

There were three GT 31s and one GT 11 (predecessor to the GT 31) on Greenbird. Three of them (#5,7,8) were provided by me and were configured to log NMEA data ‘sentences’ GGA an RMC to provide satellite signal quality data and velocity, position and bearing data. The GT 11 (Wingjet4) belongs to Richard and was configured only for RMC. The NALSA Regulations specify NMEA sentence VTG in addition to GGA. VTG provides the course direction in true and magnetic north as well as velocity in knots and kph. I chose to substitute RMC as it provides velocity, lon/lat and bearing like VTG but also time, and date, making the data analysis less cumbersome. I will recommend this and other changes to the Regulations at a later date.

In addition to the four GT 31/11s, there was a Drift Box made by Performance Logic. It is a 10 hz GPS designed for automotive performance measurement. The Drift Box records data in a proprietary format that can be exported as a .csv file for analysis in a spreadsheet. The Drift Box was configured during the runs to only record data when it was moving so there is no static data in the data file. I measured static data from a Performance Box in my back yard on 3/31. I believe it has the same GPS engine as the Drift Box. I found it has a 0.5 kph cut off filter making static measurements of little value. The Drift Box is an excellent device for what it is designed for however it lacks some of the features that are desired for a primary measurement device under the NALSA Regulations.

Deviation from the Measurement Plan: The Measurement Plan used for this trial called for the Driftbox to be the primary measurement device. In light of the limited satellite signal quality information (no hdop) and the inability to obtain unfiltered static data for an assessment of the velocity measurement uncertainty, I am recommending that the NALSA Board accept the following amendment to the Measurement Plan: Use my GT 31 #5 as the primary with all the others, including the Drift Box, as secondary measurements. Richard has agreed to this change. As discussed above, the GT 31s are fully capable as primary measurement devices. Unit 5 was chosen as it has slightly better satellite signal data although they all are very good. All five GPSs agreed with in 0.1 mph so this proposal has no meaningful effect on the outcome but follows the intent of the Regulations better.

Spurious Data: Spurious data generally comes from a poor view of the sky or weak batteries. The GGA sentence provides direct evidence with the number of satellites in view and the hdop. The primary GPS (#5) showed 10 satellites were in view and the hdop was 0.9 (an excellent value) for the 180 seconds it took to go from 25.28 mph (at UTC (Greenwich Mean Time) =18:04:00) 126.36 and back to 37 (at UTC 18:07:00). Low batteries were not an issue. The batteries were well charged at the beginning of the session on 3/26 and still showed a strong charge a week later.

‘Dropouts’ are lines of data where the GPS missed getting complete data. Typically velocity and bearing are missing from the RMC data. If dropouts occurred on the primary unit within a few seconds of the fastest speed they could cast doubt on the validity of the fastest speed values. At start up all GPSs have 2-5 minutes with dropouts as the device finds itself. The three GT31s (including the primary) showed no dropouts after a three minute start up period. The GT 11 had 42 dropouts in 11670 lines of data (0.4%) after the start up period. The Driftbox collects data at 10 hz and was configured to only record data when moving but still recorded 28807 lines of data. There were 551 dropouts (1.9%) according to the data analysis software. Almost all of them were late in the day, long after the fastest run. All the units are drop out free for the 180 seconds of the fast run. Conclusion: Dropouts are not a concern.

The Regulations require an evaluation of the risk of deception in the measurement process. This run was formally observed by four people and informally observed by roughly 100. No fast vehicles were seen on the playa that might have towed Greenbird or taken all the GPSs for a fast ride. The official observers watched and photographed the starting, resetting of the maximum speeds and placement of the GPSs. The same process was repeated when the GPSs were removed (see email from Chris Augustin dated 3/31). Three of them (#5,7,8) were packed in a secure way and delivered to me by Bob Schumacher on 3/29. I was the first person to have access to the data files on these units from the 3/26 runs. It is theoretically possible to create a fake GPS log file and place them on the GPS data card surreptitiously. This would be very difficult to do well enough to look authentic and there is no evidence of such shenanigans or deception of any sort. Finally, Richard is well known to many of us. His saga is well chronicled on the Windjet and Greenbird websites. It has been a patient, methodical quest with steadily increasing top speeds.

Data Details

The following table shows the details on the fastest three seconds

Note: The Drift Box numbers are an average of the 10 readings for each second.

The excellent agreement between all the units for speed and time offer strong support to the primary measurement of 126.19 mph. The Regulations call for that number to be rounded to the nearest 0.1 mph so the official number should be 126.2 mph.

The following shows the second by second accelerations. This is an excellent way to spot abnormal data. The data looks typical of many other acceleration graphs and does not show any problems (values over 1 G are suspect for landyachts). The waggle between 110 and 120 seconds appears to be a moderately strong application of negative angle of attack followed by two compensating adjustments. The yacht was slowing from 90 to 70 mph through this few seconds.

I have attached the text file from the primary GPS to the same email this is being sent with. The text files

from all five GPSs and analysis files (Excel workbooks) are available from me. Please contact me with any questions, suggestions or comments about this report or the data it presents. Google Earth based pictures of the GPS tracks are shown below.

Conclusion: Richard has complied with the NALSA Regulations for Speed Record Attempts. The measurement data is very complete and high quality. I recommend that the NALSA Board ratify a new world record for speed in a land yacht of 126.2 mph.

Bob Dill

All 3/26/09 Greenbird Runs

Fastest Run by Greenbird 11:O4-11:07 PDT 3/26/09