Raven (non-performance) Maximum Power Output vs 100D

[DavidO]

The dragy uses a technique called integrated doppler(carrier phase) to essentially count the number of gps carrier wavelengths between the start and finish. Carrier phase can be measured to about 1% accuracy of the 20 cm wavelength.This is also what the vbox uses, and this is the description that they gives for their distance accuracy.



Distance
GPS satellites are equipped with an atomic clock, which ensures timing stability to less than one-millionth of a second. By integrating Doppler-derived speed with this level of time signal reliability, an extraordinarily accurate distance measurement is achieved.



There are a number of tests which we have carried out over the years to verify and improve our measurement algorithms for applications in vehicle testing.

One such test is to place two reflective strips on the road at a known distance apart. Using a laser sensor connected to the trigger input of a VBOX, the vehicle is then driven between these two points a number of times, and the distances compared. In such tests, the VBOX 3i will always be within 3cm in 1000m, which is about the same as the measurement uncertainty due to the slight deviation of the vehicle during the driving.

Accelerometers are used but in a kalman filter to basically interpolate between the 0.1 second spacing of the gps data. The kalman filter uses the gps data to calibrate the accelerometers.

Thanks for the info but does Dragy have one antenna or two? It is only with the two antenna version that VBOX3i is able to achieve such precision. Otherwise the resolution is 80cm even with a DGPS base station which is far longer than the <1ft resolution needed to analyze what is happening in that critical first foot. See below copied and pasted directly from the VBOX website:

:
"Accuracy
Velocity and heading data are calculated from Doppler Shift in the GPS carrier signal, providing users with unparalleled accuracy. For example, over 10 tests between two light barriers placed exactly 1000m apart, the VBOX will report a distance within ±2-3 cm, on every test, and the average comes out to within less than 1cm. Impressive for a device using satellites 18,000 km away!

Brake distance accuracy is equally impressive, using a brake pedal mounted trigger input, VBOX 3i will measure the braking distance to within ±1.8 cm for a typical 100-0 kph brake stop. More about GPS accuracy.

VBOX 3i comes in three versions, single antenna, dual antenna and (dual antenna) RTK. In conjunction with a DGPS Base Station (Option 1) VBOX 3i Single Antenna can achieve a positional accuracy of 80 cm."

It would be interesting to find out if the 1ft rollout time subtraction is simply calculated from the acceleration essentially averaged over a longer distance (than 1ft). It would be interesting to analyze Dragy or VBOX results vs a high speed camera aimed at the wheel during launch. Has this been done or are there any volunteers?

 

[TIppy]

Thanks for the info but does Dragy have one antenna or two? It is only with the two antenna version that VBOX3i is able to achieve such precision. Otherwise the resolution is 80cm even with a DGPS base station which is far longer than the <1ft resolution needed to analyze what is happening in that critical first foot. See below copied and pasted directly from the VBOX website:

:
"Accuracy
Velocity and heading data are calculated from Doppler Shift in the GPS carrier signal, providing users with unparalleled accuracy. For example, over 10 tests between two light barriers placed exactly 1000m apart, the VBOX will report a distance within ±2-3 cm, on every test, and the average comes out to within less than 1cm. Impressive for a device using satellites 18,000 km away!

Brake distance accuracy is equally impressive, using a brake pedal mounted trigger input, VBOX 3i will measure the braking distance to within ±1.8 cm for a typical 100-0 kph brake stop. More about GPS accuracy.

VBOX 3i comes in three versions, single antenna, dual antenna and (dual antenna) RTK. In conjunction with a DGPS Base Station (Option 1) VBOX 3i Single Antenna can achieve a positional accuracy of 80 cm."

It would be interesting to find out if the 1ft rollout time subtraction is simply calculated from the acceleration essentially averaged over a longer distance (than 1ft). It would be interesting to analyze Dragy or VBOX results vs a high speed camera aimed at the wheel during launch. Has this been done or are there any volunteers?

They're talking about dgps there. The integrated doppler is done with a single antenna and receiver.

High end VBOX systems used for automotive testing (i.e. Advanced Driver Assistance Systems) can provide up to 2 cm positional accuracy. This level of accuracy is usually not required for mining users – where common VBOX products used on-site are shown below.



Product Velocity Distance Absolute Positioning* Height* Heading Acceleration
VBOX Video HD2 10 Hz 0.1 km/h 0.05 % ±3 m ±10 m 0.3° 1 %
Video VBOX Pro 10 Hz 0.1 km/h 0.05 % ±3 m ±10 m 0.1° 1 %
Video VBOX Pro 20 Hz 0.1 km/h 0.05 % ±2.5 m ±6 m 0.1° 0.5 %
VBOX Micro 10 Hz 0.1 km/h 0.05 % ±3 m ±10 m 0.1° 1 %
VBOX 2SX 5 Hz 0.1 km/h 0.05 % (±20 cm brake stop**) ±3 m*** ±6 m**** 0.1° 0.5 %
VBOX 2SX 10 Hz 0.1 km/h 0.05 % (±15 cm brake stop**) ±3 m*** ±6 m**** 0.1° 0.5 %
VBOX 2SX 20 Hz 0.1 km/h 0.05 % (±10 cm brake stop**) ±3 m*** ±6 m**** 0.1° 0.5 %
VBOX 3i SLR 100 Hz RTK DGPS 0.1 km/h 0.05 % (±2 cm brake stop**) ±2 cm ±2 cm 0.1° 0.5 %

Notice that the vbox micro 10 hz has a distance (not position) accuracy of 0.05%

 

[DavidO]

They're talking about dgps there. The integrated doppler is done with a single antenna and receiver.

High end VBOX systems used for automotive testing (i.e. Advanced Driver Assistance Systems) can provide up to 2 cm positional accuracy. This level of accuracy is usually not required for mining users – where common VBOX products used on-site are shown below.



Product Velocity Distance Absolute Positioning* Height* Heading Acceleration
VBOX Video HD2 10 Hz 0.1 km/h 0.05 % ±3 m ±10 m 0.3° 1 %
Video VBOX Pro 10 Hz 0.1 km/h 0.05 % ±3 m ±10 m 0.1° 1 %
Video VBOX Pro 20 Hz 0.1 km/h 0.05 % ±2.5 m ±6 m 0.1° 0.5 %
VBOX Micro 10 Hz 0.1 km/h 0.05 % ±3 m ±10 m 0.1° 1 %
VBOX 2SX 5 Hz 0.1 km/h 0.05 % (±20 cm brake stop**) ±3 m*** ±6 m**** 0.1° 0.5 %
VBOX 2SX 10 Hz 0.1 km/h 0.05 % (±15 cm brake stop**) ±3 m*** ±6 m**** 0.1° 0.5 %
VBOX 2SX 20 Hz 0.1 km/h 0.05 % (±10 cm brake stop**) ±3 m*** ±6 m**** 0.1° 0.5 %
VBOX 3i SLR 100 Hz RTK DGPS 0.1 km/h 0.05 % (±2 cm brake stop**) ±2 cm ±2 cm 0.1° 0.5 %

Notice that the vbox micro 10 hz has a distance (not position) accuracy of 0.05%

Again, we need to analyze the dynamic performance of the instrument which according to this site:
How Does DGPS (Differential GPS) Work? . is affected by atmospherics. Doppler is affected by atmospherics too:
"The ionosphere will add a relatively large delay, the actual number depends on receiver location, satellite location, time of day, solar flare activity, etc. A lot of those factors are estimated and allowed for, which then corresponds to a delay of up to 16 ns for the signal passing through - note that this is a rough figure, which is changing all the time. This can introduce up to a 5 m error to the captured position.

The second layer the GPS signals travel through is the troposphere. This is the ‘weather’ section of the atmosphere, so includes conditions such as clouds, rain and lightning. This adds a much smaller delay to the signal of up to 1.5 ns, which can introduce up to a 0.5 m positional error.

These delays are random delays which fluctuate. As such, there is no way to precisely measure what they will be at any given time. Each delay is also specific to every individual satellite, as they are positioned at different areas around the world, so their signals will be subjected to different atmospheric conditions."

So again, a convincing argument would be to experimentally verify how accurately the VBOX and Dragy are determining that first 1 ft during very dynamic circumstances. Are you aware of any such experimental validation?

 

[TIppy]

Again, we need to analyze the dynamic performance of the instrument which according to this site:
How Does DGPS (Differential GPS) Work? . is affected by atmospherics. Doppler is affected by atmospherics too:
"The ionosphere will add a relatively large delay, the actual number depends on receiver location, satellite location, time of day, solar flare activity, etc. A lot of those factors are estimated and allowed for, which then corresponds to a delay of up to 16 ns for the signal passing through - note that this is a rough figure, which is changing all the time. This can introduce up to a 5 m error to the captured position.

The second layer the GPS signals travel through is the troposphere. This is the ‘weather’ section of the atmosphere, so includes conditions such as clouds, rain and lightning. This adds a much smaller delay to the signal of up to 1.5 ns, which can introduce up to a 0.5 m positional error.

These delays are random delays which fluctuate. As such, there is no way to precisely measure what they will be at any given time. Each delay is also specific to every individual satellite, as they are positioned at different areas around the world, so their signals will be subjected to different atmospheric conditions."

So again, a convincing argument would be to experimentally verify how accurately the VBOX and Dragy are determining that first 1 ft during very dynamic circumstances. Are you aware of any such experimental validation?

Again, you are talking about differential gps. This is different. They aren't trying to measure position accurately, but only the change in position. Most of the usual complications for measuring position are removed by differencing the signals over a short times and positions.
They note the carrier phases at the start and current position. The difference is then projected onto the local plane of the earth.

 

[DavidO]

Again, you are talking about differential gps. This is different. They aren't trying to measure position accurately, but only the change in position. Most of the usual complications for measuring position are removed by differencing the signals over a short times and positions.
They note the carrier phases at the start and end. The difference is then projected onto the local plane of the earth.

Doppler signals are dynamically affected by atmospherics. Velocity from Doppler is integrated to determine position. Position is therefore also influenced in a dynamic way by atmospherics, unless you are using DGPS, which you confirmed Dragy is not.

 

[TIppy]

Doppler signals are dynamically affected by atmospherics. Velocity from Doppler is integrated to determine position. Position is therefore also influenced in a dynamic way by atmospherics, unless you are using DGPS, which you confirmed Dragy is not.

Its better than dgps. It's using the same antenna and radio to receive the signals. The distortion to the signals change relatively slowly over time and position. The signal processing is similar to dgps, but they don't have to resolve phase ambiguities since they don't care about absolute position. There is still differencing going on to remove any absolute errors. Anyway, their accuracy is still 0.05% at each 0.1 second interval regardless the method. As I said before the data is kalman filtered with accelerometers to interpolate. Even if there is some lag induced, it doesn't matter. We aren't interested in absolute timing only differences from indicated zero mph or 1 ft.

 

[DavidO]

The next software update will include a power increase (of about 3% for the Model S?). It will be interesting to see exactly how the Raven’s new power output curve will compare with the 100D shown here.

 

[kkillebrew]

Why don't you just get another driver and drag race the two cars?

 

[DavidO]

Why don't you just get another driver and drag race the two cars?

Because my 100D was traded in for the Raven.

 

[liludiivert]

Any updates on this?

 

[DavidO]

Any updates on this?

Nothing new I'm afraid. I was told my car is performing "normally". Still waiting (hoping) for OTA peak HP increase to bring car in line with what was achievable with my old (early 2018) 100D. I think I can conclude that the Raven LR has slightly faster acceleration off the line and up to about 40 mph. Above that the 100D is faster (so far) resulting in a Raven 0-60 time of 3.7s (with 1ft rollout subtracted), which is a hair slower than my 100D. I have not retested after OTA 3% power increase but I don't expect anything significant vs the peak power differential Raven / 100D measured. So 0-60 about the same but acceleration from highway speeds (any speed above 40 mph) was more impressive with the 100D.

 

[DavidO]

My suspicion is that my car is faster after the new 2020.20.5 update. Took it for a quick spin yesterday on <30% SOC (yes that low) and the pull at low speeds felt sharper. Just subjective at this point as I have no roads around here to fully put it to the test. Would be interested to hear other Raven LR opinions and even better, actual data.

 

[Krash]

At low speeds the car is torque limited. The SoC should only affect the car at higher speeds when it is power limited.

It is technically possible that the toque limit was increased in the latest firmware. We"ll have to see when we get newer CANBus logs.