Select to expand quote

AUS02 said..
Here's my best windsurfing alpha from Lake George (28.195 knots, with a minimum speed of 17.903 knots) and below that my best foiling alpha (21.522 knots, with a minimum speed of 14.423 knots) - still trying to get the hang of them! Certainly easier to keep the speed during the gybe while foiling and quicker to accelerate back up to full speed.





Also just noticed that Boardsurfr mentioned the Speedreader software shows both minimum speed and percent of entry speed kept at the bottom of the results table - these are mine for slalom (top) and foil (bottom), so as high as 60% for slalom and quite a bit higher (over 70%) for foil (for those two sessions):

Nice alphas! But keep in mind that the minimum speed will always be lower if you record at higher rates, as you did for foiling. If you are at the minimum speed for just a fraction of a second, the 5 and 10 Hz devices will get he "true" minimum speed, while the 1 Hz data show a 1-second average. GT-31 data are also filtered so that each data point is effectively averaged over more than 1 second.
With 5 Hz and 10 Hz data, you'll usually have several points were the true speed is very close to the minimum speed. But there's random noise of about +- 0.5 knots added to the "true" speed, which means that there's always a good chance that one of those points is about half a knot lower than it should be. Note that the same applies to the entry speed, where a single point value is likely to overstate the actual speed.

As a result, the minimum speed and percentages with a GW-60 or Motion will often be lower than what a GT-31 reports for the same turn. I've thought about changing the reported min. speed and percentage to use 1-second averages, which would be less prone to noise and a bit more meaningful, but have not done that yet. Playing with new hardware is more fun .

That said, those are some very nice u-turns on the speed gear and on the foil.

Select to expand quote

sailquik said..
Thats a truly outstanding Alpha Dave! Thanks for posting it.

It is very interesting because your out-run top speed was higher than your in-run. Do you remember if the in-run was tighter angle to the wind than the out-run?

And the bottom speed is exceptional too!

What board were you on?

I just released that the first one looks like a 1Hz GT-31, and the Foiling one looks like a 5 or 10Hz device.

Thanks Andrew, That was up in the corner at 3-mile, sailing in as though you'd be getting ready to accelerate into the flat water for a speed run. Wind was strong and so was able to put quite a bit of pressure on the fin on the way in and point slightly to help ensure I was in the proximity circle on the way back out and allow me to really make the most of the wind and flatwater on the way back out. Did quite a few alphas up in that area and that was the last and fastest one. Have circled that alpha (purple) and just below it was my top 2-second run for the day of 41 knots (completed only a few runs earlier than the alpha), so that alpha was done where the flattest water and windiest conditions were for the day.




I was on my 2010 wood iSonic 86, with a C3 Slingshot 23 fin and 5.6 Severne Overdrive using a GT31: (gpsteamchallenge.com.au/sailor_session/show?date=2015-01-29&team=27)

For the foiling session, it was a Motion Mini.

Boardsurfr, good point about the different GPS's capturing different minimum speeds at 1 or 10Hz!, so that would be good to have the 10Hz averaged to 1 second for that category.

I disagree with Boardsurfr about his use of the term 'Noise' and his contention that is is random in the higher Hz gps tracks. Tests I have done with side by side Mini Motions show surprisingly pretty good correlation of the 'sawtooth' in the speed graph. One in particular shows very clearly, and with great correlation, the change in speed in the fraction of a second when I changed gears during acceleration tests.

I don't think it is entirely random 'noise'. The correlation is not perfect 100% of the time, but it is certainly not random.

My tests were done with the two Mini Motions on my helmet.

It would seem logical to me that when you are sailing the receiver is bouncing or vibrating to some greater or lesser extent. The sawtooth we see could very well be the real manifestation of the micro speed changes that this produces.

Of course, I agree that that movement/vibration is somewhat random, and not a true indication of the horizontal speed one is travelling in that respect. So in that sense it could be correct to call it 'random noise' (qualified). But I think what the receiver is recording is very, very close to actual receiver micro velocity.

So in that respect, the minimum speed data produced by a 10Hz GPS, worn on the head in particular, during an Alpha should be a very close indication of the true minimum speed. Smoothing it to 1 second would probably be a very poor indicator of actual minimum speed.

Of course, if the GPS is worn on the upper arm, there will be more movement off the path of travel (board and sailor mass) by the GPS during the gybe, and more sudden change of antenna orientation, and I think we can see this in some speed graphs quite clearly. Data from a GPS worn on the wrist is going to show this extra movement and orientation change to a much greater degree again, and is certainly not ideal for the accurate measurement of speeds during the gybe in manoeuvres like the Alpha.

Here is the example of the sub second gear changes from the Motor Bike test:

Acceration test:


The 4 small 'steps' in the speed graph are the gear changes.

Notice that is the speed increases, so does the 'sawtooth'. That would be a logical result of increased vibration at higher speed.

Here is an example of part of a high speed graph:

I noticed that in this test, the red unit data always seemed to be a bit smoother. It could well be that the blue unit was 'rattling' around a bit more in my helmet. But it could be any number of other things as well.

I don't really like when fun park is combing to the town , but I have one experiment in mind that I want to make one day.
Get on the biggest carousel with my Motion GPS and check if that plotted tracks are perfectly round and even each time .
Can't come with other idea to have source of reliable , repeta ble circular motion for the test in mind.
Probably the model plane on the leash could serve similar purpose, but in windy condition speed may vary.Another nice experiment can be done with big passenger wheel - in this case we could observe how precise our device is when velocity is mainly along vertical axis.
If somebody has now access to one of those , will be interesting to see how the graph and track presents itself on our software.


BTW. the first question comes to mind if the track of the cabin is perfectly round circle or not really ( due to hinges and motion) .?

Altitude is far less precise on GPS than horizontal position. It is often many meters, or scores of meters in error.

But it's not positional accuracy you want to test anyhow. It's Doppler velocity. I seem the remember Tom Chalko telling me that one or more of the Locosys GPS were, (or could be), fixed into 2D mode for velocity calculations, ie, the assumed altitude was fixed at sea level to eliminate the vertical velocity errors. You experiment would soon prove if that were correct, and in which devices.

Select to expand quote

sailquik said..
Altitude is far less precise on GPS than horizontal position. It is often many meters, or scores of meters in error.

But it's not positional accuracy you want to test anyhow. It's Doppler velocity. I seem the remember Tom Chalko telling me that one or more of the Locosys GPS were, (or could be), fixed into 2D mode for velocity calculations, ie, the assumed altitude was fixed at sea level to eliminate the vertical velocity errors. You experiment would soon prove if that were correct, and in which devices.

all we know about those big wheel , horizontal and vertical: almost perfectly circular and constant speed, over smaller time increments. So any small imperfections on our graphs could be attributed to our GPS device and system. We could see instantly sudden errors , spikes, dislocations, s****ing? Like sudden 5ktn spike on our max speed could make somebody happy, because it is still probability to happen on the water. But here on the wheel it is clearly mistake. I saw already posts onGPSTC with very suspicious spikes in speeds. A bit unfair , once you know that somebody beat you by 5 knots due to error recording.
On the open water is less likely , but here on GC, if high building, moored houseboat, speed boats could provide echo signals to spof Gps?

5 knots spikes!!

Old news Macro. Almost unheard of now we use Doppler data with speed error verification. If you see any posts that you suspect, please send them to me.

I got a couple of Nice Alphas the other day at LG.

Here is the first one where I go off balance on the foot swap/boom flip. You can see the slower, delayed acceleration out caused by this in the speed graph and in the video you can hear my angst when I knew I had stuffed up. But surprising to me it was still better than the second one below.



First alpha video:


Here is the second one (first one on the video below). This one felt OK but the accelleration out was still a bit slow and I actually missed the 50m proximity circle by a few meters. I had to expand the circle to 55 m to get this to register in RealSpeed, so it was not legitimate at around 26.6 anyhow.


Here is the 3rd and best one @ 27.36Kt.(the second on the video below) This one felt very nice, the min speed was highest and the acceleration out and end speed was also the best.


Video of the second and third Alphas:

Very nice Andrew!!

Also great colour coordination with board and sail

Yes nice alpha Andrew. Would have thought you'd have gone for a wider board for max Alpha numbers in those conditions?

Select to expand quote

Dezza said..
Also great colour coordination with board and sail

LOL! That was a very happy fluke! And I love it!

Select to expand quote

Ian K said..
Yes nice alpha Andrew. Would have thought you'd have gone for a wider board for max Alpha numbers in those conditions?

This is the combo of speed and sail I can get fastest speeds (30-32Kt) crosswind on in 20-25 knots. and it still holds speed very well in the gybe and gets good acceleration.

A wider tail can carry speed through the gybe better, and accelerate quickly, but is usually harder to turn in at top speed and may not be quite as fast crosswind. there are always compromises.

i can go slightly faster crosswind on the M47 speed board, but it loses much more speed in the gybe, foot work is more critical and acceleration is slower.

My best Alphas were done on the older IS87 which was actually narrower, and with a 5m sail in 25-28 knots. But I find it easier to get those numbers in the newer board and in this wind strength. One of the disadvantages of trying to do alphas in 30 knots of wind is that you dont get the sail as neutral in the gybe, (the real wind speed is closer to your speed in the gybe) which make the sail flip window shorter and more critical. I think a more highly skilled sailor who can cope with that should be able to take advantage of the stronger wind for higher in and out speed though, and better acceleration.