Unlimited board speed test and SIC Standamaran review

[PT Woody]

Interesting that in Time 1 times, the Standamaran is just about quickest in the field but in Time 2 times, it is slowest. Surely an upwind/downwind thing?

[PonoBill]

I'm going to put a fin box in my Bullet, but I'll put it behind the rudder, because then when the rudder gets hit the trailing edge drops into the fin box instead of punching through the skin.

[maui_husky]

I noticed the same thing (slower times on the 2nd run).  I wonder if it has something to do with it's extra weight.  Maybe more strength required to get it going?

[blueplanetsurf]

Wow, thanks for all the great feedback, I'm taking notes for the next test.  I agree that more runs are needed to get meaningful data and will try to include more data, like board weight, price (I like the idea of speed per $) etc.  We originally planned to do two rounds of testing but ran out of steam after doing 12 sprints.

Yes, Run 1 times were with the wind and Run 2 times are going back upwind, so that's why Run 2 times are slower.

Regarding which boards we are used to, these boards are usually used/ owned by:
Jared: Ohana
Anders: Bark
Robert: Pang

Here are some of my thoughts:
I expected the standamaran to do well upwind with the smooth entry but in the test it did not compare well in the upwind legs.  Why?  I'm not sure but my theory is that the wakes coming from both tips and intersecting at the center of the board create a wave that adds drag at higher speeds and limits the top speed.  Going into the wind the small chops might exaggerate  this effect.  I'm not sure though, just a theory.
At normal speeds (not sprinting)  the standamaran seems to have very low friction and it takes very little to maintain a speed of around 5 mph.

All the boards have pros and cons and which board will be fastest depends on the paddler and the conditions.   So why were some boards faster than others?  There are so many variables and to try narrow it down to just the width is just not realistic even if the numbers seem to indicate that.  I have tested two 12'6 prototypes with identical length and width with the main difference being the rocker and entry and the board with more rocker was actually faster and had a cleaner entry.   Regarding length, I know that most 14' boards are significantly faster than most 12'6 boards and that most unlimited boards are faster than 14' boards but at some point (over 16' it seems to me) adding more length does not always translate into more speed.
Shaping a fast race board is more art than science, I think.  
Paddler weight is important too, as the same board will have a different entry and exit depending on the weight of the rider, so the rocker line and volume have to match the rider weight.

[PonoBill]

All catamarans suffer from hull-to-hull wave interference, the closer the hulls are spaced, the worse the interference is. I talked to Mark about this when he was building the Standamarans. He said he hoped the low friction from the very narrow hulls would overcome the effect. I don't think it can. In racing catamarans the hull to hull interference represents 20-30 percent of the total resistance to forward movement. The main reason racing cats have such wide platforms is not hiking moment, it's hull to hull interference. Interference is exacerbated by shallow water. The standamaran needs the hull spaced closely so it can be paddled.

It's a lot faster than I thought it would be.

[blueplanetsurf]

Ok, we just finished another speed test and was hoping to get some help with analyzing the data- Area 10 can you do the multiple linear regression analysis thing again and give some feedback?

John Amundson asked me to help him test two new 14' prototypes (called V1 and V2 in the spreadsheet) for next year's Aquaglide lineup.  My friend, Scott Dodds came with his  new 14'x 26" Starboard ACE, a proven race shape, which we used as a comparison board.
Without analyzing the numbers, we all felt that the speeds of the three boards were very close.  The V1 and V2 are significantly more stable than the 14' ACE and the V2 is the more stable of the two prototypes.  John is really getting the race board shapes dialed in and Scott and I were impressed by how well they worked.  He is planning to make a couple more prototypes to test and whichever one he chooses to go into production is going to be a great allround race/ touring board- stable, user friendly, and fast.
We did three runs each, one on each board, one upwind and one downwind leg per run.  The wind was gusty, which is why run#2 was slower for all of us than run#1, so that should be taken into consideration as well.

I know it would be better to have more numbers but it was getting dark and we were beat after the 3 rounds of sprints.

Here is the spreadsheet, click on the second tab on the bottom (14' test) for the results of todays test.
https://docs.google.com/spreadsheet/ccc?key=0Au9qxAnW7ZMddE8ycGlSNWxraDZBM0FBRlZZYmRtMGc

sorry, no pics or video this time, just numbers.

[Area 10]

Hello Robert. Thanks for giving me a reason to have a tea break (hey, I'm English, we don't drink coffee....much).

Again, I am very impressed with the time and effort you have gone to here, and that you are sharing the data like this. It is remarkable to see someone trying to go beyond the usual speculation and hearsay and actually try to put numbers to what is going on.

You will no doubt have already just looked at the average times across all paddlers for each board. The times for the ACE were about 3% faster than for the V1 and 2% faster than the V2. BUT it would be quite wrong to conclude from these data that the ACE is fastest. In fact the boards were so close in performance on this course and with these paddlers that these differences could just be a chance finding.

If you rank order the effects that the various factors of paddler, board etc had on the times of the runs, you get this (most important factor first):

1. Upwind or downwind - a huge effect dwarfing all others, about 10x as important an effect as any other.
2. Which run someone was on, first, second or third (influence of about 5% or so on the times), presumably fatigue.
3. Which board you were on (in the region of 3-4% difference).
4. Who the paddler was (about 1%, you are all remarkably similar in your abilities, eerily so in fact).

BUT, with so little data, the only effects that would even begin to be accepted as noteworthy by a scientist were the first two. I'm especially suprrised that the effect of who was paddling the boards was least significant of all. I have paddling buddies who are very close in their abilities, and there, on any given day someone usually has a small but clear advantage over another - even if the next day it is another person.

(As an aside, I should point out that something of course is a bit odd with your (Robert) timings, which seems to be in multiples of 5 rather than on a continuous scale like for the other two paddlers.)

I would suggest that, if you really felt there was a difference between the boards in e.g. stability etc, you could improve these test results in terms of deciding relative performances by doing the following:

1. Include heart rate data for each run. for each person. This would allow me to determine how hard each person was working. Over longer runs this would become important.

2. Test in different conditions - do both flatwater and choppy water tests.

3. Use a much longer course.

4. Paddle to exhaustion (yeah, I know - but when you are really so tired your legs are like noodles, you'll probably be a lot faster on the more stable board, and this will also show in heart rates. Might be the best test of all for how well the designer has understood a paddlers' needs)

5. I am bothered, as I say, by how similar you are all in terms of your performances. This makes me wonder if there is what psychologists would call a "group effect" going on. For instance, do you all paddle at the same time, side-by-side (or leaving at set intervals in a line)? There is nothing more motivating than seeing your buddy inch ahead of you, and equally, it is hard to lead. So sprinting side-by-side might have the effect of reducing the differences between paddlers. Again, HR data might help. But better would be to do each run only one paddler at a time, and with the person doing the timing standing on ground (out of line of sight), and not the paddler timing himself.

As I say, I do admire the amount of work and dedication you have put into this, which everyone on the Zone will I am sure appreciate. It is a hugely complex thing to test which board is "fastest". In fact, I think to do it to the full satisfaction of a scientist would require many weeks of work, many people, and therefore many thousands of dollars.

And at the end of that time I suspect we already know what the outcome would be. At a given length, the board that looks most like a surf ski/olympic kayak/{insert example of narrow tippy but fast paddle craft here}, would be the fastest.

But only for the most elite paddlers who could balance on it.

For the rest of us, in real world conditions, stability will always be an important determining factor: if the difference in theoretical speed between the fastest 14 race board and the slowest is of the order of 5% (which seems likely), then all that needs to happen is that in chop the person on the fastest board has to make 1 in 20 of every stroke a correction stroke, and that advantage is pretty much all gone.

So, maybe a fairer test if a board has been specifically designed to add a bit of stability, is to test it in conditions where stability will affect performance postively. So whilst these data might seem to show a very slight advantage for the ACE, if it is indeed the case that the V1 and V2 are considerably more stable, at only a marginal cost to pure flatwater speed, personally, I think I'd go for that since I'm not likely to be paddling in any dead-flat lakes or rowing courses, and I'll always be paddling a lot further than 0.21 miles.

Incidentally, I think that once the younger paddlers start coming through in numbers, us older geezers will be dead in the water. Perhaps literally. This is because balance ability declines very sharply after the age of 40, and at some point this is going to limit our performance relative to a younger paddler, especially as relates to the equipment we can use. So we are just able to fool ourselves at the moment that we are doing OK because there are so few youngsters at a typical race, and they don't have to opportunity to train like we do...and can't afford the best equipment...etc.

Anyway, thanks very much for these data. And do please share SOME details of the boards with us. At least you could tell us the dimensions and weights of each. Is the ACE narrower?

[Argosi]

Robert and Area10, thanks very much for your efforts. Not many others doing this kind of stuff.

I did a speed test last weekend with 2 of my boards: a Starboard NEW 12'6"x 23.5" and a custom Joe Bark 14'x25.5".

I had a third board but wanted to focus on 2 at a time to reduce the complicating effect of fatigue.

Here's my 1 man test protocol for 2 boards:
- Wait for a calm, no/low wind day.
- Use GPS mounted on the deck of the board
- Wear a heart rate monitor.

- Paddle Board 1 500m upwind. Exertion level such that I hit about 90%+ of max heart rate by about 2/3 of the way to the end of the upwind leg.
- Turn off GPS between each 500m leg.
- Rest briefly while turning board around and start 500m downwind leg when HR reaches 80%. Paddled so that my HR reached 90%+ about 2/3 of the way to end of downwind leg.
- Switch to Board 2 and repeat.
- That completes 1 round of testing of Board 1 and Board 2.

- Repeat 4 rounds of testing. That results in a total of 8km of paddling: 1km round trip x 2 boards x 4 rounds.


Results: While the differences where not huge, I could see a consistent speed advantage for the Bark. Before the test, I suspected that my Bark would be faster so I started the test paddling my Starboard. If my Bark was still faster, it would mean that it was faster despite having a slight disadvantage of more fatigue, making the conclusion more certain.

If you deviate from the protocol, it makes it harder to isolate the actual board speed versus other factors.

For example, if you rest longer before a particular run, that run should have a slight advantage due to less fatigue at the start. If there is a change in wind speed for a given run, that run will not be directly comparable to the others. If you work extra hard or relax too much on a run, the results will be affected (that's why a HR monitor is helpful in keeping you honest).

It was an interesting way to make my morning paddle different from my usual route.

In windy or downwind conditions, it would be much harder to get meaningful data on board speeds since there are so many more variables that enter the equation, such as varying wind and water conditions. In downwind conditions, all I know is that my Starboard is more fun to paddle than my Bark even though I'm not sure which one is actually faster downwind.

[1paddle2paddle]

Area 10, since you inquired about heart rate during the runs Robert posted about, and since I happened to have my trusty Garmin 305 on me for the event, I can provide the data I downloaded.  I can also state that it felt that my heart was working much harder on the upwind runs than the downwind ones, but the data doesn't exactly confirm that.  Also, I have times for 9 runs, and Robert only has 6 runs on his spreadsheet, so you have to match up the heart rates with the runs on the spreadsheet by the times given.

1(up):     avg = 161, max = 173     time = 2:13
2(down): avg = 164, max = 177     time = 2:05
3(up):     avg = 171, max = 176     time = 2:43 (not on spreadsheet)
4(down): avg = 162, max = 174     time = 2:14
5(up):     avg = 171, max = 178     time = 2:30
6(down): avg = 159, max = 173     time = 2:09
7(up):     avg = 163, max = 173    time = 2:27
8(down): avg = 152, max = 169    time = 2:09 (can't tell if this or the other 2:09 time is the one on the spreadsheet)
9(up):     avg = 163, max = 174    time = 2:32 (not on spreadsheet)

After that 5th run of 2:30 it felt as if my heart almost came out of my chest after finishing, and I needed to rest for thirteen minutes before resuming the trials (my HR spiked at 213 during that particular rest period).  But Robert appears to have been nice enough to throw out my two slowest times.

Also, your comment about group effect was interesting.  From a boots on the ground perspective I can say that the runs are short enough that I can almost go full out chasing Robert, and I think having him just out in front really pushed John and I harder to try to keep up.  I can tell you that on longer downwind runs after about 15 minutes Robert is so far out ahead that the motivation to try to catch him is fading.

And the ACE is 14' x 25", and I believe the protos were 28" and maybe a little wider on the other one?  So yes, the ACE was significantly narrower.  But I was truly surprised at how fast the prototype boards were; they did not feel slow at all compared to the ACE, even though the ACE "won" the test if you could call getting the best overall time winning.  So I agree with your assessment that for more all around fun, the protos would be great choices and would be stable enough to take on some some fairly windy conditions.  I got the ACE specifically for flatter-water training, so I'm happy it wasn't SLOWER than the other boards...!

[PonoBill]

I've been playing around with my dragged speedometer. the idea is to have something kind of like a pacing attachment to a bicycle. A GPS does a good average speed measurement but a pacing recorder gives more instantaneous numbers. My dragged speedo is a 1/4" carbon fiber shaft with a bearing block on one end, small prop on the other, and a couple of magnets on the shaft that trigger a hall effect sensor that the data recorder counts. I've been dragging it around behind my glide trying to calibrate it.

The challenging part is that because it samples 15 times per second it gives a whacky result. sometimes in a time slice it could no magnets, sometimes one and sometimes two. I want to be able to get fairly instant values for speed because I want to measure paddle impulse, work and force applied, and how much the board slows between strokes. I should be able to derive all these from speed, acceleration, and mass of the rig. But since the speed isn't very high the RPM of the shaft isn't enough to give me enough pulses for a smooth speed measurement. I tried a lower pitch prop, but I think it slipped too much. I'm going to try using a disk to put the magnets on so I can get a dozen or so going by the detector each revelution instead of the just four I'm using now.

There has to be a rational way to do this. Maybe I could count for longer periods and overlap the start for the counts. For example, suppose I started a count with each magnet going by, with ten magnets on the shaft, and counted ten pulses and measured the interval, but then started a second ten count when the second magnet came by, and so on. So for every two revolutions I'd have ten intervals completed and ten more started.

Or am I just over-complicating this and missing something simpler?

[ObviousSup]

Or am I just over-complicating this and missing something simpler?

I don't know if one of these would connect with a data recorder but it sounds close to something that might record the speeds you need.
http://tinyurl.com/speedwatch

Some of these options look more interesting.
http://www.nkhome.com/rowing-paddling/
The speedcoach seems to be well regarded in the rowing community from what I have read.

Maybe checking with Ron and see what he used since he talks quite a bit about paddle Efficiency?
http://www.frontrower.com/efficiency.htm

Or is this all old news? I've tried reading all the back posts trying to avoid the "Teaching grandmother to suck eggs" problem.

[blueplanetsurf]

Thank  you all very much for the feedback and suggestions.  Scott's GPS readout indicates another variable- human error.  I did not mean to drop your slowest times although I did not count the first upwind run, I considered it a warmup.  Anyway, we had to memorize the first time and I recorded the up and down times at the end of each run as they were called out to me (or as I heard them, in dimming light, then entered them on the spreadsheet after a few beers, so some errors may have snuck in)
I agree that the group effect can be a big factor.  I'm always faster when I test with others than solo, especially if someone is a little faster than me.  Doing a staggered start is a good idea and independent timing, more rounds and GPS with heart rate monitor for each tester.

I'll send this to John and let him decide how much he wants to share about his new boards.

[Area 10]

Argosi:
Bravo. Very interesting, and it takes perseverance and an enquiring mind to actually do these things rather than just think about doing them…

But I hope you don’t mind if I make a suggestion about your experimental design. You explained your rationale for always running the Bark in the second leg of a round. I can understand your logic entirely. But if I were analysing the data I would have a real problem, because the way you have done it you have two perfectly correlated but potentially separate factors (board, and position of run in the sequence). If you want to be able to analyse these kinds of data properly, you need these factors to be what statisticians call “orthogonal”, i.e. completely unrelated. In practice, the best way for you to have done what you did would have been to use, for instance, an “ABBA design”. This is nothing to do with ‘70s Swedish pop music, but instead means that you would run board A first, then board B second, then board B third, then board A last.

If you are doing lots of runs, I’d suggest that you try to ensure that the average position of each of the boards in the overall run sequence is roughly equivalent. So if you do several (let’s say 2 for example) ABBA quadruplets of runs, for instance, it would look like this:

Run 1 = Board A
Run 2 = Board B
Run 3 = Board B
Run 4 = Board A
Run 5 = Board A
Run 6 = Board B
Run 7 = Board B
Run 8 = Board A

Now, if you add up the serial positions for each board and each run, for Board A you get 1+4+5+8= 18 (mean position 18/4 = 4.5). For Board B you get 2+3+6+7=18 (mean position 4.5). This means that if you find a difference in performance between the boards when you look at the data overall, you can be pretty sure that it is NOT going to be because of fatigue effects. And if you use fancy statistics (or someone does it for you) they would be able to “factor out” the effect of fatigue when looking at your data. I can see from what you have said that you foresaw this issue, so I’m just suggesting an easy way to deal with it.

Of course, the other point to make is that you went into the test expecting the Bark to be faster. That expectation can make enough difference for it to become so. But that’s a story for another day: I don’t want to put you off from doing this because I find it fascinating, and the Zoners who read this are lucky that there are people like you who will go to this trouble rather than just believe what someone tells them.  So whatever the shortcomings of any one test (and there will always be something...), they are a hell of a lot better than mere speculation.

1paddle2paddle:
HR spiked at 213 during rest - Wow – you must be a lot younger than me. If my HR went that high I’d be in the ER! Thanks for the HR data, I’ll have a look at it shortly.

Of course, HR can be deceptive, since it speaks more to the aerobic than the anaerobic component of the effort – I am not an exercise physiologist, but I think that this is why paddling upwind can be very tough, but your HR can still be fairly low whilst doing it. It is why pro trainers often use “perceived effort” scales in training – the HR only tells one story. But it is a lot better to have it than not.

And thanks also for raising the issue of “data laundering” by Robert – very kind of him, but perhaps not ideal from the point of view of the stats…especially since the boards were so close in performance. It does go part of the way to explaining why you all (according to these data) are apparently of such similar ability though. The rest is presumably the group effect, since both you and Robert are acknowledging that there is something going on to that effect in the format you have used so far.

And your comments seem to bear out my speculative suggestion to use longer runs (sorry!) I know it is cruel to see your buddy leave you behind, but from the point of view of testing the boards we need to be able to separate out the “variance” (as the statisticians would call it) attributable to the paddler from that attributable to the board, and this generally works best if one can detect consistent differences.

And if a 28” board is within 2- 3% in performance to a 25” board, and is much more stable because of that extra width, then that sounds like a pretty good compromise to me. My balance isn’t anything special, and I paddle all sorts of conditions, so I’ll take the more stable board please, and run the person on the 25” down in the final sections of the race when their legs are getting tired, or when we hit chop!

PonoBill – I am not a marine engineer, but I think your problems may just stem from a sampling rate of 15/sec. I would have thought that you might get closer to what you want with a rate of about 100/sec or more (and/or more magnets – dunno your rpm), and then use some data smoothing to reduce to say 1/10th sec resolution. But I’m sure you have thought about this.

Robert – thanks for being so honest. That explains a lot. It’s VERY hard to do these things perfectly, and heaven only knows I’ve spent whole days doing comparisons myself…paddles too…so I do know what you face. But if you REALLY want to know what makes you fastest, it is, as you clearly appreciate by your actions, the only way to know for sure. So stick with it. It’s so kind of you to share these data with all of us, so that we can learn too.

[LaPerouseBay]

After that 5th run of 2:30 it felt as if my heart almost came out of my chest after finishing, and I needed to rest for thirteen minutes before resuming the trials (my HR spiked at 213 during that particular rest period). /

Not aimed at you specifically 1P2P, but good reading for us older folks.  Too many stories of MI's on the water.

http://www.surfski.info/getting-started/tips-training/item/1025-atrial-fibrillation-and-the-athlete.html

[PonoBill]

Been there, done that. About ten years ago, on rollerblades, refusing to back off when racing my 6'6" son-in-law. Kicked myself right into atrial fib. Boyum too. Getting old sucks, and sometimes it tries to kill you.