Ventral Fin and fin tests

[PonoBill]

Kind of a crappy day in Hood river--perfect for fin testing. I installed one of Larry Allison's ventral fins yesterday, and I have a bunch of other fins to try on my Blackfish.  I kind of screwed up the finbox install, I was trying to keep it neat and make the patch small, but the 4OZ I laid over the carbon got a bubble in it while I wasn't watching and I had to sand it all off and redo it. Crap. Oh well, the box is in the right place and it's laser straight, so I got that going for me.





Is this the right fin, Larry?

[Larry Allison]

Bill the small fin is the Ventral all the way back. The other fin is a kick fin for crazy conditions between the Twins. Nice work.

[PonoBill]

You're way too serious. Small joke.

[ukgm]

You're way too serious. Small joke.

Let's now hear some feedback on it !

[PonoBill]

Okay, this is complicated. There's a surprise. There's about a ten percent speed difference available by playing with fins on this board, which is more than I expected.

I set a baseline by paddling with the stock three fin setup. I tried three different paddles, expecting the Konihi 84 to be fastest as I've seen before. To my surprise the Mana was faster. Might be because I'm adding more of the Puakea stroke to my stroke.

Konihi 95   5.35 mph max, 4.94 average, 4.30 upwind, 5.35 downwind.
Konihi 84   5.50 max, 5.14 average, 4.40 upwind,  5.50 downwind
Mana 90     5.55 max, 5.20 average, 5.10 upwind, 5.55 downwind
I reversed the order of paddle use and got about the same results, so I used the mana for testing.

Stock plus ventral, Mana 90  Max 5.50  5.20 average, upwind 5.20 Downwind 5.50  The ventral fin took a little getting used to. Tracking is better, stability is better,

I tried taking out the center fin with ventral. the difference was too small to list--about the same as three fins plus ventral.

I took the ventral out and ran stock twins only.
Twins no ventral  Max 5.65 Average 5.20 Upwind 5.05 Downwind 5.65   This was the fastest setup, contrary to my earlier tests of the standard three fin with no ventral. The board was hard to control and it wandered, but it reached higher speed. I tried some larger "nub" style twins I got from Larry with no change.

Single fin and ventral Max 5.10 Average 4.95 Upwind 4.85 Downwind 5.10  This setup accelerated well but it felt like it reached a maximum speed quickly and then stopped accelerating.

Twin nub fins with small "ventral" between them, no ventral  Max 5.50 Average 5.40 Upwind 5.25 Downwind 5.50  This felt really good, accelerated well and held speed easily. Didn't track very well

Twin Nubs with ventral Max 5.50 Average 5.30 Upwind 5.20 Downwind 5.50  Tracked well, didn't feel as easy to hold speed but the results were good.

I was pretty puzzled by all of this, so I tried the Konihi 84 with the twin nub plus ventral
Max 5.65 Average 5.40 Upwind 5.30 Downwind 5.65

By then I was totally confused, and I'm not much less confused now looking at the numbers. From the way things felt, the Konihi paddle works better with the ventral fin than the Mana does. I think I need a teeny ventral fin and a teeny center fin--maybe it just functions as a gate between the two foiled side fins.

The ventral fin adds stability and tracking and doesn't cost any significant amount of performance. I think for anything but a downwind race it would be an obvious choice. There's definitely performance to be gained with multi fin setups, but it's going to take some experimenting to get all of it. I think course conditions and paddle choice may go along with fin choice. Which seems like something surfers have known forever.

[zachhandler]

Okay, this is complicated. There's a surprise. There's about a ten percent speed difference available by playing with fins on this board, which is more than I expected.

I set a baseline by paddling with the stock three fin setup. I tried three different paddles, expecting the Konihi 84 to be fastest as I've seen before. To my surprise the Mana was faster. Might be because I'm adding more of the Puakea stroke to my stroke.

Konihi 95   5.35 mph max, 4.94 average, 4.30 upwind, 5.35 downwind.
Konihi 84   5.50 max, 5.14 average, 4.40 upwind,  5.50 downwind
Mana 90     5.55 max, 5.20 average, 5.10 upwind, 5.55 downwind
I reversed the order of paddle use and got about the same results, so I used the mana for testing.

Stock plus ventral, Mana 90  Max 5.50  5.20 average, upwind 5.20 Downwind 5.50  The ventral fin took a little getting used to. Tracking is better, stability is better,

I tried taking out the center fin with ventral. the difference was too small to list--about the same as three fins plus ventral.

I took the ventral out and ran stock twins only.
Twins no ventral  Max 5.65 Average 5.20 Upwind 5.05 Downwind 5.65   This was the fastest setup, contrary to my earlier tests of the standard three fin with no ventral. The board was hard to control and it wandered, but it reached higher speed. I tried some larger "nub" style twins I got from Larry with no change.

Single fin and ventral Max 5.10 Average 4.95 Upwind 4.85 Downwind 5.10  This setup accelerated well but it felt like it reached a maximum speed quickly and then stopped accelerating.

Twin nub fins with small "ventral" between them, no ventral  Max 5.50 Average 5.40 Upwind 5.25 Downwind 5.50  This felt really good, accelerated well and held speed easily. Didn't track very well

Twin Nubs with ventral Max 5.50 Average 5.30 Upwind 5.20 Downwind 5.50  Tracked well, didn't feel as easy to hold speed but the results were good.

I was pretty puzzled by all of this, so I tried the Konihi 84 with the twin nub plus ventral
Max 5.65 Average 5.40 Upwind 5.30 Downwind 5.65

By then I was totally confused, and I'm not much less confused now looking at the numbers. From the way things felt, the Konihi paddle works better with the ventral fin than the Mana does. I think I need a teeny ventral fin and a teeny center fin--maybe it just functions as a gate between the two foiled side fins.

The ventral fin adds stability and tracking and doesn't cost any significant amount of performance. I think for anything but a downwind race it would be an obvious choice. There's definitely performance to be gained with multi fin setups, but it's going to take some experimenting to get all of it. I think course conditions and paddle choice may go along with fin choice. Which seems like something surfers have known forever.

How many times did you repeat each fin test? Wondering what the signal to noise ratio is with your speeds being spread out like that

[PonoBill]

I didn't repeat them at all. took two hours as it was.

[zachhandler]

Testing is such a damn hard thing to do. Takes forever just to test two different options. And it takes many repeats, and ideally many different testers to be really convincing. Who has time and energy for that?

[Eagle]

It is hard enough doing one max speed sprint test.  Let alone multiple.  I found the more I did -> the slower and more inconsistent the results became.  The fastest time was always the first for me.

What I found more interesting was the GPS track plotted using different fins.  My findings were quite interesting nonetheless and those inclined should do tests of their own.  My tests were done over a number of days with different boards and fins - but same paddle.  It did help provide some clarity though.  But not something I ever want to repeat.

[Larry Allison]

Alright Bill super Stoked to have you taste what I live. Try adding 12 more guys and different global locations and then compare notes from all locations on top of my water time understanding the different mind sets of everyone before making my conclusion. This in some case takes me up to 8 month like the testing we just went thru with same plan shape but different structure, Fiberglass, Honeycomb, Honeycomb/ carbon and foam/carbon. As you see for those who know me I am very anal to not contradict myself like my competitors that change their mind every 6 months. Fun stuff when reality surprises you.

[PonoBill]

The test results above are in the order I did the tests. I got my highest speed on my ninth test (six paddle tests, third fin variation) and then equalled it on the thirteenth test. I'm not a sprinter, I'm a grinder. Diesel motor.

I was pleased to see that the ventral fin doesn't interfere with turning the board. With twin fins the tail is stable enough to easily lift the nose and pivot, but it also footsteers as well or better with the ventral fin than without, and a cross-bow draw works very well and is stable.

I like the feel of the four fin setup. I think I just need to tune it up on size. There needs to be a lending library for fin testing. I'm seeing a ten percent variation in speed just with basic testing with a few fin variations. I'm sure there's a little more. Ten percent is mid-pack to podium for a lot of racers.

[Eagle]

Interestingly over distances like 5 miles or so my GPS times are very repeatable based on around 90% exertion levels.  ie. near max sustainable exertion levels.  The best is that it provides 2 split times over a 3 leg loop to compare as well.  Provides real good data points to confirm perceived vs actual effort over each leg with HRM.  That sort of speed test has worked really consistent for me over the years.  And provides good information as my times have dropped year over year.

[PonoBill]

Yup. It's really only the long tests that matter unless all you do is sprint racing. My performance year to year has steadily improved. Unfortunately, it's improved a lot slower than all those folks around me. Hell, I used to be able to beat Connor Baxter and Travis Baptiste--when they were twelve.

I've seen many times that an apparent improvement in performance doesn't hold true in the long run. The Konihi 84 paddle tests much faster than other paddles. But it only does so for me at a high cadence, which I can't maintain. As soon as my attention wanders it's a very slow paddle.

[ukgm]

Testing is such a damn hard thing to do. Takes forever just to test two different options. And it takes many repeats, and ideally many different testers to be really convincing. Who has time and energy for that?

It can take a while. I can typically test 4 changes in equipment using one kind of test reliably in a 2 hour window. I might repeat that again another day if the data is bunkum or I want re-test stats. I then need 1-2 hours for data analysis. To give you an idea, my current project to test 10 fins is likely to take somewhere in the region of 10-15 hours. It's not just the time, its the weather. You need to pick a venue and know it well enough to wait for optimum or stable weather too.

Yes I do have the inclination to do this kind of thing but then I have a job with a fair degree of flexibility (I'm an academic).

[ukgm]

Interestingly over distances like 5 miles or so my GPS times are very repeatable based on around 90% exertion levels.  ie. near max sustainable exertion levels.  The best is that it provides 2 split times over a 3 leg loop to compare as well.  Provides real good data points to confirm perceived vs actual effort over each leg with HRM.  That sort of speed test has worked really consistent for me over the years.  And provides good information as my times have dropped year over year.

It depends what you're after. Personally, I'd recommend a test speed that is representative of someones typical racing speed but sub max is a good call. Either way, I'm a strong exponent of using 'stroke index' (average speed x stroke length) as my key metric. This is a key and well cited metric in sports journal  literature that require propulsion such as swimming and canoeing. It does mean the results are unique to the user but nobody will get a robust sample volume anyway (and all anyone will care about is will it make them faster anyway). What I've learnt is not to test at sprint speeds. Whilst this can really expose the differences between equipment (due to the exponential increase in drag), it generally leads to a poor stroke quality, be misrepresentative of your racing and this can ruin the results.

For what it's worth, I don't do any less than 6 test runs (typically around eight and will often junk one) and I have been getting a coefficient of variation of my fin test data sets (standard deviation / mean *100) of 1% and that's as good as a science lab. Getting good data is possible with patience.