Boat speed improvements

893 views
Skip to first unread message

Leon Warrington

unread,
Sep 24, 2020, 11:01:38 AM9/24/20
to
So,

I have a 1997 hudson, which in its day was a top boat (it won the world champs). It's an early stern mounted wing rigger design which has been resprayed once, but still original tubular metal wing rigger.

How much faster, if at all, would people anticipate a new, top level boat would be?

I race masters B level (UK), so say over 1000m would you be talking a few seconds, 10 seconds?

I haven't weighed the boat, but it is definitely at least several kg heavier than a few year old Fluiddesign.

Cheers, Leon

carl

unread,
Sep 24, 2020, 1:33:23 PM9/24/20
to
I seriously doubt that you'd detect any difference in the speed at which
that boat can be driven. What deeply intrigues me is the near total
lack of comparative testing of equipment & the implicit assumption that
newer (or what some fancied sculler used) is somehow going to be better.

"A few seconds" would be a huge margin. For a base time of 7:15 for 2K,
a 5 sec speed increase (1.16%) would require a power increase of ~4%.
While feasible, when someone makes that claim one should demand the
evidence or, better still, conduct one's own objective testing.

Since singles are limited to 14kg (a daft rule, beloved only of
blazerati as, if boat weight actually mattered that much, the rule would
deliberately disadvantage the lighter sculler), I'd be astonished if
your boat was "definitely at least several kg heavier than a few year
old" xx. And, if I may introduce just a hint of science into the
discussion, there are compelling theoretical grounds to show that a
lighter version of the same boat may go no faster, or even be a little
slower - there's a minimum weight of boat, for any given sculler &
technique, below which boat speed tends to fall.

Rowing equipment selection seems too often based on the natural desire
to have the latest gizmo rather than on actual evidence. Shiny new toys
aren't necessarily high performers, even if someone won the last WC in
that model. What matters, still, is not the colour, nor the style of
rigger, nor the material of construction, nor the name of the maker, nor
the weight of the boat (within reason), but how well you can work with
that boat, how confident you feel in it, how well you scull & how fit &
strong you are.

Cheers -
Carl

--
Carl Douglas Racing Shells -
Fine Small-Boats/AeRoWing Low-drag Riggers/Advanced Accessories
Write: Harris Boatyard, Laleham Reach, Chertsey KT16 8RP, UK
Find: tinyurl.com/2tqujf
Email: ca...@carldouglasrowing.com Tel: +44(0)1932-570946 Fax: -563682
URLs: carldouglasrowing.com & now on Facebook @ CarlDouglasRacingShells

---
This email has been checked for viruses by AVG.
https://www.avg.com

carl

unread,
Sep 25, 2020, 7:38:13 AM9/25/20
to
I should have added: a significant part of the resistance the boat
encounters is from the other fluid that you pass through, namely the
air. If, into a strong headwind, your speed falls by 20%, then wind
resistance is now causing roughly half of the total drag. In still air
the wind drag is significant, & reducible.

The main source of that wind resistance is the sculler & the oars.

If the wind is coming from the side, then you incur additional
hydrodynamic drag due to the boat crabbing along the course to overcome
the leeway induced by the aerodynamic side-force - hulls are not
optimised for this, but again this is reducible.

carl

unread,
Sep 25, 2020, 9:10:48 AM9/25/20
to
Funnily - if anyone finds ignorance funny among those paid well to
extract the last 0.1% of performance from their crews - I've just seen a
2016 publication on wind drag reduction on oar shafts:
https://www.sciencedirect.com/science/article/pii/S1877705816306452

The wheel re-invented - yet again, but only 16 years late. We applied
boundary layer trip strips to the oars of the GBR eight for Sydney 2000.
Since when they've been quietly forgotten by our sport.

Such a shame, since science really can help in our sport. It is easier
to gain performance by applying known science than to buy it in the form
of new kit.

RunDMC

unread,
Sep 25, 2020, 11:22:46 PM9/25/20
to
Hi Carl, I've used speed strips on my oars off and on (they fall off after awhile, and it takes me awhile more to get new ones re-applied), and I only take it on faith that they help because it would take a tremendous amount of trialing back and forth between strips and no strips to detect a difference. I've read the referenced paper before and it is rather difficult for me to figure out whether the authors take feathering into account when they do their modeling. If not, they have somewhat overestimated the air drag reduction as one can only have the strips oriented properly during the feather or when square, not both (and for fairly obvious reasons we want them working when feathered). The paper is also a little sloppy with a missing equation reference on page 113. And disappointing that they didn't get experimental data with speed strips of different thicknesses. Is 1mm sufficient? Is the zig zag important? Other publications from the glider world suggest not: a thin wire may do the trick.
But an interesting and probably overlooked opportunity for free speed. Maybe we should be applying them to our riggers as well, although the gains may be smaller than for oar shafts.

regards,

Dan

Richard

unread,
Sep 27, 2020, 7:41:49 AM9/27/20
to
On Friday, 25 September 2020 at 14:10:48 UTC+1, carl wrote:

> Such a shame, since science really can help in our sport. It is easier
> to gain performance by applying known science than to buy it in the form
> of new kit.

Amen to that! Although Christmas is coming and who doesn't like shiny new kit or geeky new gadgets...? ;-)

The IET's (Institution of Engineering and Technology) Surrey local network recently hosted a webinar given by Prof Steve Haake from Sheffield Hallam University. It was great to see how he and his team had approached a range of different sporting challenges from a scientific and engineering perspective. Although his work includes aerodynamic and hydrodynamic analyses applied to a number of sports, sadly rowing didn't feature - I wish it did. Improving our understanding of the physics and engineering behind our sport and burying the myths and folklore surely can only benefit us all.

The webinar was recorded but I've not managed to work out where it's posted online. In the meantime, Steve's website is stevehaake.com with links to various other talks he's given (and a plug for his book "Advantage Play").

carl

unread,
Sep 27, 2020, 10:36:29 AM9/27/20
to
Thank you for that feedback, Richard. And I particularly note this
comment in the blurb for his book:
"... in a world where top performances are static, where we've found all
the best athletes and deployed the best coaching methods, the one thing
that might distinguish the winners from the losers is new technology."

Amen to that! But it doesn't have even to be new technology as there's
plenty of established technology that could help but never gets applied.

There is understandable difficulty, for athletes & coaches, in accepting
or comprehending the mechanical sciences which can be legally and
honestly applied to enhance performance. It can be hard to quantify, &
harder still to remember, the worth of those changes in equipment which
do bring small but significant benefits. Great coaches & athletes often
have no (or minimal, or flawed) understanding of fluid dynamics &
inertial mechanics, being driven first by the need to maximise fitness &
replicate what worked well last year.

This often does not leave them very receptive to what may be great
ideas, well accepted in other fields, with directly applicable benefits
to rowing, which are predictable but only by means of the relevant
sciences. The difficulty of wrestling with hitherto alien concepts
(which may also challenge hitherto-held beliefs), & a rational fear of
getting it wrong or being bamboozled into using useless gizmos, can
further inhibit the acceptance of good ideas. If you break with current
orthodoxy & your crew doesn't perform well, that'll be you fault,
whereas if you play safe but your crew still fails that will be forgiven
as you "did everything right but the athletes weren't quite up to it".

Jake Frith

unread,
Jan 8, 2021, 12:09:55 PMJan 8
to
I'm as frustrated as Carl is about this state of affairs in rowing.
Talking from a UK perspective, and as someone who has worked some years for a sports governing body. Ever since British Cyling started its 'marginal gains'/ AIS inspired stuff (A process that started WAY before London 2012 oly cycle, back last century with Chris Boardman and his Lotus bike and Graham Obree battling each other for the hour record etc., most UK sports NGBs have been falling over each other to 'get science'.
But rowing, which (in the UK) has the budget, and is a similarly equipment heavy sport... Nope. Not interested. Just do more 300kg deadlifts.
If British Rowing would just put its hand in its pocket and spend a few quid with a company like BAR (an America's Cup campaign spinoff now doing all sorts of interesting stuff), it would (certainly initially) pay for itself many times over.

Re: the comments on aero and oar shafts. Clearly, obviously a low hanging fruit to be had there. That is one of the obvious places I'd be looking first. No need to pick whether optimising for squared or feathered. Dead easy to sleeve the oar shafts in a carbon aero sleeve, connected to the oarlock so it stays horizontal all the time and the oar shaft just rotates inside it when squaring and feathering. Get it all nicely PTFEd up so the shaft can turn easily in it. I reckon max cost would be 0.5kgs per oar. Leave it open at the blade end so the water drains out. Maybe have a connected short section inboard too for total aero! It would work perfectly with existing oars so a low cost addition, not expensive new kit. An 8+ has got 8 of these horrendous sticks prodding out of the sides, and up by the oarlocks they are THICK! Into a headwind the improvement would surely be measureable? Even in no wind it'd be worth having.

Pat

unread,
Jan 9, 2021, 10:39:26 AMJan 9
to
Where can I buy the Aero-strips for use on my sculling blades?
I've read about the aero-strips and would like to try them out in my scull.

I also have a questions around how to test technology to statistically show a difference. I would like to know what science based protocols exist and what is currently used.

Will the technology makes the boat go faster? how to measure?

Technologies that I know of ....

1)
Oars - https://www.concept2.com/oars/how-made-and-tested/speed-testing

2)
Randallfoils - http://randallfoils.com/

3)
Shoe base plates - https://batlogic.net/why-bat/

4)
Carbon riggers


My plan
Timetrial with other scullers on a Thursday evening at 6:30pm
Pick one or two other scullers as my references
Complete 1.1K 3 times each Thursday for 3 - 5 Thursday's
Record times, and difference in my time to my two reference scullers
Make the Change - technology and adapt to technology change
Repeat
Box-plot, T test and see if my time delta has changed

From my view, I have reduced the variables
Disadvantages - the time period is extended so some people's fitness may improve more than others vs. time to accommodate the adaptation to the new tech.

Henry Law

unread,
Jan 10, 2021, 2:03:43 PMJan 10
to
On Sat, 09 Jan 2021 07:39:24 -0800, Pat wrote:

> I also have a questions around how to test technology to statistically
> show a difference.

I'm afraid we must contact your national governing body and have you
thrown out of the sport for suggesting such a thing. You should know
that rowing is all about hunch and opinion: never about data.

--
Henry Law n e w s @ l a w s h o u s e . o r g
Manchester, England

carl

unread,
Jan 10, 2021, 6:26:54 PMJan 10
to
That is exactly the question to be asking, since on-water performance
tests, while sometimes conclusive & sometimes the only way to test, can
(like seat racing for crew selection) be highly subjective.

Some things we can test, but some we can say without question will bring
meaningful benefits. Of the list above the only one we can say, without
doubt, will bring real benefit is in applying boundary layer trip-strips
to oar shafts (except in a strong tailwind). These strips - raised
ridges about 1mm deep, as narrow as you like, & running the length of
the shaft at about + & - 70 degrees from what would be the horizontal
plane for the shaft during the recovery - do assist boundary layer
re-attachment (funny old science, this fluid dynamics game) which
assists pressure recovery in the departing air flow, & reduces the width
& turbulence of the airflow wake - all of which reduce net drag - as the
shaft moves through the air, especially when the shaft is moving forward
and the more so into a strong headwind.

Narrower shafts also help.

As for carbon as a means of going faster? I'd welcome an explanation of
why this might change the mechanical & aero/hydro-dynamic performance of
equipment unless it permits the creation of more "slippery" designs.
That said, riggers can be evaluated by wind-tunnel testing - provided
you properly simulate the interaction of boat, crew & water surface with
the flow past the rigger. This can be done, & we have done it - to a
very reasonable degree. And, if you have a lot of money, it can also be
done to a reasonable standard by CFD (computational fluid dynamics)
modelling and analysis.

Otherwise tests have to be water-based. And that will require either
blindfolds on the rowers or some other means of removing visual causes
for prejudice.

Even then, anything which makes the rowing "feel" different will require
a period of adjustment, which must remove or reduce the scope for clean
objectivity.

However, if we are ever to go any faster with the same old bodies, then
it seems we should first conduct proper theoretical analysis of the
chosen options (& the means increasingly exist for so doing) & then
conduct on-water tests with as much rigour & objectivity as possible.

And we could apply similar approaches to matters of technique.

In short, rowing is a field sorely lacking in research, with its
development impeded by a surfeit of opinion, powerful commercial
pressures & a lack of technical nous. But that's only my view, & I
could be wrong ;)

Kit Davies

unread,
Jan 11, 2021, 6:02:56 AMJan 11
to
On 08/01/2021 17:09, Jake Frith wrote:
> Re: the comments on aero and oar shafts. Clearly, obviously a low hanging fruit to be had there. That is one of the obvious places I'd be looking first. No need to pick whether optimising for squared or feathered. Dead easy to sleeve the oar shafts in a carbon aero sleeve, connected to the oarlock so it stays horizontal all the time and the oar shaft just rotates inside it when squaring and feathering. Get it all nicely PTFEd up so the shaft can turn easily in it. I reckon max cost would be 0.5kgs per oar. Leave it open at the blade end so the water drains out. Maybe have a connected short section inboard too for total aero! It would work perfectly with existing oars so a low cost addition, not expensive new kit. An 8+ has got 8 of these horrendous sticks prodding out of the sides, and up by the oarlocks they are THICK! Into a headwind the improvement would surely be measureable? Even in no wind it'd be worth having.

My observation is that, when the blade is in the water, the shaft moves
forward through the air very slowly (in comparison to the recovery at
any rate), so the need for extra aerodynamic improvements in the squared
state is far less, and maybe not worth the extra complexity.

Kit

carl

unread,
Jan 11, 2021, 3:52:40 PMJan 11
to
On 08/01/2021 17:09, Jake Frith wrote:
I don't disagree, Jake, but it might be a bit of a contraption, &
possibly subject to buffeting from turbulent air currents?

I understand that at the Atlanta Olys the Japanese turned up with very
much what you describe. It might have worked well for a stronger crew
but, like so many other potentially useful devices, it seems not to have
been tried since then.

There is this problem of clutter & surface area to consider, & when
boundary layer trip strips are so simple & effective it seems anything
that is not heavily promoted is doomed to oblivion. They were first
used in rowing by the GBR M8+ in Sydney2000. No one remarked on them,
even though that crew won. Someone leapt on the bandwagon a year or so
later (I forget the name they applied to it), but then it was heard of
no more.

Such devices were used by skaters (on their legs) but I don't know if
they still are, & (I believe) in cycling?

As ever, the problem is that our sports - always looking for spectacular
gains - seem not to care about the marginal gains available from the
appliance of science, even though races can be won & lost by centimetres
(1 cm = 0.0005% of 2k, or an effective difference in useful power output
of just 0.015%).

It's not smart to ignore even small contributions to performance, &
rather more is available from a range of different tweaks to equipment.
You should place more trust in well-established science & less in
subjective assessments. Folk are sometimes surprised to learn that
aircraft are designed & their performances finely calculated long before
the first metal is cut.

And then there's technique - which is not a matter of how it looks but
how well it actually works...

Jake Frith

unread,
Jan 12, 2021, 12:02:27 PMJan 12
to
Yes, it would indeed be a contraption, which is why I never did it, and I'm someone with a loft full of carbon cloth, an airing cupboard full of epoxy, and an unhealthy interest in doing weird, performance inspired boat modifications. That and because as also covered above I would have no way of ever finding out if it had produced any benefit. It would also (if built to the lightness you would want it) shatter in a cloud of carbon shards on even fairly light contact with a dock/ another crew's oars, so it would be utterly impractical in club rowing (although that would not stop them wanting the technology if it had won at the Olympics). But I've never been in the business of marginal gains.

I was not aware that the Japanese had used them. As you say, it might have made the difference to them between 4th and 6th in a heat, but nobody would have noticed.
I'd imagine it would be the sort of 'secret weapon' a coach would wheel out in the case of a headwind. It would mess with the opposition's heads apart from anything else.
The boundary layer trip strips alternative would be much more robust for club environments, if not quite as aerodynamically efficient, but still a significant improvement over a plain tapered cylinder (standard oar).

Carl makes an interesting point on oar shaft thickness. The fear of one snapping (it does occasionally happen) must make the makers over engineer them a fair bit. But I think they are the same wall thickness all the way up, just mandrel wound I'd imagine so there's not much science gone into them. They appear to be wound in a pretty similar way to a windsurfer mast (also came in in the late 70s/ early 80s and produced in much bigger volumes, so perhaps no surprise about the similarity), but a windsurfer mast is designed to bend considerably and doesn't mind being round as it sits insife a lufftube that is teardrop shaped. Nowadays composite tubes that are loaded in one direction only in use are ovalised. Although there is the annoyance that the ovalising you need for strength is the opposite way from that you would want for aero on the feather, I'd have thought there would have been efforts made to internally ovalise them. Or even just put a strip of unidirectional carbon down the aft edge of the shaft (to provide more strength in tension) and then you can make them from thinner tube reducing both weight and air drag.

The carbon SUP racing paddle I use weighs 440g and is about 7ft long. It is slightly ovalised (28mm on thin measurement 30mm on thick). OK, it's not as long and it doesn't get bent round a fulcrum which creates a big folding force, so you'd never get very close to that weight, but I'd imagine sculling and rowing oar shafts could be made a fair bit thinner couldn't they?
Jake

Matthew Farrow

unread,
Mar 26, 2021, 6:46:19 AMMar 26
to
Hi,

How would you suggest ensuring the shroud device does not become load bearing during the drive? Obviously a shroud is simple, but deflection can be up to 300mm in a soft-shaft rowing oar near the spoon, so I do not believe the problem is quite as simple as you envisage.

If the shroud stops the oarfrom bending, it will support the load the rower is applying instead. In this instance, a 0.5kg shroud would undoubtedly break.

Matt

Jake Frith

unread,
Mar 26, 2021, 9:04:40 AMMar 26
to
Interesting thought. I would imagine it would depend on how the fibres in it were wound /orientated and the epoxy used. Plenty of ways to make very light , bendy shafts. Think fishing rods etc. The difficult/ expensive thing is making long, lightweight shafts that you DON'T want to bend.

If getting the right flex really was a problem, you could intentionally segment/ split the trailing edge in a few places so each join would intentionally bend apart a few mm each stroke, but if it was made from a material with the correct flex characteristics in the first place, this would not be necessary. If you split the trailing edges they would also probably play distracting tunes into a strong headwind!

These simple tapered sleeves would not even need to be strong enough to support their own weight when not on the oar. Indeed if they could, you'd know you'd made them too heavy.
The problem would be any kind of dock, flotsam, onshore handling interactions with other solid objects. In a rowing club with kids chucking them in a trailer type scenario they would not last 5 minutes. In a varsity boat race type oar clash scenario the affected shrouds would be smashed to dust (although you'd still have your ordinary oars underneath to continue racing with).

That's why if I was responsible for sports performance at elite level, especially in VIIIs (more oars- more gain), I'd give it a go. They would be so light and fragile, if anything went wrong on the day with the technology (such as an impact stopping them staying horizontal to the water/ oarlock) any remnants getting in the way would most likely snap and fall off and not create problems. So something slight to gain, with very little potentially to lose. I'd imagine the gain into a strong headwind would be noticeable.

You could make teflon lined, split (2 part) rotating stations of the appropriate sizes for a few points down the oarshaft and just glue on appropriate thin sheet material, wrapped round the leading edge and taped together at the trailing edge, just to see if it works. The far end rotating stations would need open architecture to let any water straight out.
I used to have some sheet material that would do, pulled it out of the inside of the leading edge of a scrap hanglider, it was there to make it a better aerodynamic shape. Not sure what it was though. Some sort of thick acetate I'd guess. It was effectively doing a similar job.

lladn...@gmail.com

unread,
Mar 27, 2021, 5:38:29 PMMar 27
to
I have had some trials adding an areo foil to an oar. This is only possible if you restrict the depth an oar is buried.

Fast ... what do you think!

https://drive.google.com/drive/folders/1WIoVQkN4-5BA_QeSCtU1P7jL7yn1NDyN?usp=sharing

John Greenly

unread,
Mar 28, 2021, 12:51:52 PMMar 28
to
It appears that folks here are not familiar with Dreher oars and sculls. I use their standard adjustable sculls. The shafts are circular internally to accept the sliding adjustable handle section, but externally they are slightly elliptical. They are more than twice as thick on the sides aligned with the blade faces as they are the other way and use unidirectional fiber in the layup as well, to take the bending load during the drive. This construction results in a lighter and very stiff shaft.

Dreher also makes more aero-efficient elliptical-shaft oars with the wide axis of the ellipse parallel to the blade faces, so it's aligned into the wind during the recovery to reduce drag. Those shafts are also laid up much thicker on the correct sides (the narrow direction) to take the bending load during the drive.

I have no connection with the company, but I think it's unfortunate that they don't seem to be very widely recognized. As well as their innovative engineering, their manufacturing techniques and quality are absolutely top-notch, with flawless surfaces and smooth joints. The shafts are not mandrel-wound, but use various prepreg cloth and fiber layups that are autoclaved under pressure (according to their website). I can attest to the toughness and longevity of their products; I bought my sculls already well-used, and then over the last 10 years I've bashed the shafts on all sorts of solid objects including (shamefully) a steel navigation buoy on the ocean in the Blackburn Challenge, and they're still good as new.

Cheers,
John

Ian Randall

unread,
Mar 28, 2021, 2:24:02 PMMar 28
to
On Monday, March 29, 2021 at 3:51:52 AM UTC+11, John Greenly wrote:
> It appears that folks here are not familiar with Dreher oars and sculls. I use their standard adjustable sculls. The shafts are circular internally to accept the sliding adjustable handle section, but externally they are slightly elliptical. They are more than twice as thick on the sides aligned with the blade faces as they are the other way and use unidirectional fiber in the layup as well, to take the bending load during the drive. This construction results in a lighter and very stiff shaft.

Of course!
Just realising the potential of the design concept.
Full aero shafts are simply not possible until shafts are out of the water during the drive.
This is only possible with a hydrofoil applied to the top edge of the blade, or some other blade depth limiting device.
I see this as the next stepping stone in blade/shaft design.
The full aero shafts trialled were to mirror the aero profiles achieved by modern bicycles.... anything less is a compromise.

Message has been deleted

Matthew Farrow

unread,
Mar 29, 2021, 6:04:50 AMMar 29
to
Hi John,

I am familiar with Dreher blades, but from measuring the chord to thickness ratio I do not believe the reduction in aerodynamic drag is the same as they may lead you to believe on their website. From scaling up the image and measuring as accurately as possible (unfortunately this is as good as I could get), the chord to thickness ratio seems to be about 3:2. You don't observe the same reduction in Cd as you would with a much longer chord length, such as the example posted. Obviously they are an interesting product, but they aren't doing enough for me to fully capture my interest and build a brand as being the aerodynamicists of rowing.

As for the makeshift shroud posted, next step is to get some data!

Matt

Jake Frith

unread,
Mar 29, 2021, 11:02:36 AMMar 29
to
Part of the challenge with ovalising the shafts to make them lighter/ stronger AND more aero at one fell swoop is the obvious one that the way you would want them ovalised for aero is 90 degrees different from the way you would ovalise them for lightness/strength. That's probably part of the reason Dreher is not going for longer chord, more aerodynamically efficient ratios. If you are obtaining your aero shape from the layup of the shaft as a whole, rather than optimising the shaft structurally then adding additional lightweight shrouding parts for the aero as I have suggested, it's likely to start to get pretty heavy if you need to combine enough strength in the drive with aero on the recovery and wind the looms using similar systems and technologies to your other products. There's also the fact that they probably don't want to frighten their customers with anything that looks too radical.
But Dreher should be praised for moving things in the right direction.

Of course the ultimate good design solution would integrate structural and areo elements, such as high end bicycles do, but my view is that there are already a lot of round looms out there in use, and they don't often snap in normal use so some sort of additive aero could work quite well. Additional weight would be equivalent to a pair of wet rowing shoes versus dry.
While the benefit of shrouding the lower looms with fixed aero covers as shown above is a great way to start. The weight / complexity challenge of additionally allowing the shrouds to rotate on the looms would be minimal. Basically, you don't glue them to the shaft, and you have a simple fork and pin system that fixes them to the front of the oarlock.
The benefits of allowing the aero shrouds to keep their horizontal orientation throughout the stroke cycle are:
- They can cover the whole loom. You will not need to end them some distance before the blade, (as shown above) as during the drive they won't be trying to work as an aditional section of blade moving against the flow- as is the case if you fix them on in the feathered orientation. Having to end them before the blade, leaving unshrouded cylindrical loom exposed is penalised quite heavily- as even though it is the narrowest part, this is also the fastest moving part of the loom.
-You will see (slighter) benefits, more so in strong headwinds, on the drive phase as well as the recovery, as the parts of the blade further inboard are still moving through air at getting on for the speed of the boat, even during the drive phase.
-If you square early, it won't penalise you for it any worse than you would be with standard circular shafts. If you square early with deep aero shrouds fixed to your shafts, you'll have the significant additional drag of squaring those shrouds early too. (If the highly unlikely event you're in a very strong tailwind this advantage gets reversed)
-You might as well. It's not technologically demanding and is likely to be slightly advantageous.

The surmountable disadvantages are that they would need to be made slightly stronger (and therefore heavier) than fixed shrouds so they don't twist down the length of the loom, and the fact that you are introducing a simple additional mechanism, and even simple mechanisms can, very occasionally go wrong.

My guess is the advantages of rotating shrouds would outweigh their disadvantages over fixed, but it's only a guess, that only a scientific testing procedure would shed light on. Either option would likely be better than round looms or Dreher's slightly eliptical ones.

Matthew Farrow

unread,
Mar 29, 2021, 12:29:28 PMMar 29
to
Completely agree. Stationary shroud with blade rotating inside is almost certainly the way to go. However, after giving it some actual thought it is difficult to think of a good solution which allows for bending of the oar, whilst forcing each shroud segment to remain horizontal. The challenge is to connect the segments via a method resistent to torsion, but free to bend with the oar. I can see plenty of solutions leading to some sort of twisting, whereby the further outboard segments would rotate and not maintain the desired position, as the ~2m cantilever really is a challenge to overcome. At the very least, the outboard segments would freely rotate to some degree and some "play" would be present in the system. Remember the segments must remain horizontal under forces from the wind, the constant acceleration of the oar+boat. A rower would simply get annoyed with a wobbly shroud attached to their blade which catches the wind and makes a racket. Some more detail on the fork and pin system you envisage would be great! I see how it would work for the first segment nearest the oarlock etc, but 2m outboard with the requirement to allow for 300mm of bending is a challenge!

It is also very important to not enfringe too much on the bending characteristics of the oar since they are crucial for power delivery and mechanical efficiency. Big changes in the "feeling" of rowing would probably also lead to rowers choosing not to adopt the development, and stick with their trusty shroudless Concept2. I'd say a good design criteria would be to not alter stiffness properties whatsoever, then a genuine solution might fall somewhere in the "negligible" ballpark.

I also wouldn't say the shroud could go the entirely up to the spoon, and some space must be left instead. At least 200mm of the loom is covered during the drive, and I can see a large aerofoil segment being rather annoying in this regard getting stuck in the water etc. This is especially in headwinds and wavy conditions, which is the exact time you'd want your aero devices fitted!

I am currently doing a university project on the topic, hoping the quantify the advantage of a fixed horizontal shroud vs non-shroud. I am not focusing too much on the actual design of the product, more the aerodynamics and rowing modelling side. However, when I have sat down and tried to get a general concept hitting the criteria I mentioned above then it becomes a bit tricky.

Not shooting any idea down, just developing the discussion which is v enjoyable! And if your idea is valid and would hit the concerns I've mentioned above then that's awesome, and I'll happily hold my hands up haha.

Jake Frith

unread,
Mar 29, 2021, 4:01:41 PMMar 29
to

> Completely agree. Stationary shroud with blade rotating inside is almost certainly the way to go. However, after giving it some actual thought it is difficult to think of a good solution which allows for bending of the oar, whilst forcing each shroud segment to remain horizontal. The challenge is to connect the segments via a method resistent to torsion, but free to bend with the oar. I can see plenty of solutions leading to some sort of twisting, whereby the further outboard segments would rotate and not maintain the desired position, as the ~2m cantilever really is a challenge to overcome. At the very least, the outboard segments would freely rotate to some degree and some "play" would be present in the system. Remember the segments must remain horizontal under forces from the wind, the constant acceleration of the oar+boat. A rower would simply get annoyed with a wobbly shroud attached to their blade which catches the wind and makes a racket. Some more detail on the fork and pin system you envisage would be great! I see how it would work for the first segment nearest the oarlock etc, but 2m outboard with the requirement to allow for 300mm of bending is a challenge!
>
> It is also very important to not enfringe too much on the bending characteristics of the oar since they are crucial for power delivery and mechanical efficiency. Big changes in the "feeling" of rowing would probably also lead to rowers choosing not to adopt the development, and stick with their trusty shroudless Concept2. I'd say a good design criteria would be to not alter stiffness properties whatsoever, then a genuine solution might fall somewhere in the "negligible" ballpark.
>
> I also wouldn't say the shroud could go the entirely up to the spoon, and some space must be left instead. At least 200mm of the loom is covered during the drive, and I can see a large aerofoil segment being rather annoying in this regard getting stuck in the water etc. This is especially in headwinds and wavy conditions, which is the exact time you'd want your aero devices fitted!
>
> I am currently doing a university project on the topic, hoping the quantify the advantage of a fixed horizontal shroud vs non-shroud. I am not focusing too much on the actual design of the product, more the aerodynamics and rowing modelling side. However, when I have sat down and tried to get a general concept hitting the criteria I mentioned above then it becomes a bit tricky.
>
> Not shooting any idea down, just developing the discussion which is v enjoyable! And if your idea is valid and would hit the concerns I've mentioned above then that's awesome, and I'll happily hold my hands up haha.



Great to see that someone's taking these things further. Yes, the 300mm of bending is indeed a challenge. Segmenting the shroud in some way has to be one way ahead. Segments could have overlapping/ underlapping skins, or if separated by a few inches, just lycra or similar covers to keep aerodynamic integrity over the gaps. The leading edge (trying to compress/ shorten during the drive) will be quite close to the oar, so its change during the stroke versus the oar will be quite small. The trailing edge extension and compression, assuming an efficient (deep) foil chord, will be pretty big.

It's the sort of thing that would take a lot of trial and error.


Segmenting the shrouds, each one would need an airfoil station, or rib (which would be the part, probably in two pieces, (so you can get it on the oar), containing the low friction bearing surface rotating on the loom), at or near each end. If you split the shroud into 3 or 4 segments each one will not have to deal with much bend of the loom within its length. It can even be designed with enough fore and aft space inside it to allow the loom to do its bending completely unmolested. Minimising the twist in each individual segment should not be too much of a worry. If I was doing it, I'd include in the laminate unidirectional filaments of aramid diagonally in both twist directions. Aramid's excellent resistance to tension loads, mean that if laid up right the twist will not be noticeable within each module.
The problem, as you have identified, would be taking out the play from segment to segment. One could imagine a staircase of slightly further out of line segments progressing towards the blade, regardless of whether the segments were coonnected by telescoping or scissoring linkages. If you think about it, the segments stiction to the loom at each station will mean the sections up by the blade will have a tendency to over rotate when you feather (nose down airfoil attitude), so on the recovery they could be giving highly undesireable downforce up by the blade!

As the segments are getting smaller as you progress down towards the blade, so smaller chord and thickness, might they conveniently telescope inside each other a few inches at each joint?

Any segmented sections further inboard can be quite long, as most of the oar's bending is down at the far end. Inboard of the oarlock (I'd shroud that bit too- might as well!) has so little flex and is so relatively short, I'd say flex compatibility of shroud and oar won't be an issue there.

Whatever 'staircase effect' you get may well be reset each drive phase, so in that respect it could be a blessing as well as a curse. The bending of the oar along its length happens in the plane you need it to to bring all the segments back in line with each other. In that respect you don't want too many segments as allowing the curved loom to push back into each segment, straightening it up on the drive might be a benefit. If segments are too short, you won't get this effect.

If you're going to have to use segments, don't worry about pins and forks etc. at the oarlock at this stage. That's only going to look after your first segment. Theres plenty of ways you can attach it to the oarlock that will have low enough play to keep the first segment in the right place and still allow the blade upward and downward angle movement WRT the water that it needs. If you had to you could put the whole oarlock in a hinge with it's pin horizontal and arranged on fore and aft plane (like some old riggers used to have as outboard pitch adjustment, but able to move). Then you could clip/ bolt the first shroud directly to the oarlock, but that's a fair bit of engineering. It's the subsequent lower segments and how you link them that you need to worry about. The more important aspect of attachment at the oarlock will be doing it in a way that will make getting the oars in and out of the boat no more difficult or time consuming than it currently is- but that is a simple/ surmountable challenge.

Anyway, you could slightly, intentionally underrotate the top segment by the oarlock, (with a simple string or similar mechanism) knowing that this one is aerodynamically less important, but do it in such a way that it helps reign in the lower segments' tendency to overrotate when you extract and feather.

Re: burying a horizontal aero loom by the blade during the drive phase and extraction. That's an area no CFD program or anyone can tell you as I don't think anybody's done it. Trial and error is the answer there. It might not be as bad as you think! Who knows?

So the big problem that needs adressing is the 'staircase effect' of all the play between the segments giving you a twisted foil shape.

Let's assume for a moment we could exactly control the position of the bottom segment. The one furthest away down by the blade. The one which is probably most important as it's the fastest moving, yet is the one likely to be most affected by our staircase effect as its furthest from the oarlock.
We've got control of our top segment because it's connected to our oarlock , and we've got control of our bottom segment because were saying we have. There's only going to be one or two segments between these, and we'll still have to minimise play between them, but I think it's then going to be a manageable system.

So, lets think about whether we can more actively control the position of our bottom segment. And maybe, depending on how we do it, we can constrain some or all of the other segments on the way down there too. 2mm dyneema line has 120kg break load and minimal stretch and enough of it to stretch the length of both oars weighs less than the shoelaces in our rowing shoes. Each of our stations needs to rotate no more than 90 degrees on the loom. Underrotating can be easily constrained by a ring with a pin in it next to any one of our rotating stations. But, going from squared to feathered our shroud will try to rotate with the shaft, ie., over rotate
We know our lowersections need to snap to thier (foil horizontal) stops as we feather so that they do not overrotate (following their tendency to rotate with the shaft).

Could there be a system where we run our dyneema line from a side of our oar collar, or a short lever fitted to the collar run down holes in our stations to a bottom section squaring/ feathering mechanism. (This is making my head hurt without a cardboard model!). But I would think it's possible. The return of the bottom shroud to its 90 degree stop when squaring could be met by a light (3mm) bungee inside the module if neccessary.

Working out which way each component is trying to go and how to constrain it is very hard to visualise without models, so I might have made mistakes above concerning whether you need pins in tracks, bungees, cords etc. But you are moving a mechanism in a pretty fixed set of directions, and the point that the rotation of the bottom section needs constraining/ needs a helping hand will be at the same point each stroke cycle.

The energy required, even with some system loss to flip a lightweight carbon shroud into position during drive or recovery will be unlikely to be of measureable level, as we're moving about 60% worth of our body weight up and down the slide, and it's a tiny proportion of this we'll be using to pull a light string at the opportune moments.

Aware that every sentence above adds extra weight and complexity though.

If I was going to try to work this mechanism out and demonstrate it, I'd put a boat in trestles with an oar in and make a little hollow 6 inch long foil section down by the blade and a station each end of it. I'd then look at it in the flesh and contemplate/ test where and when you need to pull a line to flick the section to each of its end stops. I reckon I'd get a working system together in an afternoon, or at least prove its unfeasability. Once you can keep a little mock up section at the far end of the blade where you want it at all times you've largely cracked it. You can repeat the mechanism in as many sections as you want- it's going to have negligible weight and all internal so no air drag. But as I say if you can can control the top and bottom ones, you're probably sorted.

As a final point, in view of all the above complexity. The drive phase is the point where you are least worried about the Cd of the looms. So a system where you let the trailing edge of a single part foil intentionally split apart slightly every drive phase as the curve of the loom starts peeling its way partially out of the middle of the foil trailing edge might not be the end of the world. It could come as an extrusion that's already foil shaped, tapered and naturally snaps together at the trailing edge. Because it can be flexed over the blade you just fit rotating stations to your loom and snap the extrusion over the blade. Only connect the semi rigid skin to the bottom, top stations and oarlock end. Might work. Might not. Would depend on very low friction at the bottom rotating station- suggest millionaires tape round loom and ptfe inner bearing surfaces on the station parts! If it worked it would be cheaper, removeable etc. though it might hold a bit of water in it at times.


Jake Frith

unread,
Mar 29, 2021, 4:19:25 PMMar 29
to
Ps.
Before anyone picks up on it, the phrase:
" as we're moving about 60% worth of our body weight up and down the slide, " is nonsense as we're in a moving boat, not on an ergometer, but the sentiment is correct - that we have relatively a lot of energy flying about the system and in comparison the amount to tweak a light carbon shroud into a better position on a low friction shaft is negligible.

Ian Randall

unread,
Mar 30, 2021, 2:03:53 AMMar 30
to
Current FISA rules will not allow any moving parts on a rowing oar, thus an aerodynamic shroud as described above would be permitted on the shaft of an oar.

However, this design concept has been solved ... see images and plans here.... https://drive.google.com/drive/folders/1WIoVQkN4-5BA_QeSCtU1P7jL7yn1NDyN?usp=sharing

Test it yourself ... it noticeably reduces the wind resistance on an existing oar. Free speed.

The problem is not the effectiveness or legality of this design or a host of other speed advancements in our sport but the very nature of innovation and adoption of novel designs.

I made this presentation at an Australian STEM conference earlier in the year on the development of the RANDALLfoil ... you may find it interesting.

https://drive.google.com/file/d/1zmtLH02oDHZjOTd3vquX6AeknL1PiktV/view?usp=sharing





Andy McKenzie

unread,
Mar 30, 2021, 4:32:57 AMMar 30
to
Surely (legalities/cost/practicality aside) the way to go would be a aerodynamic and structural oar shaft, with fixed orientation, which could be given the appropriate structural strength and bendiness? Have a spoon on the end that rotates, and carry the rotation from the handle through the fixed shroud by a flexible drive shaft. That way you maintain your aerodynamics and avoid all the fussing with flexible shroud elements.

Jake Frith

unread,
Mar 31, 2021, 6:18:43 PMMar 31
to
If FISA doesnt allow any moving parts on a rowing oar, that's that, and does rather limit options aerodynamically. But the fixed shrouds linked above should see benefits that are worth crews looking at. If I was a serious rowing crew, I'd at least have a second set of oars fitted with shrouds like this that I'd pull out of the bag if there was a headwind.

Thanks for the Randall foil presentation. I found it quite interesting and inspiring especially the stuff about innovators having almost a duty to progress thier ideas. I've been sitting on and (far too) slowly developing a few novel, patentable ideas (outside of rowing) and feel the same sort of responsibility/ guilt. Indeed to the point that the frustration of not having developed any of them has made me sometimes regret conceiving them in the first place, especially every time inferior products enter the market.

My thoughts about the Randall Foil itself (which I accept nobody has asked for) are:
Has diving of oars been proven to be a problem? I can see how a novice rower diving a blade 18 inches underwater is clearly an issue (alters the gearing, oar has to be dragged back out again at the end of the stroke, upper part of the loom is a brake against the oncoming flow). But the difference at elite level between an athlete burying an oar 3 inches under versus another burying 1 inch under. Is that even measureable? The oar and that first foot of loom isn't moving very far through the water, and it's mainly twisting in its own stationary puddle isn't it?

Also the leading edge of the oar (if we are considering it a foil) at the catch, which from experience is where the diving tends to be an issue, is its far end, surely. Does this mean that the behavior of a Randall foil is very dependent on the angle of the top edge of the blade onto which it is fixed? If diving one's oars or sculls is as costly as your presentation suggests, wouldn't an alternative design be a shorter wing shaped surface riding foil fitted at an upward angle to the outboard quarter or third of the blade face, so that the rower could only bury the blade at the catch by massively forcing the hands upwards? Not sure whether a penalty would be paid at blade extraction though... While the surface riding foils would stay at the surface at the catch, when the flow over them reversed later in the stroke, would they get forced back under again and then be a pain to extract, or would the much more turbulent water the blades are working in later in the stroke mean that it's not an issue? Maybe you'd need to have a larger upward kick at the end of the blade top to accomodate it although that's likely a whole blade design rather than an additive component. So maybe something like your shelf type design but fitted to a blade top the shape of a shallow smile, that way the leading edge has rocker keeping it above the water like the front of a surfboard at the catch, and then again from the other end when the flow reverses later in the stroke.

I thought about this stuff quite a bit a few years ago when I was doing a lot of sculling- at one point I double sided taped pieces of 2 inch deep by 3/4 inch thick closed cell foam strip to the top of my blade faces- my thought being that even a small amount of buoyancy so far away from the boat would help with diving at the catch and also help with balance in rough water (coastal sculler). In some ways similar in sentiment to the Randall Foil. It was not really noticeable in either aspect whether they were on or off the blades and I took them off when something started eating the foam (rats perhaps? - I used to store my blades in the garden). I went for buoyancy rather than a 'foil' type device as it was my belief that for much ot the stroke the blades shunt back and forth a bit and twist in a stationary puddle, but I finally concluded that the volume of buoyancy you'd need to actually make a difference to the diving would be unmanageable out on the blades. I managed to cure my diving with the tried and tested tricks of technique changes and stern pitch.

I liked the benefits that the Italian sculler found and appreciated that the commentary picked up on the device. Looking at his strokes on the video though he didn't seem to be rowing in a particulary shallow fashion. The foils didn't appear to be sitting at the water surface (if that is their intention?), but hard to tell for sure unless you had a following camera.



John E

unread,
Apr 7, 2021, 5:00:29 AMApr 7
to
I've just seen this discussion and found it interesting as we have kept ourselves motivated during lockdown by developing a few products to add to our (Active Tools) product range, one of which gives an aerodynamic drag reduction. We started looking at Height Spacers to try and solve the issue of swapping them one handed when out on the water and the answer to that seemed to be a ‘pull tab’ and as that inevitably elongates the design we then wondered whether making the nose of that tab ‘aero’ (as the cycling fraternity say) and adding a Kamm tail (again, very common on bike frames https://www.trekbikes.com/gb/en_GB/inside_trek/kammtail_virtual_foil/) might give a small drag reduction.

What we found, using CFD analysis, was that a short stack of these had a drag of only 19% of that of randomly orientated conventional spacers and that that would give a top flight Eight a 0.4 metre advantage over a 2K race... The caveats on this are that we did not look at cross wind performance, although the is no reason to believe that will be poor, and forward mounted riggers, or riggers with C cups that shield the Oarlock/Swivel, would not benefit from using these.

This is what the device looks like and we have initial samples from the tool on the way to us now https://drive.google.com/file/d/1GkUtmQcqGHZydN4QO5n1TIOqhh8bSrU_/view?usp=sharing

The 0.4 metre figure makes one wonder what gains you would get by extruding aerodynamically profiled backstays, or adding thin Carbon Fibre covers to standard ones, and also whether a profile with a Kamm tail might work well for oar shafts?

On another point, in response to a comment in the thread, I believe that all the major manufacturers make their oar shafts from Pre-preg material? The spiral lines you see are the witness lines left after the Nylon tape they wrap them with before curing is removed.

Peter

unread,
Apr 7, 2021, 5:37:22 AMApr 7
to
I assume someone has looked at the drag factors of the crew themselves? - Hair, clothing, sunglasses and hats and cox positions - particulary in cross-winds...
pgk

Jake Frith

unread,
Apr 7, 2021, 10:46:38 AMApr 7
to
John,
I would imagine a Kamm tail profile could work very well indeed for oar shafts if Trek's claim of comparable drag to a 1:8 foil is correct.
In fact I think you, or perhaps Trek, or whoever they got the idea off might have stumbled upon the next generation of rowing oar.

Think of the advantages:
1) You could get the correct orientation of stiffness on the drive from an oarshaft of that profile without a large weight penalty. (It might even be lighter than a round shaft - The flat could actually help your drive phase stiffness- like it does Trek's lateral frame stiffness)
2) The penalty in the water of burying the loom would be much smaller, so you might be able to extend the aero section all the way down to the blade. (Not have to end it a foot early and lose all the aero where you actually need it most like in the fixed arero shroud shown somewhere above)
3) Similarly the shorter tail would mean you would not have to worry about contact with the riggers like you would with a long foil section, so you could continue your aero right up to the collar. (and upwards of the collar as far as the handle if you wanted). More of the potential issues of conventional foil shrouds neutralised.
4) The fact that these types of foil have better performance in misaligned flow suggests you might be able to do things like deliberately arrange it slighly misaligned on the recovery, so it's less badly aligned on the drive. Depends on what misalignment these Kamm profiles would absorb before the drag starts climbing. Because it's a short foil profile, even squared (90 degrees out of whack), it's going to be way better than a conventional long tail foil profile.
5) It will work as a fixed part of the oar shaft (so within FISA rules)
6) The technology is not IP protected and not patentable (been around for years- such as on Kamm tail sports racing cars of the 1960s). Plus I note Trek did not try to patent it- which suggests someone else thought of it for bikes first. For FISA, this is an advantage- they do not allow patented technologies. If there was a Rowing specific IP to be had (unlikely), its been blown by this thread anyway (publication- prior art).
7) In the way that Trek has found out that this is less of a 'wing' than a full length foil (so it does not get moved as much in sidewinds), any problematic behaviour of the looms of the oars catching the wrong bit of air, creating lift and trying to take off, or plumet into the water might be usefully reduced.

It's a known fact that fatter foils will take a larger angle of attack than thinner ones before stalling (hence sailing dinghies having a relatively skinny centreboard and a fat rudder profile). In combination with this I remember reading somewhere that at certain flow speeds, truncating a foil like this sometimes does not majorly increase drag, as the air pretty much describes the same path it would have if the back half of the foil was present. In a wind tunnel the upper and lower air streams desribe a foil shape, coming back together in much the same place it would if the back of the foil was present. Reading between the lines of their marketing waffle it looks like what Trek has found out is if you have a fat foil and get rid of the back of that foil, the air will then describe its own foil type shape, only aligned better to the flow than when a full foil is redirecting the flow in a direction it doesn't necessarily want to go- so for them better in crosswinds. So in the context of a bike in a crosswind the Kamm profile effectively provides a 'virtual foil' correctly aligned to the aparent wind. But similarly an oar is not always perfectly aligned with the oncoming air flow- particularly when alternating between squared and feathered.

If your profiled height washers buy 40cm over 2K, try riggers and 8x oar shafts in a CFD package!

Some potential there I'd say...



Andy McKenzie

unread,
Apr 8, 2021, 4:53:59 AMApr 8
to
I had to look up 'Kamm Tail Profile' and when I did I sort of had to laugh, because I thought 'I'm sure I've seen these in the back of the boathouse'. Old wooden hollow blades have almost exactly this profile, although from an aerodynamic viewpoint 90 degrees out of phase, as the flat back is orientated sternwards during the drive, and towards the water on the recovery. Still - should deal with the 'prior art' element of the patent lawsuit.

Jake Frith

unread,
Apr 8, 2021, 6:24:51 AMApr 8
to

> I had to look up 'Kamm Tail Profile' and when I did I sort of had to laugh, because I thought 'I'm sure I've seen these in the back of the boathouse'. Old wooden hollow blades have almost exactly this profile, although from an aerodynamic viewpoint 90 degrees out of phase, as the flat back is orientated sternwards during the drive, and towards the water on the recovery. Still - should deal with the 'prior art' element of the patent lawsuit.

Yes, I thought that too. I have an old pair of Suttons sweeps as bannisters in my house. Apart from being 90 degrees out of whack, they are a bit too rounded at the edges, but there is a similarity there. I guess they did it for ease of construction, and like a laminated wooden longbow, the wood on the flat back section is a darker, harder and presumably stiffer/ whippier wood. I guess it also helped in that they didn't have to have a complex seperate moulding for the collar- they were just a section of white heat shrunk plastic that conformed to the shape, mainly held in place by the bolt on collar and a bit of electrical tape top and bottom.

Andy McKenzie

unread,
Apr 8, 2021, 8:52:52 AMApr 8
to
On Thursday, 8 April 2021 at 11:24:51 UTC+1, frit...@googlemail.com wrote:
> > I had to look up 'Kamm Tail Profile' and when I did I sort of had to laugh, because I thought 'I'm sure I've seen these in the back of the boathouse'. Old wooden hollow blades have almost exactly this profile, although from an aerodynamic viewpoint 90 degrees out of phase, as the flat back is orientated sternwards during the drive, and towards the water on the recovery. Still - should deal with the 'prior art' element of the patent lawsuit.
> Yes, I thought that too. I have an old pair of Suttons sweeps as bannisters in my house. Apart from being 90 degrees out of whack, they are a bit too rounded at the edges, but there is a similarity there. I guess they did it for ease of construction, and like a laminated wooden longbow, the wood on the flat back section is a darker, harder and presumably stiffer/ whippier wood. I guess it also helped in that they didn't have to have a complex seperate moulding for the collar- they were just a section of white heat shrunk plastic that conformed to the shape, mainly held in place by the bolt on collar and a bit of electrical tape top and bottom.

Remember that prior to heatshrunk plastic the collar would be riveted leather.
Reply all
Reply to author
Forward
0 new messages