> 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.