gregf83 

An interesting discusion. I don't have a bent bike but am curious about how much of an advantage you could potentially have. Doesn't anyone with a bent have a power meter? That would seem to be a simple way to settle this discussion. If you can output 300W at threshold on an upright bike can you also output the same 300W on a bent? If you can then the online calculators provide a pretty good estimate for the performance improvement. Am I missing something?

Jeff Wills

Some of them have power meters, but for most recumbent riders it's silly to have one. As several people have pointed out, there aren't that many people racing recumbents, so there isn't that much incentive to train with a power meter. A better place for this discussion would be over on the recumbents.com "Technical" forum: https://www.recumbents.com/forums/forum.asp?FORUM_ID=5 - there's a few people who ride different bikes at a fast pace.

There's also an database of recumbent bike speed records: https://www.recumbents.com/wrra/records.asp . It shows that the hour record for unfaired recumbents is at 32.3 miles- granted, that's with a disk rear and front tri-spoke wheel, but still darn fast.

My 10-mile TT time is at about 26 minutes, but that's with indifferent (if any) training- about 23 mph. At one Human Power Challenge 1-hour road race, Rob English was passing me every other lap - he was riding 3 laps for every 2 of mine. I have to assume he was averaging close to 30mph for the one-hour race. Granted, his machine is a lot lower than mine, but bloody heck that's demoralizing.

RChung

Though you'll have to admit that even there very few use power meters or were interested in measuring either their rolling or aero drag. The few data that are available suggest that most unfaired recumbents don't have total drag lower than a well-positioned rider on a good TT bike. The exception may be low racers, though I have not seen reliable estimates of CdA there at all so that's just my speculation.

Some of the claims made in this thread of a 17% increase in speed due solely to decreased aero drag would require that the CdA of a recumbent be just short of half that of the rider's CdA when on an upright. That's believable only if the rider's upright position were especially poor or the recumbent position were especially good, or both.

purplepeople

Here are photos of your typical carbon TT bike and a carbon low racer, both of which have a 700C rear wheel. Using the equal size rear wheels for scale it's pretty easy to estimate that the frontal area is about half for the low racer. Even at 60% of the frontal area, we're well within 17% increase in speed since CdA is inversely proportional to cube of velocity when keeping power constant.

AFAIK -> 1/(1.17^3) = 0.624

So, since Blazing is riding a similar low racer, I can believe he is getting his 17% increase, especially if he is comparing to his old DF bike, which probably was not a late-model aero TT bike when you consider that he has been a recumbent rider for many years.

:)ensen.

cod.peace

What he said, pictorally:

Laurel Lane

Jeff Wills, about the record one hour time, take those discs off and run spoked wheels. Then you lose 1-1.5 mph. Now you are back even with a conventional dropbar, spoked wheel DF at 30-31 mph.

I'm not a trained monster rider but my time trial times on a cyclocross bike with 32 slicks is the same as your recumbent 10 mile times. And I suspect you are a male. I average 22.5 mph for 11.3 miles and I'm a woman. My best speed was 23.1 mph when half the course had a big west wind.

I like the bent but remain unimpressed when you match the two frames together with the same wheels and no fairings. This does not make a bent bad, just different. (more comfy)

RChung

Um, no, I'm afraid you're mistaken. That was a photo taken on 23 Jan 1984. That was the 51.151 km record. You can tell that was taken on 23 Jan 1984 and not four days earlier on 19 Jan 1984 (which is the first time he broke Merckx's 1972 record with 50.808km) by the name on the file: it's labeled "moser_230184.jpg"

RChung

I believe I wrote: "[...] most unfaired recumbents don't have total drag lower than a well-positioned rider on a good TT bike. The exception may be low racers [...]". Are you suggesting that the drag area of low racers is representative of the drag area of most recumbents? (BTW, notice that Boonen is riding a bike with spoked wheels). You can show all the photos you want but the bottom line is that the eye isn't really good at estimating CdA since CdA is not the same thing as A. For example, tailbox fairings on recumbents have a big effect on drag even though they have essentially no effect on A. Likewise, lenticular disks on conventionally tubed bikes work better than flat disks even though their A is increased. You and I recently mentioned some data for a semi-faired racer [edit: I don't know if it qualifies as a low racer or not]. That rider's CdA was estimated at around .21 m^2. I've got data files for well-positioned riders on TT bikes with CdA in the .19-.22 m^2 range. Unfortunately, as I said, I have no data on unfaired low racers.

Laurel Lane

I believe you about the photo. But the list of records set by various riders specifically called out what they were riding. It said spoked wheels, both wheels. The whole point of the list I was reading was to show how the bikes compare. What Moser did at every bend in the road is unknown to both of us. I do know he used all kinds of equipment at different times. Your photo is nice but it's irrelevant to the topic. You could probably find a photo of him on a motorcycle. What would that prove?

RChung

Well, it's hard to deny the evidence: what it proves is that the list you've been relying on is unreliable.

Laurel Lane

No, goofball. It means the photo is irrelevant. Moser used all kinds of equipment. He set the record in a class specifically calling out spoked wheels. If you are too stupid to understand that statement I cannot help you. Don't get so defensive about a bike. God Almighty are you dumb!

RChung

Perhaps so. However, in this case whether I happen to be dumb is also irrelevant; the list on which you have been relying is unreliable.

atom bomb

There's one thing that cannot be disputed. The UCI banned the recumbent riding position in bicycle races under their jurisdiction because it offered an aerodynamic advantage over the conventional upright riding position. Presumably they knew what they were talking about because they're in the bicycle racing business, so that's good enough for me. And that was in 1934, so this is old news, no?

That L.L. has only ever witnessed slow 'bent riders and passed them easily doesn't surprise me given her fitness and the statistical possibility of coming across a fast 'bent rider in her area when there are so few 'bent riders to begin with. Most of the (very few) 'bent riders I have come across on the road were also pretty relaxed in terms of speed, too. Just like with uprights, there are riders all across the spectrum that you will encounter on the road.

Goodness, gracious... it's all explained here:
https://www.youtube.com/watch?v=1ZhozlAvGdk

57rebike

When people use the term "Recumbent" in a question about speed, the term is too broad to provide a meaningful answer. The ONLY thing recumbents have in common is that you are generally seated in something more like a car seat and less like a bicycle seat. One can safely say that because of that ALL recumbents are MORE comfortable than DF uprights.

As for speed, the conversation needs to be split several times:
1) Trikes (3WR)
2) Two wheeled (2WR)

I don't know enough about Trikes to comment about how they should be further subdivided. I'll leave that up to the Trike experts.

In the two wheeled Recumbents (2WR) there are:
2a) Short Wheel Base (SWB)
2a1) 26x26, High Racers (HR)
2a1a) SWB Under Seat Steering (USS)
2a2) 26x20

2b) Long Wheel Base (LWB)
2b1) LWB High Bottom Bracket (HBB)
2b2) LWB Low Bottom Bracket (LBB)
2b3) LWB Racers. . .Rider rides just off ground level with axles at least 1/2 wheel diameter behind riders spinal chord and at least the same distance ahead of crotch.

Whether a Recumbent is faster than a DF varies with weight and wind profile.
Weights vary with basic design features. . .
(Trikes pretty much have to be heavier than two wheelers)
(I don't know what the weight spread is between a heavily carboned trike and a cheap steel recumbent)
Weight will have a huge (negative) effect climbing and a benefit descending. Very small weight penalty on the flats)

Wind profile matters at speed. The higher the speed (whether through rider power or downslope) the bigger the advantage for a small wind profile.

From that standpoint, short of fairings, nothing beats a LWB racer, especially fully faired.
Getting more conventional, the SWB:HR is a close second.

I'm sure we could work through opinions and sort the list, eventually reaching the conclusion that there are some recumbents, that on a typical mission, even with a fully trained rider, there is no gain, in fact a penalty. Those bikes are ridden simply because they are more comfortable. Not a thing wrong with that.

I'm told that USS has a bigger wind profile than a HR, but I haven't seen data.

I am a couple months into my Volae Expedition, a 2WR:HR. It is absolutely faster on the flats and down hill, and absolutely slower on the long slogs uphill. Rollers are getting better as my legs come in, but I don't think it will ever climb a long slow slog as fast as my DF upright will.

So. If your typical mission is fairly flat to rollers, there are many recumbents that will add speed. There are many more that will not. On long (<8mph) slogs, I don't believe there is recumbent on the planet that will match a descent DF watt for watt.

I believe all recumbents are more comfortable than DF. I did my first Recumbent Century (101.5) yesterday, and my neck, arms, wrists and all parts between my belt and my knees are totally comfortable. I'd guess my speed is about 2-3 mph faster all in. Unfortunately my new bike computer does not segregate stop time, so I cannot compare ride speed directly.

57rebike

chucky

I'd like to see this too. My pet theory for tying together all the facts is that the wattage is probably lower, but I'd love to see measurements which confirm it.

Yeah, but is it fair (no pun intended) comparing a well-positioned rider on a good TT bike to a poorly-positioned rider on a recumbent? The problem is that anyone that cares about aerodynamics on a recumbent knows to use a fairing, so we're comparing apples to oranges (well-positioned riders to poorly-positioned riders).

Also, where you do draw the line on what constitutes a fairing? Is an aerohelmet a fairing? Disc wheel? Consequently, should you be comparing a TT bike with an aerohelmet to a 'bent with a tailbox, etc? In the end all the only thing we can say is that the fastest possible bikes (without artificial rules) are recumbents. Everything else is semantics.

RChung

Possibly not, but I don't know enough about positioning on recumbents: I get the impression (though it's just an impression) that aerodynamic drag is thought of as a design issue for recumbent riders and that once the format of the bike is decided then the rider's position is relatively fixed. OTOH, conventional upright bikes (even TT bikes) have restrictions on design and form so rider positioning is decidedly unfixed and by far the greatest source of drag. IOW, on the same TT bike a rider can be poorly-positioned or well-positioned and the people who care about speed go to the trouble to position themselves well, and use wind tunnels and power meters to double-check that. Recumbent riders who care about speed don't seem interested to the same extent in measuring their drag. As I've made clear earlier, I'm not at all knowledgable about recumbents; I do happen to know a bit about power meters and measuring rolling and aero drag in the field.
That's a good point, but it emphasizes rather than refutes my point above: TT helmets turn out to be very position-specific, so TT riders spend a great deal of time testing different helmets. For example, some TT helmets work better for a flat back while others work better with a slightly different head position. Meanwhile, tailboxes are fixed in position. Once again, it appears that recumbent riders think of tailboxes as a design element while conventional TT riders think of aero helmets as dependent on positioning.

purplepeople

Interesting that assumptions are being made about the design parameters. As shown by the sheer variety of recumbent models, from the heavy models with upright seats to the low racers with disc wheels and smallest frontal area.

As a builder I can say that recumbents come in different models with specific purposes for often unrelated reasons. Many homebuilt recumbents are LWB simply because there are the easiest to build. The SWB low racers are the hardest to design because there is very little room (for some riders!) between their legs to put the fork and front wheel drive. While the overriding concern for most designs is comfort, for a select few, speed is the goal. To that end, the current debate is whether a low racer is faster because of it's aerodynamics vs a high racer with the lower rolling resistance of it's dual 700C wheels.

As agreed, not having the data is the critical problem here. Moreover, a big part of that problem is that rider physique plays a huge role in performance. But lack of evidence does not make our assertion untrue, only unproven. As further proof, I offer my experience with downhill running. Understanding that total vehicle weight plays a small but largely insignificant role, we all know that the best DF descenders on a straight downhill are those that can contort themselves into the best aerodynamic shape. And yet, just coasting, I begin to outrun them on my low racer. Without power meters or wind tunnel tests, it's the best way I know of to show that my ride is that much more efficient at speed.

@57Rebike: The low racer with the wheels between the rider's legs are considered SWB, even though the wheelbase may be approaching 60". In general, the nomenclature refers to whether the front wheel is ahead or behind the cranks.... and why there is another configuration known as MWB (or CLWB) in which the cranks are very near the head tube (and yet still do not have interference with front wheel swing).

:)ensen.

RChung

Wouldn't actually measuring the drag between those two designs help to quell that debate?

chucky

I wasn't trying to refute you. I agree with you and was just emphasizing, as you said, that in the recumbent world aerodynamics is typically tackled more in the design of the bike.

On the one hand this is perfectly understandable because that's obviously the best way to beat the records, but on the other hand I remain frustrated because I think the records are very poorly designed tests of vehicle speed. While in reality I think that fairings present a major hinderance in actual vehicle speed due to the difficulty of accelerating the extra weight, I don't necessarily think that simply barring fairings, disc wheels, or other aerodynamic aids makes for a good test of speed either. IMO something like RAAM should ideally balance all the relevant factors, but it's interesting to note that we don't see recumbents typically coming out ahead there either.

purplepeople

Yes.

But I am not surprised that one of the mfg hasn't yet done this. AFAIK, there are only two mfg of carbon low racers that I know of, Velokraft and Zockra and together they might make 10 of these each year on custom order, so there would hardly be a spare bike to take to the wind tunnel.

:)ensen.

chucky

One of the tricks for making an upright TT bike more aerodynamic is to bring the rider's legs closer together. Is it possible that an optimally positioned rider on an unfaired high racer could be more aerodynamic than on an unfaired low racer? Even aero bars were not used on uprights until the end of the 1980s, IMO the recumbent world has a long way to catch up in the use of aerodynamic rider positioning.

purplepeople

Actually, no. At the bleeding edge, riders are using low-Q cranks and custom 2" hubs to get their cranks past the front wheel. They are even use smaller cranks lengths, typically around 155 mm, which give an advantage of smaller swept front area and also keep the knee in it's highest leverage zone. Many of them are riding with their heads so low that they can just see past their knees. I've heard talk of using a remote vision system on an unfaired bike just to get the head behind the leg shadow. On at least one bike, the handlebars have been modified for the straightest possible arm position in front of the head.

Everybody takes from every possible source. For instance, what works for the streamlined speedbikes also works for the low racers. For instance, the Varna Tempest is actually less wide than it's rider, who must contort his shoulders to fit. Consequently, there is now talk of using techniques similar to the compression swimsuits to physically squash the lowracer pilot into a smaller size.

Bear in mind that I'm not anywhere close to being an expert on this stuff.... those people are all over at the WISIL forum, where the discussion might sound similar to talk in the labs at Cervelo.

:)ensen.

chucky

I'm sure that's all true, but the lowracer pics you posted above look to have very wide Q and the knees in particular are very far apart. Just sayin...

RChung

Um, that's past the "talk" stage for TT racers. There are those who already use shoulder girdles or "shrug bras." In this case the aero advantages have to be balanced against power loss (a similar thing applies for low q-factor BBs and "circus cranks", i.e., cranks around 140mm or so); but the point is that racers have actually been doing field testing with power meters to measure the net effect on their races.

purplepeople

Wow, there really are a lot of misconceptions and pre-conceptions about recumbents. While full fairings do present an extra load on the rider during acceleration, they also offer an inertial advantage on slight inclines and a major aerodynamic advantage in most wind situations. While it is not possible to sprint in the same way that an unfaired bike can sprint, for all-out speedbikes, the limiting factor in acceleration is the transmission which itself must be limited to only a few ratios to keep vehicle width to a minimum.

For example, here are Sam Whittingham's lap times for the 1-hour record he set this summer at the Ford 5-mile oval. Note that the final distance ridden was about 90.6 km over the hour.

Lap 1, 74.70 kph
Lap 2, 91.02 kph
Lap 3, 91.99 kph
Lap 4, 92.15 kph
Lap 5, 92.21 kph
Lap 6, 92.09 kph
Lap 7, 91.74 kph
Lap 8, 93.02 kph
Lap 9, 93.13 kph
Lap 10, 93.03 kph
Lap 11, 94.38 kph

Since it was spent accelerating from zero, it's clear that with an average of 80% of his target speed, he is likely very near his record average by the end of that first lap. Under full acceleration he brought that same machine to over 133 kph within the 5-mile run of the Battle Mountain speed course.

If you really want to test the speed differences between different bikes, the only way to do that is to put the same pilot on each one, use a calibrated electric motor for the motive power and have a go. This way, we are not measuring the human/machine combination but the vehicle by itself.

Barring aerodynamic aids is not about speed but athleticism. The UCI has clearly given the message that their sport is about the athlete and not the technology, never mind that aerospace technology is required to be competitive. I'm quite sure that lighter and more aerodynamic bikes magnifies the difference between riders so limits on the technology serve to level the playing field, with the clear effect that UCI racing is much more exciting from a competition perspective.

Unlike the UCI (or the USCF for that matter), recumbent racing is clearly about the absolute performance capabilities of the human/machine combination.

Interesting you mentioned RAAM, since the recumbent teams are not only beginning to place, but win. IIRC, last year, John Schlitter rode a Bacchetta to first place in his 50-plus category and this year, Rans took the 4-man team honors. And never mind that Lightning and Easy Racers fielded streamliners against each other some years ago, with Lightning setting some crazy fast record for the crossing. From what I heard, the inside scoop was that Easy Racers dropped out about halfway because the downhill speeds getting so dangerously fast, the support vehicles couldn't keep up... something like 85-90 mph.

:)ensen.