The Costin Swift Speedbike

The Costin Swift Speedbike

In 2012 Sean Costin started work on a new speedbike that would be raced at Battle Mountain and that he would use for years to come in HPRA racing. This bike would use Sean's Velokraft NoCom as the chassis. Sean collaborated with Raymond Gage to who designed a shell inspired in part by Matt Weaver's Cutting Edge design which incorporated flexible membrane( sheet latex) that allow the knees to protrude from the main hard shell during pedaling. This design allowed for excellent vision of the road while retaining a very nice aerodynamic profile. This bike is now the World's fastest streamliner based on a production recumbent bike.

The bike was not exactly stock equipment. Aside from other modifications over the years , for this project, Sean narrowed the bottom bracket to 60mm and moved the idler forward so that there would only be one idler for the huge 104 tooth chain ring. No jack shaft was required which improved the drive train efficiency.

The shell was designed by Raymond Gage using Solid Works. For production, Raymond digitally sliced the shape into sections which would be cut out of 2" foam and reassembled. Dr. David Sianez of CCSU (Central Connecticut State University) cut the foam shapes on his large format CNC router. The shell was laid up by Sean and Dan Ayala with carbon on the outside and then the foam cut out and laid up inside. Mountings were extremely simple and light. Threaded rod extensions from the rear drop outs and a tie rod from the boom to the shell that tensioned the lower half of the shell against the bottom of the bike.

4/15/2012 Sean writes:
"Raymond Gage is close to finishing the shell design for my new bike and he has a few pictures to share. The design concept was inspired by Matt Weaver's Cutting edge bike. Perhaps with improved tools, knowledge and experience we make a very fast bike.

I'm steering away from camera vision and looking to explore the realm of the knee pop design. This will involve some tinkering once the shell is complete.

Matt Weaver's knee pops were made with a latex skin on the exterior and a lycra underneath and I plan to use the same.

We are working to make this bike as small as possible. Raymond was able to model my pedal stroke base on video I sent so we could tighten the pedal box as much a possible. I'm using a 60mm crank spindle and 150mm cranks"

"We are just about at the stage where water jet cut pieces will be made from files generated from this model. This should improve accuracy and speed the building stage. That is unless someone offers to CNC the whole thing.('')

I sincerely thank Raymond for creating a much more involved and expert model than what I had envisioned. I feel very fortunate that he took me up on my request to make this dream a reality. Though the pictures look effortless, I don't think they can convey the software challenges he faced and the multitude of iterations he explored in pursuit of this design."

5/17/2012 Ramond Gage writes:
"I had Sean video his pedal stroke through many revolutions, and send me his shoe measurements. I then picked frames out of the video that represented the worst case, mostly from heal drop, and used those frames to generate the models shown in the image here I then created an offset surface of my fairing design to check clearance to a certain tolerance. This is an evolved approach from what I used on the Orion, which worked out pretty well.

Of course, if I got any part of that process wrong, printing models of it wouldn't tell me anything, as they would be to that error, instead of reality. Cutting out full size templates that Sean can mount up to his Nocom, and then pedal in, should tell us all we need to know."


Raymond sent the design to Dr. Dave at CCSU for 3D printing.

6/6/2012 Sean writes: "The CNC process has begun on our shell. Raymond Gage has been racing against the clock to complete the design and slice it into 2" sections which can be cut on a 8 x 4 x 5" Z machine under the direction of Dr. Dave at CCSU.

What is particularly nice about this method is that the parting line for the top and bottom half is already built to a slice and the interior will almost all be easily popped out due to perforations that are holding everything together during the shipping and the outer layup.

Relatively light trim work of the foam will be necessary to get the interior ready for an internal layer of composite.

Here is a screen image and a 3d print of the design."

Here are the CAD slices sent to Dr. Dave at CCSU.

Slices Exploded

Slices Together

"non-virtual fairing parts from Dr.Dave at CCSU!
Looks like I won't need to do any sanding on the exterior.

Thickness surfacing will be done last. Note the slight lip at the top edge. "

CNC'ed fairing slices completed.

6/24/2012 Sean writes:
"I met with Dr. Dave Sianez and picked up the mold yesterday from CCSU. Dave is pictured with the bike sort of fitted together. We took a few pictures, shrink wrapped it up and popped it in my truck. Everything looked really good and now I'll be doing some touch up sanding of flash and hopefully I can do the first layup next weekend. The transitions are really good- top and bottom. Far better than what I could have done by hand. I'm really happy with the way this is turning out. "

Sean bonded the layers of foam together and gave the foam plug a light sanding.

"Bottom half almost ready for carbon"

7/21/2012 - First layer of carbon.

Raymond writes:
"The fairing was designed around the 155mm cranks and narrow BB Sean is currently running on the Nocom. There is no need to have heel and toe holes in the fairing with this setup, unless I screwed up the design or the measurements weren't precise enough. The only holes planned are for the knees. If Sean could put shorter cranks, say 130mm, then the knees would not pop up through the fairing as much, but he would need to lengthen the boom again to account for proper leg extension. I would be curious how short the cranks would need to be in order to keep the knees entirely inside the fairing, and what/if the power drop would be. That might be a future design study."

7/24/2012 Sean writes:
"Grand Canyon of foam. Those inner sections just pop right out once I hot knifed the perforations. "
"It's so nice to have that parting edge. As much as it was difficult to have the plug made up of so many slices, it made today much easier. "
"Moving right along now with great motivation. Everything seems to be fitting perfectly thanks to Raymond. Inside foam nearly finished to prepare for the inner layer of carbon. I hate to cover up the light blue. It looks incredible against the black carbon. I think I'll paint the carbon light blue if I get the chance. The four blocks will have some sort of alignment pins for the top. With the inner cuts of the foam from the perforation as a guide it was relatively easy to sand the interior to a consistent thickness. However, after 3 hours of continuous sanding my entire right hand and forearm cramped up. It was time to take a break and make some pictures."

"Earl Russel dropped of my new rear wheel with the Capreo hub yesterday which will give me a 9 tooth rear cog to go with the 104 T front Chainring ( could be wishful thinking)
We're going to make some wheel covers and Raymond will dynamic balance it. "

Front view

7/29/2012 Sean writes:
"No name decided on yet. Keep the suggestions coming. I just put another layer of carbon on the inside and it's far stiffer than the Coslinger The weight is part of a guessing game currently, but I can tell you that the total bike weight should come in the ballpark of 45lbs. "

"The bike is now mounted to the lower shell and I've done some test runs without the top. I'm taking it to the velodrome on Wednesday for some higher speed stuff."
"Big honking holes."

"1/2 hour in the velodrome with a self start no less ( not easy) . Chain noises to resolve, Shave off a little here and there for one knee, but the bike handles great. Crosswinds were limiting my comfort factor at speed, so 34mph was my top. With no windshield, no covers for the knees, no body work and no tape. The bike was doing just under 33.7 mph on 140 watts steady state. I'll need a drag chute to get any serious training in. BB held steady. I felt by the knee holes while riding and very little wind comes in there. Knees only stick up less than 2" .

I felt by the knee holes while riding and very little wind comes in there. Knees only stick up less than 2"

Into the final stretch. "

Video of testing Sean's speedbike on the Nothbrook Velodrome


9/07/2012 Sean writes:
"This photo taken a few hours ago after high build primer applied. Do you think I'm cutting it close or what?  Rendezvous with Ollinger and Mowett in La Salle Peru, IL in early AM.

Looking just like the computer model "

Sept 2012 - Sean strikes a pose in front of the Swift on SR305 near Battle Mountain, NV

More pictures on Sean's Photobucket site

Sept 2012 - Sean sets a personal best of 70.4 MPH at the 2012 world human powered speed championship.

Sean was instrumental in the creation and organization of the world human powered speed championship starting in 2000.

Picture by Wim Schermer

9/15/2012 Sean writes from the Battle Mountain WHPSC event:
"This is my power tap reading for Thursday night which I broke 70mph for the first time. We came to the realization that If I could gain more speed very early in the course and build with relatively low watts, I could rest a little and get my breathing under control before the big push at 1 mile to go. At the road inflection at 2.5 miles to the finish, I put it in a higher gear and pedaled my rpms's up from there. Having power data was extremely helpful at improving our speed and optimizing our power usage on the course so that we had an improvement nearly every run."

"Ultimately we built a world class streamlined bike out of a production lowracer with modifications to the BB to make the bike narrow. We demonstrated an efficient fabrication process that minimizes build time without compromising quality. Key features are Datum mounts with funnel shaped locating pins, Flat wrap wind shield, latex knee pops, quick mounting release and tensioned front fairing mount."


Picture by Wim Schermer

"Our team developed rather naturally and it grew out of specific tasks required and areas of expertise. This could not have been done without this combined effort. This bike has superb all around characteristics for a streamliner and I'm looking forward to racing it for many years."

Sean and Raymond Gage are happy after Sean's 70 MPH run.

Fast forward to summer 2013

Sean raced the Swift at the June 2013 Michigan races at the Waterford Michigan sportsman's park. The bike ran well on the twisty course with a big hill.

Picture by Dan Glatch

Here's Sean while working on the bike in the pits during the Michigan races.


Picture by Dan Glatch

Rich Meyers launches Sean during a test run at the Michigan races.

Picture by Mike Mowett

Sean also raced the Swift at the Northbrook and Kenosha velodrome races in July. Sean did well in the 100 lap race with an average track speed of 35 MPH. Actual bike speeds were a couple MPH higher.


In September 2013 Sean raced the Swift at the Iowa Hawkeye Downs 1/2 mile track. Sean's Power tab hub showed that his speeds in the 24 lap race averaged close to 40MPH, bike speed. It also showed his technique of cranking around 400 watts in the straights, and coasting through the corners.

Picture by Steve Fuller

Here's a good picture showing the gigantic front chainring and the cone body locator devices that easily align the top and bottom of the shell.

Picture by Steve Fuller

Here's a picture of the rear of the bike in the shell showing the rear mounting and the close tolerances.

Picture by Steve Fuller


Back to Sean Costin's projects