Buyers Guide Forums Information Buy/Sell Communities Build/Tech Racing Site
MARS projects - VFS - Finished streamliner
What's New



Mailing List







Click HERE to go back to the top Vacuum Foamshell page
Click HERE to go back to the main Projects page
Click HERE to go back to John Tetz's Projects
Click HERE to go back to the Foamshell Links Page

Finished Streamliner

Basically, I am pleased with the vacuum-formed results. The nose came out very well, stiff and stable, helped by all those compound curves. But when I had to spread the shell halves apart to clear the handlebars and feet (because of the triple crank), it caused a distortion which shows up as horizontal waviness through the middle. I spread only the middle, not the nose, as I didn’t want to make the nose too fat and blunt. Therefore this buckled the middle, but luckily the waviness is horizontal, in line with the flow. The foil shape through the middle is somewhat flat, which aggravates the waviness, since flat surfaces are inherently weak. It’s a shame, because the purpose of doing all the vacuum-forming work is to get smooth foil shapes. The good thing is that the waviness is well beyond the transition to turbulent flow, and turbulent flow is less affected by unevenness.

An outline of one of the internal braces is also visible half way down. This didn’t happen on the OFS, so I recommend gluing a piece of HD across the open U-shape of the brace. This would also help stiffen the brace and increase the glue surface.

I tried to heat and push toe bumps in the nose, but couldn’t make them large enough. So I made a quick cutout in wood, clamped a sheet of Zote, and used heat and air pressure to form the toe bumps. Oh, how I hated to cut big holes in that smooth nose! Then I had to heat and push heel bulges at the bottom. I also had to use hand-held local heating around the handlebars to get just a bit more clearance. The great thing about foam is that it can be heated and reformed into what you need.

I added adjustable blue NACA (National Advisory Committee for Aeronautics) vents to cool the hands, as there are many blood vessels around the hands and forearms. Other cooling vents will be installed later.

Because the tail is removable, the fairing around the front of the wheel is short so the tail can be withdrawn backward, decoupling it from the internal braces. The tire helps form the leading edge of the wheel fairing.

Even with the shape and surface non-uniformity, this bike moves so much faster than the old OFS; I’m not quite sure why. I’m seeing speeds that I never saw in the OFS, easily cruising at 25 mph instead of working a bit at 23 mph in the OFS. Coming down Main St. in my town, I’m doing 30 mph for the entire length, whereas I could start fast in the OFS but gradually lost speed down to 27 mph by the end of a mile. Quit pedaling, and this vehicle rolls and rolls and rolls. I got this thrill in the very first streamliner I rode (Gerry Pease’s F-40 in 1991), and the thrill is still there after ten years of streamlining.

On the original Momentum molds, the rider was completely inside looking through a plexi bubble. In a road streamliner I feel it’s important to hear traffic, which means open cockpit, which meant I had to lower the area in front of the rider to be able to see the road. I did this when the shells were still in the mold halves. Although this nose shape is very sexy, I don’t like it as much as the squared-off nose of my OFS. With a square nose, there is no need for toe bumps or heel bulges. Where will I mount a decent headlight on this sloping nose? And a square nose is shorter by more than 6”.

But is it the nose shape that makes this bike faster than the OFS? Does the lift generated by the sloping nose do something to the air going through the gap between the bottom of the shell and road, such as canceling negative lift? I’m hoping it doesn’t. I’m hoping it’s the better treatment around the head and the slower tail contraction, along with a thinner tail cutoff and cleaner gaps, that make this bike faster than the OFS. Both bikes have similar frontal areas. The VFS is a bit lower, so maybe the front wheel being more inside the shell helps. The foil shape on the OFS looks a bit better with no flat section through the middle. The bare LFWD is faster than the RWD of the OFS. I plan to do some on-road comparison tests to see the actual differences, and will publish these results later.

This shows the tail contraction and a little of the helmet-to-turtledeck coupling.

A close-up of the helmet area. Note the mirror inside the windshield. I spread the windshield out just enough so the mirror would fit inside rather than stick out in the air. The mirror is about 3” from my eye, so it gives a wide view even though it’s small. The wider windshield helps reduce the wind noise a little. The turbulent air coming off the windshield circles in at the ears and goes forward towards the inside of the windshield, hopefully making a turbulent bubble, a kind of virtual foil shape with the air traveling over the outside surface of that bubble onto the turtledeck. Good theory, right? Need to do some tuft testing.

The cockpit.

This inside view of the cockpit shows my shoulders and arms up against the shell. The first few times inside it feels extremely tight and confining, but as the days pass it is beginning to feel more and more roomy, typical of streamliners. My body is learning where to be and what to do. Getting up off a seat that’s 7” from the ground takes effort. On occasions I have had to help people off the seat on the bare bike even though the feet can be spread apart and moved below the CG of the rider. Can’t move your feet like that in the shell, so it’s like doing a squat thrust to get up. Sure develops those muscles.

Another angle of the inside. My knuckles are actually not touching, and there is room for gloves in winter. Note the knee-to-derailleur clearance. I love the look of all that mechanical stuff. When laypeople look inside they are taken aback by the magic of it all.

The tail straight on.

The nose straight on.

Coming down the road.

I have to admit that working with foam is much more labor intensive than, say, Coroplast. It’s somewhat closer to the effort of making a hardshell. But for me it has enough redeeming features that I have concentrated on learning how to work with it. I know I can build some kind of a shell using simpler materials and get some results. It’s almost as if you can put a garbage can on a bike and get an aerodynamic improvement. I have been through plastic nose cones, a fiberglass nose cone with spandex to the tail (Lightning F-40), a tiny hardshell and, over the last several years, heat-shaped foam.

I’m at a place in time where I can work towards a higher level of both performance and aesthetics. First of all, I like the foam’s lightness. Because I live in a hilly area, and because I’m at an age (71) where I can’t put out the power, I’m forced into designing around lightness. At 7 lb., this is well below any other shell technique.

Also, crash protection is important. The spandex machines were reasonably light and gave reasonable performance but zero protection in a crash. Losing some skin led me into searching for better materials and finding foam. The hardshell I built was quite light, 11 lb., but turned out to be very fragile. I spent a lot of time doing body work (and each time adding weight).

I also like the quietness of the foam. That hardshell banged and rattled. And the foam is conducive to compound curves. But I don’t quite like the fragileness of the surface. Easily scuffed, and lack of reparability means once scuffed, always scuffed. However, grease and dirt do wash off easily.

Final comment: this thing needs suspension, so I have dug out the sketches I made of the RWD version I did at the same time I was designing the FWD.


shell front section 5 lb. 8 oz.
tail 1 lb. 6 oz.
turtledeck 7 oz.
total shell 7 lb. 5 oz.

shell length 90.5"
shell maximum outside width 20"
shell height in front of windshield 32.5"

To obtain the front/rear weight distribution data, I weighed the bike + shell, first with the scale under the front wheel and the rear wheel on a block of wood level with the scale, then the reverse, with the scale under the rear wheel, and added the two weights. Front/rear weight distribution (no rider):

bike 26 lb.
shell 7 lb.
bike + shell 33 lb.
front end 23 lb. 23/33 = 70%
rear end 10 lb. 10/33 = 30%

This shows that 70% of the bike + shell weight is on the front end.

I then used the same procedure and reweighed the bike + shell while sitting on it. Weight distribution (with rider):

rider 160 lb.
bike + shell + rider 193 lb.
front end + rider 110 lb. 110/193 = 57%
rear end + rider 83 lb. 83/193 = 43%

This shows that with a rider, the weight is more evenly distributed (57%/43%).

Next I compared the relative weights of the bike + shell:

bike + shell 33 lb.
bike 26 lb. 26/33 = 79%
shell 7 lb. 7/33 = 21%

The figures change when adding the rider’s weight to the total:

bike + shell + rider 193 lb.
rider 160 lb. 160/193 = 82.9%
bike + shell 33 lb. 33/193 = 17.1%
bike 26 lb. 26/193 = 13.5%
shell 7 lb. 7/193 = 03.6%

This shows how small (3.6%) a portion of the total bike + shell + rider weight is attributable to the shell.

Next I compared speeds at similar power outputs (~0.1 hp):

average speed, no shell 14 mph
average speed, with shell 20 mph
increase in speed with shell 20/14 = 1.43 = 43% increase

So, for a 3.6% increase in total rider + bike + shell weight, the speed improvement is 43% on flat ground for the same power output.

I will be collecting on-road data as the summer progresses. Stay tuned!

John Tetz
Succasunna, NJ
May 2003

Next section: Appendix: Welding Zote Foam