Winter Trike Construction
Final editing performed by R. M. Horwitz
After spending 27 years living in California (LA and SF Bay areas), riding virtually year round, I moved to Green Bay Wisconsin to accept a job offer with a nutritional supplement company. All went well in my daily bike riding regimen when I arrived in August as Wisconsin Summer and Fall weather is fine and country roads are very safe. Motorists on country roads where I mostly ride are courteous, giving riders virtually the entire lane in which they are riding when passing. Comfortable riding came to an abrupt halt when the weather changed from a pleasant riding temperatures to sub-freezing virtually overnight. The first year I was in Wisconsin, it snowed in November and this snow with more added to it along with sub-freezing temperatures lasted until the end of March. So this 5 months of cold weather and the discomfort of riding in it necessitated rethinking how to get back to uninterrupted Winter riding in the colder climate.
I tried dressing warmer with layers of clothing, more socks, gloves, face masks, neck warmers, and even heated clothing. All this helped but it was unacceptable compared to the relatively lighter weight and unencumbered clothing needed to ride in the milder temperature climates. A better alternative had to be found since foregoing riding through the Winter months and losing all semblance of physical fitness from 27 years of uninterrupted riding was not an option. Furthermore, riding stationary equipment indoors has never been able to maintain my interest as it is excruciatingly boring and lacks the demands of road riding.
I began to research buying a bike or trike which was enclosed, i.e., built like a mini-car with an exterior fairing which would shield against the wind and weather. To this end I found numerous alternatives I would consider, all built in Europe. Ones considered were fully enclosed such as the Go-One and the Aerorider. These machines looked like they would have a good chance of performing well in relation to comfort, speed and handling potential. I used to do road racing, have ridden over 150,000 miles mostly on high performance bikes that I either bought or built myself. Unfortunately, commercial enclosed machines ranged in price well outside what I was willing to pay, the better ones at the time existed only as prototypes. So I began to research what it would take to build a fairing that met my needs.
Fast forwarding through the details of options I considered, I settled upon building a fairing which would fit onto a trike having the “tadpole” design with two steering front wheels and an in-line rear drive wheel. Reasons for choosing the trike configuration were mainly two: 1)the need to be completely enclosed for maximum cold protection with no way to put a foot down for balance at stops without an opening or “bomb bay” doors in the bottom; and 2) the need for the machine to be stable on potentially icy surfaces in the Winter. Other design features I decided upon included the following:
· Hell-Bent Bicycle front steering knuckle with Ackerman compensation
· Underseat steering using Hell-Bent pushrods
· Diatech stub axles with 20” Vueltra Aero rims for the front wheels. Tires used were ERTO 406 20” x 1.5”
· Standard 25 x 700C rim for the rear wheel
· British made Hope hydraulic disc brakes in front, J-brake on the rear
· Self-designed suspension rear-end
· Use of an intermediate drive sprocket (two chains are used to provide the drive with one running from the pedal chain ring to an intermediate sprocket, the other from the intermediate sprocket to the rear derailleur.
· Sachs rear derailleur and shifters
· Fabricated hydraulic brake fluid “splitter” used to evenly distribute fluid to each wheel cylinder. This permitted activation of both front brakes with only one caliper
· Sachs dual drive internal rear hub with integral three speed gearing along with 8 speed rear sprocket. This eliminated the need for a front shifter, multiple chain rings and the greater chain management complexity that would have required.
· Easy Racer recumbent seat.
While I likely could have found a suitable commercial trike that would work, I wanted to build the frame for the machine myself because: 1)the machine had to be built free-standing and rideable but easily accessible for a 230 lb. rider (me!); 2)it had to be compact enough so it could be fitted into the yet to be built cockpit fairing and 3)I could. The choice to build the frame out of steel was made since I had previously taught myself to braze and had an inexpensive but functional oxy/acetylene set-up. I had a number of old steel bike frames from racing days that were bent from crashes plus components which would serve as parts for the new frame. In order to decide the frame design elements that would be required, I chose to make scale drawings of the trike mechanics as well as the exterior fairing. My drawings showed I could use a single 2” steel tube (0.065” thick from Aircraft Spruce), bent into the designed shape (Figure 1).
Figure 1. Side view showing central frame tube
After bending, the central tube was then saddle notched on the underside and fitted with a 1 ½” tubular steel axle. This combination was envisioned to be of a configuration that would permit insertion through the hatch opening in the top of the yet to be built exterior cockpit fairing. The 2” steel tube was bent to the two designed angles of 151 deg. (front) and 110 deg. (rear) at a local muffler shop for a cost of $10. It was cut to length and notched then the axle was brazed while holding both members in a simple wooden jig mounted on ¾” particle board. Next an English thread steel bottom bracket shell was brazed into a hole cut in a predetermined position at the front end of the 2” frame tube. In turn, a second bottom bracket shell with short section of seat tube still attached from a salvaged frame was brazed into a hole cut into the central 2” tube on the bottom side of the rear bend. A short section of another tube was notched, fitted and brazed into position behind the first attached tube thereby forming a triangle with the 2” central frame tube. These details can be seen in Figure 1.
It should be noted that frame design, layout of components, heights above ground for pedals are factors that need to take into account their full, unrestricted functioning inside the exterior cockpit fairing. So for example, a machine not intended for use with a fairing can be designed much closer to the ground than one operating with a fairing. This is because the fairing itself needs to have adequate ground clearance and the pedal/foot combination needs to clear fairing interior without rubbing or changing the normal foot angle that is comfortable to the rider while pedaling.
Though no different from any good frame, it was important that all brazing was done while paying close attention to alignment as the performance of the machine will totally depend on maintaining this alignment. It isn’t hard to do, it just takes paying attention and checking measurements and alignments. The outside edge of each bottom bracket shell must be even and perfectly parallel to the centerline of the frame. Similarly, the axle must be at a perfect right angle to the frame centerline and spaced in the middle of its length. The axle member was cut so the overall wheel to wheel width when completed was 36”. This is a bit on the wide side but was chosen since the external cockpit fairing needed to be fitted and the machine had to be stable in potentially slippery road conditions.
The final major frame feature fabricated and installed was the steering column. Sitting in the mounted seat and experimenting with different configurations showed an underseat steering configuration would work so that was what was designed and constructed. This merely required drilling a hole in the 2” central tube frame member behind the cross axle and brazing into place a standard steel head tube cut to proper length. A steel steering column from a salvaged bent fork was cut to proper length and brazed to a fabricated steering arm made from ¼” plate steel (See Figure 2).
Fig. 2. Steering column and push rod set-up
Mounted with salvaged cups and bearings, standard handlebar stem and aluminum mountain bike steering bar cut to proper length completed the steering system I used. Mountain bike bar ends were mounted on the steering bar to accommodate brake calipers, one for the two front hydraulic brakes and the other for the mechanical rear J-brake caliper. If salvaged bike parts are not available most if not all these parts can be purchased from a builder’s supplier like Gaerlan (www.gaerlan.com ).
Once this much of the frame was completed, the rest of the main frame construction was fairly easy. Further work on the frame included cutting the axle ends at an angle permitting 16 degree kingpin inclination then brazing steering knuckle mounting plates onto the ends. These plates were made of ¼” steel and each was drilled such that the steering knuckle could be mounted with a 12 degree caster. Rickey Horwitz describes details in a series of excellent websites Trikes 101 part 1, and Trikes 101 part 2 on how to do this. Similarly, seat mounting brackets were fabricated, frame was notched to accommodate and these were brazed in place. Finally a friction fit spring clamp was fabricated and brazed to the frame in front of the steering column to hold a simple cylindrical aluminum hydraulic fluid “T-splitter” which was lathe fabricated. This splitter performed the function of permitting hydraulic fluid to be evenly distributed from one hand caliber to both front brake cylinders. (Figure 3).
Fig. 3. Hydraulic fluid splitter
Rear wheel frame triangle design and construction
Once the main frame was constructed, next was to construct the rear wheel triangle. The design for this was as an “A-arm” with the pivot at the rear bottom bracket (Figure 1). The pivot point was a bit tricky in that it had to permit a pivoting action but it also had to accommodate the intermediate sprocket. Fortunately this is easier than it seems by taking advantage of the fact that a threaded freewheel has the same thread as a bottom bracket bearing cup. By brazing two right sided bearing cups together while holding them in perfect alignment with a bolt, the combination provides a means for adapting the freewheel to the bottom bracket. One this adapter fitting was made a threaded freewheel with the ratchet dogs removed provided the intermediate drive which would turn either direction. Through use of adapters and a couple of custom made spacers turned from round aluminum stock, two equal cog sprockets were fitted onto the center of the freewheel, spaced sufficiently far apart that two chains could pass each other without interference (Figure 4).
Fig. 4. Bottom view of intermediate drive sprocket and pivot
Construction of the intermediate drive sprocket permitted determination of the length axle needed to function as the pivot. One was chosen which extended beyond the outbound freewheel edge by about 3/16”. This, in turn sets the width that the rear triangle mount needed to be fabricated.
Construction of the rear triangle was the hardest part of building the trike frame since it had to be strong enough not to undergo torsional flexing while only supported by the pivot point. This triangle accommodates use of a standard rear axle, 700 cm rim and rear derailleur. Since there was a desire to avoid the challenge of variable chain tensioning of the front drive train due to shifting onto different size chain rings, it was decided to use a Sachs dual drive internal hub rear axle and derailleur. This hub has integral three-speed gearing and 8 speed sprocket thereby providing 24 speeds.
A fixed chain tensioner (Figure 5) was designed and fabricated to keep the front chain taut and prevent side sway and derailment.
Fig. 5. Chain idler/tensioner (front view)
The 2” idler wheel was lathe turned from Delrin and uses an integral sealed bearing salvaged from a wheel and was mounted on a ½” bolt. This in turn is mounted on a fabricated aluminum plate which saddle clamps onto the 2” central frame just in front of the axle (Fig. 5). Another view of this follows in Figure 15.
The design features used have proven themselves through hundreds of miles of riding and eliminated the need for multiple front chain rings and made chain management much simpler and more dependable.
A further feature of the rear triangle and the actual reason for the need to pivot is the use of a shock absorber. Since this project had as one of its objectives to minimize construction costs, I used a rubber bumper from a ’56 Buick “A”-arm as the shock absorber (Figure 6).
Fig. 6. Rubber bumper shock absorber
Since the rubber bumper has an integral threaded bolt, it was a simple matter to drill a hole and place a mounting plate on the angled vertical rear portion of the 2” central frame tube. The rubber bumper screws into this and abuts a plate integral on the rear frame triangle. At first this shock absorber proved too stiff but that was easily remedied by drilling a 3/8” hole in the rubber partially through from the tip toward the base. This removal of material weakened it sufficiently to provide more give. Though crude, it actually performs as a fine shock absorber with a travel of ½”-3/4” to reduce road bumps and vibration, an important need in a fully faired machine which transmits all road shock into it creating a “wonking” sound inside the enclosure if too severe.
One additional brazed on feature added after the frame was built was the addition of a short length of tube at the front of the central 2” tube. This had as its purpose to permit clamping in place a fabricated bracket to be added later (see Figure 7) which supported the front of the fairing. The clamping is made possible by brazed-on bolt clamps which can be seen in Figure 7.
Fig. 7. Front fairing support bracket with integral headlights
With the exception of cable guides and other minor braze-on features this completed the frame. Once fabrication of all frame parts was completed, all components were assembled in place, location of cable guides and stops was made and these were brazed in place. After making all necessary adjustments, the trike was completely disassembled, brazing was filed and smoothed, frame was sanded, primed, painted and clear coated with lacquer. With no fairing at all it made for a very fun ride, handled and stopped well and had decent speed for a trike. Most people who rode it always came back with an ear to ear smile on their face variously commenting “it was like riding a go-cart”.
Up to this point, all parts of the construction with the exception of a few were fairly standard for anyone who has ever done metal frame construction. Fairing construction is a completely different matter. Researching it on the internet, I found websites which addressed male and female mold construction, crude male mold approaches with sprayed on styrofoam followed by shaping, etc. to create the building form on which the fairing is layed up. I decided to build my fairing using fiberglass cloth laid up with epoxy resin onto a male mold. The resulting fairing could then be cut off the mold and put back together by bonding strips of fiberglass/epoxy inside and out the length of the split. An excellent depiction of details of male mold construction similar to the methods I used can be found on a website (www.wisil.recumbents.com/wisil/barracuda/barrcaudafairings.htm) written by Warren Beauchamp.
I started by taping a sheet of paper to the wall that would serve as a pattern for determining the side view outline of the fairing. By placing the completed trike with me on it in riding position in front of this paper pattern, my wife traced the outline where the fairing had to extend to clear my head and body. A further need to be accommodated was the diameter of the arc created by my feet when the pedals were fully rotated. The spaces between the upper and lower pedaling extremes were able to then be sketched in to create the completed side profile. After this, we similarly created a frontal view pattern which established the outline the fairing had to be set to provide side to side clearance for my arms, elbows, head and shoulders.
I knew I wanted to support the fairing with the frame so it bore no load whatsoever. This is important since the thickness (and hence overall weight) of the fairing had to be kept to a minimum. The fairing would be supported on the frame at the front with a “U”-shaped bracket, near the ends of the axle and at the rear angled vertical section of the central 2” frame tube. It turned out I guessed correctly that these points would work to adequately support the fairing on the trike frame.
Once I had a completed pattern I began gluing together sheets of pink 2” construction grade Styrofoam. It took 10-4’ by 8’ sheets in all to complete the form (Figure 8).
Fig. 8. Glued stacked foam ready for shaping
Foam panels were glued together with 3M Super 77 multi-purpose aerosol glue sprayed on the layers which were held down by weights while drying. Once I had enough layers glued together to extend just beyond the limits of my pattern, I began the tedious job of shaping the form. This involved cutting with hand saw, filing with large, coarse rasp/files, sanding, cutting, more filing, etc. interspersed with countless measurements to ensure symmetry was being maintained. (Figure 9).
Fig. 9. Shaped foam fairing mold
It was necessary to always make certain a centerline on the foam was maintained throughout this shaping process since this is a critical point of reference needed to maintain symmetry.
Once the rough shape of the fairing mold was achieved, it was sanded as smooth as possible then gouges, holes, etc. were filled in with a lightweight sandable epoxy-based filler called Aeropoxy Light. The entire project used nearly a gallon of this filler, most of which was sanded away in creating the smooth surfaces. All epoxy and fiberglass products referred to in this section were obtained from Aircraft Spruce (www.aircraftspruce.com ). Once this was sanded and the building form was fair and smooth it was covered completely with one layer of fiberglass deck cloth impregnated with Aeropoxy epoxy resin. Once this was cured, sanded, low/thin spots re-fiberglassed and/or filled with epoxy filler, it was finish sanded then painted with acrylic flat black paint. Once dried this was then covered with 3 coats of carnauba wax based mold release. The fairing mold was now ready for fiberglass lay-up (Figure 10).
Fig. 10. Finished fairing mold
The fairing was laid up with two layers of 8 oz. Crowfoot weave fiberglass cloth impregnated with epoxy resin. Aircraft Spruce sells a knobby roller which is very effective in application, spreading and smoothing epoxy resin. The curves and nose area were covered with a extra layer as these areas provide much of the lateral and longitudinal strength. Once completely laid up, the fairing was sanded, low spots filled with epoxy filler, then resanded. Though my description of this part of the construction is short, this step took several weeks to complete.
After weeks of tedious work, it was time to cut my completed fairing in half so I could get it off the building form. There was much uncertainty in my mind about the answers to many questions… Will it release? Will it be strong enough? Once off the form will I be able to get it back together into one piece? It turned out, it was much easier than imagined, but there are several tricks I discovered that made it much easier.
The first part involved cutting the “hatch cover” off the form. This part will serve as the hinged canopy to permit getting in and out the fairing. This was planned to come off the building form in one piece requiring no bonding back together. After measuring/marking where I wanted to cut, the cutting was done with a thin abrasive wheel and hand held Dremel motor. The size of the opening not only had to accommodate rider entry/exit, it also had to permit insertion/installation of the completed trike frame. In short, I made the hatch opening as large as I could while still keeping its size manageable since it would need to be opened and closed every time I rode the trike. The hatch cover, once cut and pried outward around its largest perimeter, popped right off the building form without any problem whatsoever (Figure 11).
Fig. 11. Hatch cover separation from building form
Next the remaining lower section of the fairing was cut down the centerline completely around the building form. With a little outward prying of the fairing to loosen it, both sides popped right off the form (Figure 12).
Fig. 12. Side panel separation from building form
the fairing halves
The task of bonding the two halves of the lower fairing back together proved challenging. The difficulty revolved around the fact that the fairing halves are very flimsy at this stage and tended to sag or distort with the least pressure applied. This problem was solved by cutting the building form (yes, hard to make myself do this since it took weeks to construct) into two approximately equal length sections…a nose and a tail section. Having the two fairing halves duct taped together onto the front half of the building form overcame the distortion problems. Ability to remove the front half of the building form by pulling it out of the hatch opening was pre-tested with the fairing halves taped together. With the knowledge this would work, I proceeded to bond the two halves together, front half first followed by rear half using 2” wide strips of fiberglass cloth and epoxy resin (Figure 13).
Fig. 13. Bonding fairing halves back together
Once the two halves were bonded completely around the exterior of the joint, the interior was similarly bonded. The interior required considerable sanding to remove all traces of mold release stuck to it, a step if not done was found to result in delamination of the bonding strip. After sanding of the joint, some minor repairs of the surface with epoxy filler, further sanding, the fairing reassembly as completed.
Considerably enhanced strength resulted from bonding the two fairing halves together. But the fairing walls and the hatch cover still remained too flimsy and needed further stiffening. To achieve this, 1”-2” square lengths of Styrofoam were cut and sanded into triangle-shaped strips. Various lengths were bonded along the weakest most flexible sections of the fairing. These were rounded over and cove shaped where they abutted the fairing then covered with strips of 8 oz. Crowfoot weave fiberglass cloth then impregnated with epoxy. Once cured these stiffening
members resulted in a decidedly rigid fairing without much increased weight. Examples of stiffener members can be seen in Figure 14.
Fig. 14. Inside fairing stiffeners (pink) and frame mounting
Mounting the frame
The next challenge was to fit the frame inside the fairing while simultaneously providing mounting means to support it. As previously indicated, the front of the fairing was to be supported by a “U” shaped strap fabricated and bent to fit the contour of the fairing nose. This was tedious work crawling into and out of the fairing doing the bending and fitting. Once the “U” strap fit perfectly it was bolted with machine screws to a fabricated aluminum adapter which was drilled then bonded with epoxy (JB Weld) onto the end of a 7/8” aluminum tube. This tube was then placed into the tube clamp brazed onto the end of the 2” central frame tube (Figures 7 and 14 show details of this design element).
In order to fit the frame into the fairing, there was need to cut holes large enough to permit passage of the ends of the axles and the rear pivot fairing at the base of the rear of the frame. It was decided to cut fairly large holes to accommodate these members since the plan was to use plastic soccer cones as supporting members to lay up an fiberglass cone which wrapped around the frame just inside the external members of the frame (axles and rear pivot bracket). Figure 15 shows how this was accomplished. Further details on this part of the construction follow.
Fig. 15. Plastic soccer cones for fairing lay-up around axles
Finally, the rear of the fairing was to be supported by bonding it to the rear angled vertical metal frame section. To prepare for this, a section of fiberglass and epoxy tube was fabricated by spiral wrapping fiberglass cloth around a scrap piece of 2” steel tube coated with release compound. After impregnating the fiberglass with epoxy then letting it cure, it was cut longitudinally then pried off the tube. This split fiberglass tube was sanded smooth then fitted over and bonded onto the angled vertical metal frame section using epoxy. Making the fiberglass tube in this manner using a scrap section of 2” steel tube permitted it to be neat, clean, well fitted and epoxy bonded to the frame tube as opposed to the unpredictable outcome if the fiberglass were laid up directly on the frame tube. Figure 16 shows detail of bonding the fairing to the rear of the trike frame.
Fig. 16. Bonding of rear frame tube to fairing
Prior to final bonding and fastening of the frame to the fairing, the holes for the headlights were cut in the fairing, then front turn signal lights were mounted. Turn signals were needed to be street legal since hand signaling would not be possible with a fully enclosed fairing. Wiring was connected to the leads from these lights and brought rearward in a wiring harness running along a side longitudinal stiffener rib. These features were completed prior to final frame mounting since the frame was to be bonded permanently in place and would impede access to the inside front of the fairing.
The frame with front “U”-strap and integral attached headlights (Figure 14) was slipped in place through the hatch opening, axles placed through the fairing holes then the entire assembly was slid rearward. The rear pivot bracket was slipped through its fairing hole and the angled rear vertical frame tube was abutted to the rear of the fairing and temporarily held in place with large C-clamps. A hole was cut in the fairing to accommodate the rubber shock absorbing bumper. The fabricated frame tube clamp holding the “U”-strap at the front of the frame was loosened and the entire assembly was slid forward to mate with the fairing. With the frame supported inside the fairing, the “U”-strap was epoxy bonded and fastened to the fairing with machine screws. The rear angled vertical frame member was epoxy/fiberglass bonded to the fairing as previously described using fiberglass strips layered across the tube. Before this was done a cove surface was created using epoxy filler where the tube met the fairing such that the fiberglass cloth had no unfilled corner gaps to bridge (Figure 16).
The last part of bonding the fairing to the frame, that of building cone shaped members extending from the fairing to the frame seems tricky but it really wasn’t. This was achieved using a split plastic soccer cone cut to conform to the fairing shape then placed over the frame member. A depiction of this can be seen in Figure 15. This was then laid up on the inside with epoxy/fiberglass, cured and the cone was then removed from the outside after finish cutting the outside fairing to match the perimeter of the largest section of the cone. Where this part of the cone met the outside fairing, another layer of fiberglass cloth/epoxy was applied to further reinforce the bonding. Filing and sanding of this structural element completed the attachment of the fairing to the axle ends. Though the cone shaped fairing around the rear pivot bracket was slightly more difficult to do it was done essentially in the same manner as that used for the axle mounts.
Lights and finish wiring
Without going into a lot of detail, the electrical system was completed by attaching customized taillights with integral turn signal lights to the rear of the fairing. The turn signal lights were included by modifying an existing round taillight through internal mounting of a second socket and bulb. A wiring diagram involved a somewhat complicated circuitry which permitted each headlight to be turned on independently while simultaneously activating the taillights, and a separate circuit for front and rear turn signals activated by a standard automobile turn signal relay. Both headlights and turn signals were activated with DPDT toggle switches. Though a motorcycle battery will work just fine, the system was powered by a custom made 12 volt NiCad battery pack inside a tubular PVC housing which is held in place with a nylon strap on a hard foam rubber mat under the seat.
The top hatch was similarly stiffened in the same manner described for the lower fairing. Three window holes were cut into the hatch at the appropriate level and the cut-out piece was used as a pattern to mark and cut the window. The window was fabricated from 1/8” thick smoke colored acrylic plastic. Once rough cut slightly larger than the window opening, each window section was placed on a cookie sheet then heated in a kitchen oven set at 325 deg.F until it just became pliable. It was then removed from the oven and quickly placed on the appropriate area of the building form. With the assistance of a helper each acrylic window section was hand pressed into the exact contour of the fairing. This worked very well to permit exact matching the compound curvature of the fairing.
Once shaped, cut and filed to exactly fit the hatch window cut-out openings, each window section was inserted in its respective opening and sealed with black silicone (Figure 17).
Fig. 17. Hatch with installed windows
The silicone alone proved inadequate to hold the windows in place due to fairing flex so pop rivets and washers were installed spaced every 2” around the perimeter of each window opening. Drilling at the joint between the window and fiberglass fairing permitted the pop rivet/washer assembly to span both members to hold the windows securely in place.
The mating edge between the hatch and the lower fairing was not able to stay aligned without a flange. The design for this was simple but somewhat lengthy to construct…a bonded flange about 24” long made from epoxy/fiberglass which was laid up then cut to size using the building form to provide the shape matching the lip contour of the fairing opening. This flange was epoxy mounted on the inside of a fabricated fiberglass spacer on the lip of the hatch fairing creating a gap between itself and the hatch fairing. A similar flange mounted onto the inside of the lower fairing mated with this hatch flange gap when the hatch was closed. The flange section on the lower fairing was fabricated to run completely around the entire lower fairing opening, especially important to keep the hatch from sagging inward at its front edge. This flange arrangement not only provided positive hatch alignment, it also functioned to keep water out of the fairing if caught in a rainstorm while riding. Figure 18 shows a schematic of the cross-section of the hatch opening flange.
Fig. 18 Schematic of hatch opening flange
NACA Duct and Interior Moisture Control
Air flow through the fairing during extremely low temperatures was found to be critical. This is not only for rider comfort but also to eliminate moisture expired during breathing. With extreme cold conditions, this expired moisture if not exhausted from the cockpit was found to condense on the windshield and fairing interior in general and will form a solidly frozen layer difficult to keep clear on extended rides. To facilitate air flow, a NACA duct (see www.wisil.recumbents.com/wisil/nacaduct/naca-duct.htm for additional design and construction details) was fabricated into the hatch using fiberglass/epoxy construction. The duct shape was achieved by laying up the fiberglass on a shaped piece of styrofoam temporarily glued to the interior of the hatch. Once cured, the exterior opening leading into the duct was cut away, shaped, filed and sanded.
The duct also included a center pivoting butterfly valve at the rear edge which permitted both air volume and directional control. Two 3” holes were cut into the rear of the hatch to function as air exhaust openings. This duct proved adequate to permit keeping the windshield clear as well as keeping the interior controllably cool during riding. It might seem counter-intuitive but air flow through the cockpit fairing even at temperatures as low as –20 deg. F was found to be necessary to avoid getting over-heated on extended (greater than 10 mile) rides. The finished NACA duct can be seen in Figure 19.
Fig. 19. NACA duct/butterfly valve in hatch cover
For longer rides once I start to sweat more profusely this duct alone was not found adequate to exhaust enough moisture during extremely cold days to eliminate all windshield fogging/icing. After experimenting with battery powered forced air exhaust fans, cracking the hatch opening, etc. it was found that avoidance of interior icing could be achieved with more air flow through the cockpit fairing. However this was achieved at the expense of rider comfort. Returning home after a twenty mile ride at –10 deg. F with sweaty hair frozen stiff proved this point. Another approach had to be found. To this end, a commercial paint mask with carbon canisters was obtained at a local hardware store. This mask was modified to permit one side to exhaust outside the fairing through a 1” flexible foam tubing like is used to insulate pipe. An adapter was lathe turned to permit the mask to be affixed to the breathing tube. Though initially somewhat odd to wear while riding, this set-up proved quite satisfactory to control internal moisture/icing and is not at all uncomfortable to wear during riding. Probably because of the novelty of the machine in general, it does seem to accentuate motorist/pedestrian gawking when I pull to a stop sign and open the hatch while still wearing the breathing mask.
A few other details needed to complete the fairing construction included mounting the hatch to the lower fairing using a length of continuous hinge, cutting holes for the front chain to engage the intermediate sprocket, drilling/cutting holes for cables and hydraulic brake lines, steering pushrods, etc.
Hatch hold-downs were fabricated from two sets of slotted wooden blocks mounted on the fairing body and hatch interior next to the joint where they come together when the hatch is closed. Holding the hatch closed was achieved by use of a Velcro strap used to cinch between the two blocks.
Before final assembly the last major task was to paint the fairing. With this project resulting in a machine decidedly “over the top” in relation to its unconventional design, it was decided to give the fairing a radical air-brushed flame paint job. This was done with acrylic primer, lacquer paint and final clear coat. The resulting paint scheme is shown in Figure 20.
Figure 20. Finished fully faired trike
The last step was to reassemble all parts removed prior to fairing fitting, assemble front crank axle and cups, crank arms and pedals, rear frame triangle and wheel, front steering knuckles and wheels/brake assembly, chains, cables, steering and pushrods, etc. Completing the set-up included hydraulic brake system filling and bleeding, adjustment of wheel alignment using the adjustments on the HellBent steering knuckles and steering pushrods.
The most asked question I get is “how do you get in and out of that thing”. That was a question I asked too even before building the fairing. The first thing I discovered was that the trike needed to be held steady to permit entry. Movement from stopped position could spell disaster as the trike rolled during entry attempts. Solving this problem was fairly easy with a low tech fix by placing a short bungee cord loop around the rear brake lever (Figure 2). Entry was then straight-forward by standing on the seat, reaching back and holding the top of the seat back with my hands. Next step is to place the heels of my feet onto the angled frame members attached to the axle (Figure 15) then slide down and in. It turned out these angled frame members not only are necessary for axle reinforcement but are critical for foot placement to permit getting in and out of the cockpit without risking slipping and the disastrous outcome of punching a gaping hole in the bottom of the fairing. Getting out of the cockpit is simply reversed. Doing this is easier and slipping is less likely without shoes on. There are probably other ways to do this but the only serious precaution is to avoid putting weight onto the thin fairing floor or sides.
As might be expected, riding the enclosed trike for the first time was an unusual experience. Notably, the sound is the first obvious difference from an unfaired bike/trike in that frame/road sounds are amplified by the resonance of the cockpit fairing. Even though the Hope hydraulic brakes are very smooth, the friction they generate creates sound that is transmitted through the frame and further amplified by the fairing, making for a pronounced dragging sound. The enclosure at first creates some level of anxiety due to its claustrophobic nature and restricted view at first until it is gotten used to. Cornering takes some getting used to as well since the machine does not lean like a bicycle and body leaning is restricted by the interior width of the fairing. One needs to slow down somewhat around corners to avoid tipping or lifting a wheel, a risky proposition especially while in an enclosed fairing. The acceleration is slower due to the increased mass but the ability to crank the speed up to a comfortable cruising speed of 20 miles per hour on the flat was not difficult. For comparison, this speed is about the same if not slightly slower than my comfortable flat road cruising speed on a two-wheeled unfaired recumbent. While this machine was not designed for speed but for Winter riding comfort, higher speeds are attainable but take extra effort. It is a certainty that the effort required to maintain speeds over 23-25 mph for extended periods of time will assure staying in physical shape throughout the cold Winter months.
The steering is very quick and takes some getting used to. Learning how to more deftly “guide” the steering with a soft touch as opposed to hard gripping the bar ends on the handlebars and pulling/pushing soon becomes natural and makes the steering much smoother once mastered. All in all, it’s a fun ride and very responsive.
The most notable difference with the enclosed trike (aside from industrial grade motorist and pedestrian gawking) is the warmth it affords. This was the major reason for going to all this trouble in the first place so I was pleased to find that objective was achieved. On my first ride I wore sweat shirt and pants but quickly became overheated and had to return home to change into lighter clothing. Now I wear exactly the same clothing worn in the heat of the Summer on an unfaired bike and stay quite comfortable even in sub-zero temperatures. The coldest temperature in which I have ridden so far is minus 15 deg. F. Wind chill experts can calculate what that comes to at 20 mph road speed. Suffice to say it is brutally cold. I remained quite comfortable after about 1 minute of warming up while wearing only shorts and T-shirt in spite of frost building up on the inside of the fairing from freezing vaporized perspiration and respired moisture. I suspect I would remain comfortable at even colder temperatures.
Though it has never happened I am concerned about a mechanical breakdown in extreme cold miles from home which would force me to stop for an extended time for repairs. One would become hypothermic very quickly at sub-zero temperatures covered with perspiration and inadequate protection. I carry a cell phone as a precaution against such an occurrence. There is also plenty of room inside the cockpit fairing for jacket and pants if one wanted to drag around the extra weight. My preference is to keep my machine in good running order so there is little chance for a breakdown. I run kevlar tires with an internal thorn proof strip to minimize chance of flatting.
Flat tire repair of the rear wheel is just like a regular bicycle with its vertical drop-outs. Repair of the stub axle front wheels is actually easier since the entire machine can be wheeled into the grass, tipped on its side and tube changed in place without wheel removal.
While a large project, my objective of being able to ride outside in cold Winter months has been achieved. I ride regularly now throughout the Winter with the only thing stopping me from doing so being an occasional large storm that dumps a load of snow on the roads. Wisconsin public works crews usually clear roads quickly so I have generally been able to get back outside within a few days at most after a large Winter storm. With three wheels the chance of tipping over if hitting a patch of ice is virtually non-existent and can actually be quite exhilarating as long as there is plenty of space in which to do this type maneuver. Frozen lake beds clear of snow are particularly attractive for this though it helps to have an incline like a paved boat ramp to go down to get speed up without the rear drive wheel spinning.
Total out of pocket cost for me to build this machine was in the $1200-$1,300 range. Many parts, particularly metal tubing, were salvaged from wrecked bikes otherwise expense could have gone several hundred dollars more. Much of the expense was in the front wheels with their stub axle hubs, hydraulic brakes and steering knuckles. While not necessarily inexpensive, my total cost was considerably less than one would have to pay for a commercial racing quality machine which can easily run in excess of $4,000.
Finally, I hope my experiences and insights are of some help to others in the more extreme cold climates of the world who wish to continue to enjoy outdoor Winter riding but do so in comfort. This is a big project so start it early in the Spring if you want to have it ready for Winter.