With the addition of the NACA duct and fans, I seem to be getting enough air flowing through the cockpit that I can leave my Lexan visor installed during a ride without overheating. Up until recently, I still left my side windows open so there was an easy path for air to exit the cockpit. Then I decided to install some Lexan lenses over the side windows to improve the aerodynamics (at least in my head). This seemed to choke off the flow of air coming from the fans and the NACA duct.
Then I remembered the small exit vent that I constructed for the aero hood that I made for Peter B’s DF. That hood had no side windows so it needed to provide a way for air to exit the cockpit. Here is that vent under construction…
Fortunately, I made a small mold from that vent so that I could reproduce it easily. So I thought it was worth a shot to try installing a copy of that vent on my own homegrown hood. Here, you can see the progression of grafting the vent into the hood.
It seems to noticeably increase the air flow through the cockpit. However, without proper testing, I don’t know if it has any negative effect on the aerodynamics. My list of features to test continues to grow.
One of the features of my Milan GT that I really liked was the LED lights in the mirror covers. These acted individually as turn indicators and together as emergency flashers. I’ve been meaning to get around to adding LEDs to my Snoek-inspired mirror covers. There were a couple of small electrical issues to overcome but I think I’ve got them figured out.
Without going into a lot of detail, I came up with these 12V LEDs from Amazon that worked out well. I had to add a little structure to the interior of the cover to mount them. I 3D printed these covers using clear PETG, then painted the blue on all but the point. I also had to deal with the fact that the DF electrical system is based on a 7V battery. These LEDs require 12V. So I had to do some funny business to bring in 12V for these LEDs and the ones in the hood side markers.
I used hollow Sturmey Archer brake cable adjusting screws to mount the mirrors. This allowed me to pass the wiring to the LEDs through the screw without having to drill any more holes in the body.
This is not my original idea. I came across this somewhere on the German Velomobile forum. It sounded easy enough to do so I thought I’d give it a go. The idea is to add another layer of Lexan to the interior side of the visor, leaving an air gap between it and the visor. This layer (or insert) becomes the interior surface. The air gap prevents the cold temperatures from transferring to the insert so condensation (fog) is less likely to form. This is the same idea as employed by the Pinlock system used on motorcycle helmet visors.
Plexus Cleaner – (Optional) This is a great cleaner for plastics but pretty pricey.
Design a pattern for the antifog insert that’s smaller than the visor opening in the hood.
Trace the pattern on to the Lexan sheet’s protective film.
Using some large scissors, cut out the pattern.
Remove the protective film from one side of the insert. This will be the side facing the factory visor.
Apply the double sided tape to this exposed side of the insert near the edge. Leave the red backing film on the double sided tape. Use a razor to remove any excess tape that goes beyond the edge of the insert. Be careful not to introduce any smudges or fingerprints to the tape or to the exposed side of the insert.
Antifog insert with double sided tape with red protective film still in place.
Remove the visor from the hood and clean it with something like Plexus.
Remove the red backing film from the tape. Position the insert carefully over the inside surface of the factory visor without letting the tape touch the visor yet.
Once you’re happy with the position of the insert, start at the centers of the upper and lower edges of the insert to press the taped edges to the visor, working towards the sides of the insert. You want the insert to follow the curve of the visor leaving an air gap between the insert and visor.
Insert installed with protective film removed.
Remove the remaining protective film from the non-taped side of the insert.
Reinstall the visor on the hood. That’s it.
The antifog insert has worked surprisingly well during a couple of early morning rides.
I’ve been playing with various configurations of NACA ducts, extension ducts and electric fans on the DF to add a bit of airflow through the cockpit. The more that I tinkered, the more complicated the cockpit got. So I decided to try to organize the cooling and sort of wrap it all up in a dashboard. The idea was to provide a single piece that surrounded the vent from the NACA duct, incorporated the fans and organized the fan and switching wiring. Here’s what I came up with…
Below you see the plug on the left, made from 1″ pink foam sheet, bondo and paint. The resulting mold is on the right. I should have made a 2-part mold. Extracting the plug and then the part from the mold was very difficult.
The resulting part is on the left. On the right is the backside of the part with the fans and switches installed
This shows how the ducting and dashboard pieces fit together.
The dashboard installed:
A minor update… I’ve added some extensions to the fan outlets. These turn the air inward a bit directed more towards my face. I’ve also added an extension to the NACA duct outlet to carry the air further into the cockpit. I’ve also added a butterfly flap in that extension that allows the air to be directed up or down. All of these extension (white) are 3D printed with TPU filament. This is a very flexible plastic that won’t cause any harm in the event of a collision.
This modification falls under the category of “that guy must have too much time on his hands”. I’ve admired how the Alpha 7 foregoes the use of screws to mount the front access panel and the rear derailleur panel. The DF has unsightly M4 screws holding both panels in place. I’ve already converted the mounting of my DF’s front access panel from screws to magnets. However, my rear derailleur panel was still held on with M4 screws. OK. This is really a pretty trivial thing. I know. But, with the encouragement of my friend, Doug, I decided to convert my panel to mount securely with magnets. This modification had to be reversible so no permanent changes would be made to the DF.
I first made a mold of my existing rear panel so that I could make my own copies and keep my factory panel intact. I laid up my first part with a layer of gelcoat and 2 plys of carbon fiber twill. This part came out slightly thicker than stock by still usable.
Next, I had to come up with a way to mount the magnets on the DF body in a non-permanent way. I ended up doing something similar to my approach with the front panel. I stretched some thin flexible abs plastic sheet (the orange pieces in the photo below) across the screw holes and taped some magnets on the backside of the plastic. I used some M4 button head screws and nuts in the stock holes to mount the strips. The magnets were held on with gorilla tape. The magnets on the panel were glued in place with E6000 glue.
The panel mounts with a loud “thunk” and seems very secure. I’ll wait until I’ve had a few successful rough rides before I declare victory but I’m pretty confident that the panel will stay in place.
The Snoek is a new velomobile from Velomobile.nl. It is extremely small, light and supposedly fast. One of the details that caught my eye is its aerodynamic mirror cover. It is sort of a half cone that extends from a half round mirror. I believe that the mirror is the ubiquitous B&M Cyclestar mirror cut in half. This is the Snoek mirror…
I decided to come up with a similar mirror setup for my DF. It’s important to me to make few if any permanent changes to the DF. I always want the ability to revert back to stock. So whatever mirrors I came up with, they had to mount in the stock mirror holes.
I started with 2 B&M mirrors that I pilfered from one of my trikes. I cut them in half using a diamond cutoff wheel in my Dremel tool. I needed a way to mount the mirrors and the covers in just one hole per side. Below are models of the mirror mount (left) and the mirror covers.
To attach the mirror to the mirror mount, I cut off a piece of the stock B&M mirror’s ball/stem and cut M6 threads into its outside diameter. M5 screws are used to attach the mirror cover to the mount and to attach the mount to the body.
Here are the mirror covers that I came up with. I 3d printed them, sanded and painted them. I was concerned that they may be too far inboard to be effective. However, in my first test ride, I found that the visibility rearward was fine.
I’ve since enhanced these covers on the DF to include lighted turn indicators.
Update: I’ve since added these mirrors to my Milan SL. I found a flat area on the sides of knee humps where they fit without modification. Visibility is good.
After noticing how much more air is passed by the Milan’s small NACA duct than the DF’s, I took a long look at the larger duct on the DF. One of the main differences was the depth of the ducts. The floor of DF’s duct seemed to rise so that there was a section of the duct that was very shallow. My guess is that this reduced area was causing some turbulence which reduced the air flow. I came up with this deeper duct. I also added an extension to the duct to feed the air closer to the rider. The air flow was improved greatly by these pieces.
I showed these changes to my friend Doug (the gentleman who sold me the Milan SL). He thought that it would be nice to make the duct closable with a flush flap of some sort. There are 3D printed NACA ducts sold in Germany for the Milan that have a closable flap. However, these NACA ducts won’t work directly on the DF due to the flat shape of the Milan’s top panel vs the curvy, recessed shape of the DF’s access panel.
I took Doug’s suggestion as a challenge and this is what I came up with. I first laid up a new NACA duct and extension duct as shown above. Then I laid up 2 new DF access panels – one to house the NACA duct shown above. The other panel was used as the donor for the flap. The flap needed to have the same curvature as the duct panel in order to perfectly fill the opening when closed. Extensions were added to the tip and trailing edge of the flap to allow it to lie flush with the top of the duct panel. The tip extension of the flap inserts into a loose slot in the front of the NACA duct to form a sort of hinge. All of the pieces in their rough, unpainted form are shown below.
To operate the flap, I needed a mechanism to push it up flush with the panel and down to touch the floor of the duct. The solution that I came up with was a simple stick screwed to a carbon fiber clevis (not shown) on the bottom of the flap and protruding through the floor of the duct. I 3D printed a piece to attach to the bottom of the duct to provide a rigid slot for the stick to ride in. The stick could easily be reached by the rider. I used a small piece of 5mm bungee attached to the stick to force it and the flap up and back. To lower the flap, the stick just needed to be pulled down. The friction caused by the bungee pulling the stick back against the 3D printed slot holds the flap in the down position. If needed, I can grind several notches in the stick to hold it more firmly in some in between positions. Below you can see the stick exiting the plastic slot attachment and the model of that slot attachment.
So how did it turn out? It works surprisingly well. I replaced the wooden stick with a nicer carbon fiber piece. I added a couple of very thin magnets to the bottom side of the panel and to the rear flange of the flap to locate the flap more precisely when being closed. Was it worth the trouble and complexity? Maybe. For the most part, I just ride with an open duct. So it would be a rare occasion that I’d want to be able to open and close the flap. That said… it was a fun exercise.
Here are some photos of the “finished” duct panel.
The second version of my Milan hood was pretty successful. However, it still had a small problem that I should have addressed in an earlier version. Since I barely fit in the Milan SL, my knees can touch the inside surface of the knee bumps from time to time. Over a long ride, this can get a bit irritating. So I decided to make one more version of my hood with taller knee bumps.
Rather than build a new plug, I decided to modify the previous (V2) plug by adding more volume to its knee bumps. To do this, I laid up some knee bumps from the V2 mold and grafted them on to the V2 plug. Here is the progression of cutting and shaping the bumps.
Here you see the V3 plug, set up with flanges, before making the new mold. This time around I decided to incorporate the pesky leading edge lip into the main mold rather than as a separate mold and part as I had done on the previous 2 molds.
It’s difficult to notice the enhanced knee humps visually on the V3 hood. After all the work, I gained about 3/4″ of extra clearance for my knees.
After completing the new aero hood for the Milan SL, I began taking a more critical look at the way it fits. You’d think that I would have been more tuned into the fit as I was making the plug, but it didn’t really strike me until I was photographing the finished part installed on the Milan.
There is a large gap between the trailing edge of the hood and the head fairing on the main body. This gap is there in the stock hood that mine is based on. But, it just didn’t look right to me.
I thought about modifying the actual hood to close that gap. But that would leave me with a one-off hood that couldn’t easily be reproduced. So I decided to make another plug, mold and part to close the gap.
To do this, I made a fiberglass copy of my aero hood from my mold. Since this would end up being my new plug, it made sense to use the much cheaper fiberglass materials rather than carbon fiber. I then cut an approximately 180 deg slit around the base of the visor area. I then taped this sliced hood onto the Milan with the trailing edge taped down to the head fairing and the leading edge to its normal location on the front area of the body. This opened that slit along the base of the visor area. By filling that gap with layers of fiberglass and body filler, I was able to produce the plug, mold and hood with the proper gap along the trailing edge.
Compare the gap on the trailing edge of the old hood to that of the new hood shown below…
During a short shakedown run, I noticed that the airflow from the front NACA duct was reduced from what was produced by the NACA duct on the previous hood. The NACA ducts are identical and positioned in the same place on the hood. My hunch is that the smaller gap at the trailing edge of the hood is restricting air exiting the cockpit. I may try to improve the air flow out of the cockpit by adding an exit duct to the low pressure area of the top of the hood similar to the duct on my aero hood for the DF.