Project notes for a hovercraft/moving floor design:

Intro

It is common knowledge that almost any household appliance can perform seemingly amazing feats when properly harnessed. Who here doesn’t remember the pictures in our elementary science books of a car lift consisting of nothing more than an Electrolux ‘sausage’ vacuum and some trash bags? It’s eye opening at first, but mathematically speaking, really not hard to conceive at all. The principal is very similar to common torque multiplication; or in simpler terms, spreading out the load. So when I found myself in need of a moving floor, that very text book came to mind (Heath Science, 3rd grade, for anyone who didn’t attend Tilden Hogge in the 80’s). Grab some bags and a vacuum, plug it in and go, right? Wrong. Right off the bat I ran into some issues:

1. How to reset? The floor needs to settle back to the original position quickly. The only escape for air in a fully sealed system as in the basic lift experiment is back through the vacuum itself. Most likely the vacuum’s blower would act as a kind of check value and make the return very slow, if at all.
2. Exactly how were the bags connected in the first place? A single trash bag certainly wouldn’t do the job, and I simply could not fathom how multiple bags were interconnected.
3. Multiple use. To work as I need it, the floor lift will have to be reliable; not a single function demonstrator. Again, daisy chained trash bags would be inadequate.
4. Pressure relief. In addition to self-resetting, the floor must also be self-regulating. During work phase, the vacuum will be unattended and continue to supply air that must be bled off. Once again trash bags are not adequate; the vacuum blower will eventually fill them to bursting. Finding a bladder strong enough to withstand and ultimately block further inflow from the vacuum is also untenable. Most household vacuums use a portion of their blower’s airstream for cooling. If airflow is blocked by a full bladder apparatus, the vacuum motor will quickly overheat and fail.

As far as I know, the “car lifter” experiment deals with none of those issues, and if it does there is (or rather was) certainly no explanation. Google to the rescue!! Except, not so much. I couldn’t find anything on the specific experiment, and while there were some designs here and there, none of them suit my needs. In the end, I was left to come up with a design of my own. The biggest problem was still a check valve system. After some thought, I more or less landed on an epiphany; why not allow the air to escape from a series of holes in the bladder, using an O ring to center them in a dome shape? As the dome inflated, the holes would be sealed against the ground, giving maximum lifting strength. Afterward the whole unit would sit on a cushion of air, allowing (and requiring) a constant inflow to maintain pressure balance. It would be fully self-checking and self-resetting, just like I needed. It was only afterward I realized the design is literally a hovercraft. Duh! As an extra bonus, I had never even noticed my old shop-vac converts to a blower, not only giving me slightly more power than a common house vacuum, but also less worry about cooling, since the shop-vac’s motor has a separate cooling fan (though the hovercraft design mostly resolves cooling anyway). All that was left was to test the concept, get a design, gather the parts together and start building.

2011_09_24 – Test I, Concept

The first model is intended to and will only lift; it has no means of locomotive propulsion. It will serve as a prototype for a self propelled iteration in the future. Absolute lifting capacity is a given; what I am worried about is speed. The idea is to lift a human adult about 12cm within a few seconds of start-up. For this experiment I used the ballast bladders from my Kayak, a simple piece of wood, my own weight, and my hands as a check valve. It worked beautifully. Even with the tiny volume of air that could get through the bladder’s filler hose and nothing but my hands acting as a seal, the bag filled to capacity under 10 seconds, lifting me well off the ground in the process.

2011_09_27 – Parts

Plans Drawn, parts on the way. I should be able to get started in the next week or so. I guess it needs a name. I’ll call it the DCHC Model 1. 🙂

2011_10_18 – Model I, No Dice

DCHC Model 1 constructed, tested and… absolute failure. The skirt attachment (staples + tape) simply cannot withstand enough air pressure to lift significant weight. The center spire underneath also failed. Not such a big thing and easily fixed, but the skirt failure is an obvious deal breaker.

2011_10_23 – Model II, Final Test

Improved design with following modifications:

• Plastic lid stapled to bottom with pencil holes replaced by cut tin disks mounted with eight screws each, ringed by six large air vents.
• Skirt attached and sealed with combination of staples, hot glue and taping.
• Addition of a 6cm plywood sheet mounted atop the primary deck board. This is to create a kind of sandwich effect, forming an impregnable seal to hold the skirt in place. It also has unplanned benefits of protecting the skirt assembly from tears while walking on the craft, freeing up deck space and improving the aesthetics a bit.

Testing of this second design (DCHC Model 2) proved far more successful. A minor flaw is a bit too much air escaping toward the back of the craft compared to other areas. I apparently mis-measured and cut the skirt a bit short in that area. Still, it works as intended. The craft itself weighs around 18kg and will easily lift my 90kg body weight. I’d estimate the maximum total operating capacity is around 135kg.

More to come…

DC