I put everything back together with help from Michael today. I think we're almost ready for another field test!
We added one enhancement for the lift fan. A little bit of rubber padding between the engine mount and the safety grill. I hope that this will avoid any rattling between the two. And it should also take out some of the vibrations.
Rubber padding installed
Close up of the rubber padding during the installation process
Our simple design is reusing some used RC car tires. They're cut and splayed open. And then zip tied to the engine mount.
One thing we noticed and were surprised about was the amount of rust that's accumulated on the engine mount. There's nothing structural, but it was more than we were expecting since the craft has always been in the garage.
One other thing that we had to do was to bang out the small indent where the lift fan cowling was slightly dented from its past life with wheels mounted there.
We also tested again with the sound meter. Even with the padding the lift fan is still at similar levels. Mid to high 90s db at idle and almost 105db throttled up.
Here's me trying out the new mount with the fan and exploring the envelope of CG which is now much larger.
We tested out the thrust fan too. Again the engine noise is basically all the noise and it ranges from the mid 90s up to just over 110db at full throttle. It also appears to be the majority of the noise coming from the engine. Of course the
With a trip to the store to get a handle bar and reusing some of the pieces from the oarlocks I've successfully mounted the thrust engine twist throttle at the front of the cockpit.
The twist throttle mounted
It took a while to find a part at the bikeshop with a long enough straight section. But cruisers have some long segments in their handlebars ...
The handlebars I repurposed. (note I used the longest straight section below the first bend.)
And here's a quick test:
I also thought about mounting it on the tiller too. I didn't have enough slack in the control cable at the moment. But I think that the current mounting will provide a nice solid handle for the pilot to lean against.
The engine needs to warm up a bit before it can throttle up effectively. You can see some lag, it was way better this time than the first time I tested it completely cold. Opening up the throttle completely flooded the engine.
With the new propeller I also purchased a new throttle linkage. So I needed to find the best way to hook it up. So to get the best look at the possible throttle mechanisms I ended up tearing down the engine. Now you can see the governor and throttle linkages clearly.
The twist throttle linkage
Top view of the engine with gas tank and air filter removed
The air filters for reference
The engine with the gas tank and air filter removed, and screws on the ground
The end results of my research was to identify that the best place to attach the throttle is actually on the exterior handle.
My final solution
The bare end
A little bit of electrical tape for grip
Three zip ties to hold axially
Another view of the zip ties, also showing the zip tie with the screw attachment point.
After using pliers to tighten the zip ties
The zip ties trimmed
The zip tie mount installed on the front of the engine
Demonstrating the mechanism
With the friction released the external handle does not hold position. And removing my other throttle mechanism we're now able to control the throttle with a twist of the wrist.
It's a little bit more sensitive than I'd like, but being able to change throttle quickly will be nice. And I suspect that most control will be bang bang style anyway. Right now there's not much spring return, but I believe that once the governor spins up the return spring will be tighter and pull it back. Since at stop the governor is leaving the throttle wide open the throttle linkage only tensions at the very end.
Next up I need to get a handle bar to mount the throttle onto. I'm not sure if it should be on the tiller stick or not. A single joystick control would be nice, but possibly confusing/ergonomically challenging if it requires continuous torque.
Today was exciting. I swapped out the lift fan from the Universal Hovercraft kit with a Hascon Wing Fan cut to size 24.5" that I ordered from Slipstream Hover. The size was picked to give 1/4" of clearance on on the 25" interior space I measured and talked with the representative about optimal size to select.
Unpacking the new lift fan
The new fan is a 5 bladed fan. With a cast hub with mounting points for each blade, and a machined collet flange. Each blade is molded plastic cut to length.
I also have a new twist throttle control from them to try out.
The two propellers. Old foreground new background.
When I removed the old fan, it just slid off once I released the shaft end bolt. Clearly the collet was not clamping correctly.
The new lift fan fit check on the engine shaft
Installing the fan was easy. The collet was nicely layed out that I could easily put the bolts in from the bottom. So I could use a ratcheting socket and tighten it easily and evenly. The shaft is a little worn these days and it tightened further than optimal, but the collet has a full cut instead of just the lower part of the flange which gives it more travel. And after testing the collet appears to be holding successfully. I put the end nut onto the shaft as well which will keep the propeller from sliding off the end of the shaft, but it does not have a plate that prevents the fan from riding up the shaft.
The hub is much more compact and the overall blade is a little bit lighter. But the most important thing is that the blade is noteably more balanced. Looking down when spinning the old fan by hand, the plywood circle could be seen oscillating with at least a 1/4" asymmetry. Which meant that it vibrated the whole system quite a bit.
Here's my first test of the new fan. I spun it up to idle speeds briefly and then inspected it before trying to throttle it up.
A few interesting notes are that with the new fan, it actually was effectively hovering even at idle.
Here's me trying out hovering and moving my weight around on top of the hovercraft with the new fan throttled up.
For reference I filmed a similar segment earlier with the old fan. You can see more vibrations, as well as notice that the vehicle does not lift off even at idle. I was previously worried about the vibrations and was considering more reinforcements for the lift engine mount but I don't think that's necessary anymore.
Noise level:
I thought that the new fan might be quieter, but apparently not. I think that most of the noise is actually from the engine. I tested both the new and old fan, and idle and full throttle. And I found that it was about the same.
95 db at idle lift only UH fan
105 db at full hover throttle, lift only UH fan.
Here's my testing again with the Hascon fan. My phone was giving me trouble turning on the camera. But the readings were not significantly different than the priors so I didn't bother figuring out why the camera wasn't starting properly. You can see that these measurements were taken between 1 and 2 meters away from the noise.
So although I probably don't need to reinforce the engine mount. A housing to contain the engine noise and protect the fan, as well as provide ram air pressure.
One other note is that the new hub is notably smaller which means that the engine fan is now actually about 2" higher. I was previously considering making the baffle take up more of the lift fan area, however despite the larger gap between the diverter and the fan blades, we're still getting much better performance. Until it's the limiting factor I'll probably drop the priority of refactoring that for now.
One new piece of technology I've deployed now is some tamper marker paint to see if things are vibrating loose.
Tamper paint as first applied
Tamper paint after being run before fully dried
Above you can see that I didn't quite wait long enough for it to dry. But it's showing that the nuts are not moving.
Additional dot on the shaft to detect upward movement.
As mentioned above the end shaft bolt only keeps the hub from coming off the end of the shaft, so I added this dot of marker paint to show if the hub rides up the shaft at any point.
The hovercraft is surprisingly dirty after just a few minutes flying around a "clear" parking lot.
Dirt collected under the thrust engine
More collected on the side
Another view of the dirt at the rear
There's even dirt in the center of the fan. Clearly there's a lack of airflow there...
After cleaning off as much dirt as possible I set to work to improve the hover performance.
I took two approaches. First I started by clearing out more of the grill wires. It made a large difference for the thrust fan. And since the lift fan is actually well protected, I cleared out 3 of 4 on the bottom, and 2 of 3 on the top except for the outer ring where I did 3 in a row where I expect the highest speed.
In addition there was a gap between my metal diverter plate and the board that is attached to the skirt attachment leaving about a 3/4" gap across the whole width. And with the wires removed I could now relatively stretch some Duct tape across the gap. I think this might be my first correct use of Duct tape ever!!
Sealing the diverter with duct tape front view
Duct tape rear view
And reassembled
The propeller is much mor eexposed. i plan to build up a higher housing around the lift engine to keep me and other things farther away from the lift engine to avoid any problems with foreign object entries. I'm even thinking of a shroud/hood that will scoop forward air and also dampen the noise hopefully.
As I identified in the last post there's quite a bit of pressure buildup in the thrust fan. So I've been reducing the drag induced by the protective grill.
The 1cm width is narrower than necessary. So with help from Wim, I've removed every other wire from the front. And 2/3 from the rear of the thrust fan housing.
With these gaps I still cannot reach the propeller even if I try with my hands since it has 3+" of clearance at the front and 6+" inches of clearance at the back.
View from the front of the reduced lift fan grills
Back left view
Back right view
In this process we removed a lot of clippings. The ones I collected weighed 2.4 lbs. Not bad for decreasing the weight at the back.
Most of the clippings removed
Clippings bundled in cling wrap to be weighed. (2.4 lbs)
With so very unscientific testing of standing downwind of the fan, it felt noteably stronger. And throttling up and down did not result in the front grill flexing like it used to do.
We held back on the lift fan since it's both closer to the pilot as well as lower volume due to having back pressure. With the improved fan mounting the craft is floating much higher, but the CG is still notably aft. To take a passenger I'll probably need to add weight at the front with the current configuration.
Since I was having trouble with the lift fan propeller and I've tested it more than the thrust fan I decided to do the same operation for the thrust fan just in case.
Dissassembling it there were no major signs of wear from spinning like the lift fan which had shavings everwhere and clear galling and grooves. There is a little bit of wear.
The thrust fan shaft after
The collet freshly removed
Since I used the end cap in the middle of the lift engine hub I needed to make another one.
Here's the end result, not pretty, but effective enough.
The hub assembled with longer bolts extra washers and locknuts.
The final assembly from slightly behind.
The hub before assembly with the keys installed.
The keys and bolt in a fit check.
And here's the evidence that the fan stayed on. For at least 5 minutes....
While running this test I was watching closely for wear and tear and any malfunction. And I noticed that the grill protecting the fanwas flexing quite a bit when I throttled up. So I setup the camera in a different vantage point. Sorry it's a little bouncy since the tripod is now on the hovercraft. But you can see the grill flex when I throttle up best by watching how much of the engine shaft you can see.
TODO lift
I also did some continued testing of the lift fan. I found two more points of failure. Since i've dissassembled and reassembled the lift fan so many times, two of the screws holding the grill down wiggled out of place. it was kind of exciting to watch one of them start bouncing across the grill and then disappear between the wires never to be seen again. One more reason I'm glad that I have the metal housing. And there's the thick piece of wood between me and the lift fan too.
I filmed the lift engine tests but unfortunately the gopro had operator error and so I have two unexciting pictures for each test. One before the test, and one after the test. During this test I was excited to realized that the tilting of the hovercraft does effectively allow me to move it side to side by releasing air out from the side of the skirt. By leaning I could choose to loosen or tension the docking lines on either side of the vehicle.
The other interesting result of the lift engine test was the below grill failure. As far as I can tell this wire was vibrated free from its housing. It does not appear to have had a traumatic experience from cutting and just sprung itself loose from a weakness in the weld under the high vibration environment of the lift engine.
The one piece of wire that has proven to be not well attached. You can also see the hold where the sheet metal screw has self vacated.
The whole loose strand, and both screw holes which were vacated.
I am going to be looking for ways to decrease the vibrations of the lift engine. Mounting it on the two horizontal bars is generally strong enough but leads to a lot of resonances which I think are making everything harder to keep in place. And also makes things louder and less comfortable on the craft.
