Better hovering

With the improved engine mount we can hover much better. There's much more airflow and the skirts are kept at a noteably higher pressure. Thanks to Adrian and Michael for their help today.



The flight went very well. It hovered low at idle and made it up to full height at a mid level throttle. With the full power I was easily able to balance the craft. And Adrian reported it was very easy to slide the vehicle around.

Here's a slow motion version. It's great to be able to see the air gap under the skirt. Also watch the fun dust cloud as it gets started running.

A little bit more about the process below. There was some basic assembly to get started. And then we tested the engine while still on the blocks.

The new engine mount at the time of the first test

After the first test we needed to disassemble it again so I removed the extra metal now that the engine had been alligned and bolted down in place.

Metal removed

Trimmed engine mount

Collet with locktite before tightening

 When we ran the first test there was a repeat of the fan blade coming off even with the lock tite.

Scrapes observed on the bottom of the engine.

 We found some scrapes on the bottom of the engine where the fan had actually ridden up the shaft due to the air pressure and was rubbing on the bottom of the engine. I guess it still had enough friction to keep spinning too.

The collet spacing after extreme tightening

So after the fan came off again we remounted the fan, and this time I tightened it as hard as I could. I'm not sure it will dissassemble, but it should at least stay assembled. The collet is down to 2 and a little threads showing vs the original 4 or so when untightened.

We ran it through several tests and it stayed on. Here's a video of the last test where we throttled up to test it at full RPM



Before we got started today we added some handles to the corners for easier moving when off. Here's the timelapse through the first engine test. The battery died during our dissassembly and cutting down the engine mount extra parts.

Here's the view of reassembling the fan and installing it.



The final test on the blocks was interesting as we throttled it up, and in the process we watched the cowling counter rotate from the propeller. I'm not sure exactly the phenominon which caused that. Watch below. The yellow cowling has marks on the sides for alignment.

Skids for the bottom

Today the hovercraft received skids on the bottom as well as reinforcment for the lift fan housing.

The plans call for 1.5"x1.5"x6' skids on the bottom. I instead used the left over 2"x2" stock from the cockpit on the bottom. (Of course And I had 10' lengths so I made two 5' skids, but that seemed really limited for balance. So instead I made another pair, and overlapped them just slightly since the interior space is about 9.5'. Have a look below at the final result.

The skid strips installed

Here's the timelapse of today's work. 

At the beginning we added some reinforcements to the housing around the lift fan to make it even more solid. Now it has plywood in all three directions so it will be much stiffer to any disturbance or vibrations from the engine. 

New Lift Engine Mount

With the help of Michael, we recreated the lift engine mount today.

From the first hover trials the available surface area of the lift fan was significantly limited by the large surface area of the engine mount. A rough estimate is that it was covering 70% of the surface area of the fan, leaving only some small slivers of intake areas open. You can see that all but a 6" chord at the front and back of the fan were open in this post.

For a quick comparison you can see the new mount preassembly on the left with the parts cut away on the right.

The parts of the new mount next to the parts of the old mount.

Here it is on the fan, you can see the front and back has close to double the area, as well as large areas on the sides open as well. The main blocked area is basically under the engine now, and is also  near the center of the blades which is not an efficient area anyway. 

Fit testing the engine mount before gluing.

Here's one last shot of it waiting for the epoxy to harden.

The new engine mount in the clamps waiting for the epoxy to dry.

 The construction technique was to cut down the boards previously used to a 10.5" square around the engine mount. Then aluminum c-channel is fitted side to side and along the sides of the board. That aluminum c-channel is then bolted to 1" steel square tube. The c-channel is epoxyed on the sides of the boards which are sanded down to a press fit.

Looking at the design I think the aluminum c-channel all the way to the edges might be overkill and just more air resistance. I may go back and trim the outside 6" off since the steel should be strong enough on its own.




In this process we had quite a bit of epoxy left over. Michael noticed it was getting hot enough to be either smoking or steaming. In the following video you can see the epoxy puck which had been created by the hardening epoxy still in the bowl. In the first second you can see a little bit of the steam/smoke right by his hand in front of the cabinet. I guess the perimeter is still not yet hardened since it's being cooled by the environment.

It hovers!

In a very momentus event, my hovercraft took it's first flight today! Many thanks to those who were here and gave a hand keeping things under control: Adrian, Ayse, Eric, Jenny, and Wim!

Here are a couple views:

Top corner view:


Lower corner view:


And slow motion side view:

We did start with a fan test and we could see that the bag was successfully inflating.

 It's been a long time getting to this day. It's really awesome to see it all finally come together and hover. It's been a long project and I've gotten distracted many times by many things and pushed this project onto the back burner. But now it's hit a big milestone. There is still lots of work to add the thrust motor as well as better controls. Manually reaching forward and increasing the throttle is not optimal.

There was some excitement in the tests. At the end of the flight the lift fan did fall off of the engine shaft. That was the big bang at the end.

It did validate my design choice to use an enclosed metal duct. Despite the fan coming off while still spinning, there is almost no evidence of it being a problem. After disassembling it there were a few wood shavings in the fan housing, and the end of one blade is a little bit scuffed.

The scuffed fan blade having fallen off the engine shaft.

Wood chips in the bottom of the fan housing upon dissassembly.

The fan blad resting at the bottom of the housing after the first flight.

This reaffirms my concern about the loosening bolt from the previous tests. I'll need to look more into how to make the collet more robust to vibrations. In this case again one of the bolts had again loosened during the testing.

It was interesting that I think it only fell off at the end due to the positive pressure provided by the thrust. I suspect that part way through the test it had loosened enough to start slipping but it was only partially slipping and was held up by the back pressure.


As a side note upgrading the engine mount will definitely be valuable for decreasing vibrations as well as opening up more fan area for through flow. Right now the engine mount is covering most of the fan's surface area. With that more open it should be able to hover higher. It was adequate for this test on very flat ground but I think it will need more in the future.

I'll also take the time to add some more sides to the cockpit near the forward engine to stiffen it up.

The vibrations in the redwood are amazingly large in amplitude. They can be seen clearly in the engine test videos. 

Lift fan running smoothly

So after a little debugging I discovered that there was slight eccentricity between the ducting and engine. I'd checked the blade everywhere, but the problem was with the metal grill where the hub went through it.

So I ended up having to take the whole system apart again. I made a slight mistake in that I mounted the blade slightly higher than before which made inserting the removal bolts much harder.

I also noticed that the grill works better with the radial ribs upward so it fits around the engine mount bolt heads better.

After disassembling I confirmed the impingement was between the collet bolts and the grill.

Top bolt with visible nicks from abraiding with the grill.

The abraided fan grill. The ones at the inner ends are from this incident. The loss of paint at the 3 and 4 inch radius areas are from the original electric fan motor mount.

When disassembling it I noticed that the nicked bolt was actually completely loose. Before going too far I will want to add some lock tight.


At the end of the timelapse you can see that I enlarged the clearance. However the bolt cutters left large sharp stubs that I felt were suboptimal.

So I stopped progress and went to the store to fix that.

The grill after using bolt cutters on it.

You can see my grinding down and cutting off of the stubs, followed by filing them down with hand files.

I reassembled everything. I'm getting much faster at it now. With the larger interior hole it's much easier too. And I made sure to not mount it quite as high up the engine shaft for clearance for future dissassembly.

Once it was all reassembled I then tested the engine again. It ran smoothly and was drawing air in. I think we're on track for trying to hover tomorrow.

Lift fan testing

So once I finished testing the lift engine the next step is to test the lift fan.

Here's a video of the first two test runs of a few second each.


The first one went well and I stopped it conservatively.

The second one developed a loud rattle so I stopped it asap. I'm still debugging where the sound came from in the second run. 

Lift engine testing

So I'm getting close to operational and I wanted to bring up my engine.

I found the serial number on the engine and entered it. But no dice,"Found 0 results matching "120502-0116*"" I followed the video, and found the

Briggs and Stratton manual finder failure.

Despite the fact that I followed their video tutorial on how to find the number and I'm quite sure it's right:

The Briggs and Stratton model number printed on the sheet metal above my muffler

I ended up pulling a PDF off of manualsonline.com it wasn't quite the same model, but it at least gave me the basics. Though it did not tell me what the kill switch was.

It's the little wire coming out the front that you need to ground to kill the spark.

I started by stripping down the lift engine to test it unloaded first.

Then I tested the engine. See the timelapse below. The engine was pleasantly quiet unloaded. The throttle worked fine, however the advise I got from the manual about travelling all the way down on the throttle was incorrect. So I needed to find my aligator clips to short out the coil and stop the spark as I thought the extra little wire was for that.

The one thing about starting the engine that was a little disconcerting was that it was quite smoky obviously burning oil. Since I didn't have the real manual I had no idea how much oil to put in. And I was pretty sure the dipstick wasn't working well without turning things over. After a brief search I found the alligator leads and disabled it.

On inspection though I found this pool of oil below the muffler:

Oil pooled below the muffler after the first run.

Now I think that this is ok. The oil I put in is clear and this is defnitely black. So it's not the oil I put in. Which means that it's oil which was already inside the engine. And several times during the assembly process I had turned the motor completely upside down. So I suspect that I put oil into inappropriate places, such as the muffler. After running the oil on the dipstick looked to be right at the full line so I think that's ok. And by the time that I found the alligator clips the exhaust was much cleaner as well so I think the excess oil has mostly burned off now. The manuals aren't very useful for defining first time bringup procedures. But I'm betting that this also might be expected for the first run due to preservatives added during shipping.

 As a side note, the new "eco safe" jerry cans are terrible. I had almost as much fuel running down my hand as I did going into the fuel tank. I don't really know where it was coming from either. Maybe next time I'll try inverting it and relying on their seal to stop it until its pressed into the tank. But standard pouring just doesn't work. 

Adding baffles around lift duct

I added baffles around the lift fan to block the air from coming right back up next to the duct. This fixes the problem of a round peg in a square hole. I also added a panel on the front to help strengthen the cockpit structure. Below is a timelapse.




You can see the 4 corners glued into place such that the air will not be able to flow back up the corners around the fan duct.

While I was at it I also glued and stapled the small spacer pieces which hold the fan up above the top of the fan housing. 

The engine mount spacer block freshly stapled and glued with the corner baffles also recently installed.

Reinforcement at the front.

I also cut a piece for one of the sides, but I realized it would impinge on my ability to access the engine bolts so some affordances will need to be added before adding it as well as the matching one on the other side.

A close up  of the corner baffle. I ran some extra expoxy along the edge to help protect it from wear since the unsupported plywood felt a little vulnerable.