View Full Version : Any ideas why my experimental hovercraft doesn't work?

3-Aug-12, 09:34 PM
A friend and I are trying to build a hovercraft as a summer project, and we're hoping you guys can offer us some advice!

We bought two 6.5hp ventilation fans as used by the fire brigade at a great price, and plan to use one for thrust and one for propulsion. Obviously it won't be speedy, but we just want to make a working hovercraft.

We've put the lift engine at the front, canted down and blowing into a bag skirt, which then vents underneath the hovercraft. Once we've got this working we'll add the second engine on the back for thrust. Pictures are attached.

We used the lift calculator here (http://www.hoverhawk.com/lcalc.html), and the skirt design was based on this (http://www.astounding.org.uk/ian/hovercraft/skirt.html) and [URL="http://4wings.com.phtemp.com/tip/bag.html"]this[\URL].

Unfortunately, the hovercraft does not produce enough lift for a person and a second engine, and even when lightly loaded and hovering, the friction is pretty considerable. We've tried adding holes to the skirt, and added holes directly into the area underneath the hovercraft from the fan, but this has not dramatically improved performance.

From the pictures, does anyone have an idea how we can improve the performance of our hovercraft? We feel like the performance is there somewhere, but are now in danger of aimlessly making alterations!

Many thanks for any advice you can offer.


4-Aug-12, 06:27 AM
welcome aboard, I am no expert and others will be along that are far more knowledgeable than me. You will also get many differing opinions; it's the nature of the beast.

I am assuming you have used 12mm ply, so weight could be one enemy, another no doubt will be your fan choice. It may well be that you don’t have enough air pressure to gain lift. My understanding is that air volume and air pressure are two different beasts you can have plenty of one without the other. As I said earlier others with a greater understanding of these things will be along to explain things far better than me.

Yet another enemy could be the surface you are using it on. A flat hard surface like tarmac will give best results. If you are using it on grass then it will need to be VERY short not to lose any lift through the surface.

I take it the craft is not for use on water, with that in mind... Are those pictures taken with the lift motor running? If not then it would appear that you have a lot of structure under the craft you could lose.I would remove everything below the top sheet and attach the skirt directly to the single sheet of ply that is left.

what ever you do please report back as it would be nice to see if your efforts bear fruit.

Keith Oakley
4-Aug-12, 08:33 AM
I'm surprised - it ought to work at least on the flat concrete surface you show in your pics. Is it hovering in the pics or is it at rest? It looks like the fan isn't developing enough pressure to lift the craft. 6.5bhp should be enough but it may be set up more for flow than providing pressure. Whats the tip clearance like between fan and duct? - more than say 0.5inch will loose pressure. I presume theres no air leaks between the fan and hull. How heavy is the whole craft and whats the size of the hull? - weight/cushion area defines the pressure.

Keith Oakley
4-Aug-12, 09:08 AM
Just looked at the website you quoted. First time I've seen it and to my amusement this section is entirely a (nearly word perfect) lift from the HCGBs Light Hovercraft Handbook which I edited in the 70s and is now out of print. But crucially they've left out all the diagrams!
The reason I asked about whether the craft is hovering or not is that I'm not sure you've got the skirt set up for enough hoverheight - referring to the missing fig6 attached what are your dimensions H and h ?

4-Aug-12, 01:54 PM
Thanks for all the advice! I'll try and clarify a few points:

- Gaz
We have used 9mm ply, two boards separated by 8cm wood planks around the edge and in the middle. The weight is certainly not insignificant, but I don't think it's excessively heavy compared to the weight of the engine.

I think you have a good point about air volume and air pressure. When the engine is running and the hovercraft is on the ground, around the rim of the fan air is flowing back away from the fan out of the hovercraft. This doesn't happen if the hovercraft is lifted up away from the ground. We've had a look at the fan clearance, and at most the tip is ~5 mm from the fan case; we're not sure if there's a robust way of reducing this. Perhaps these fans have been designed for operating with a low pressure differential?

We don't know much about different fan designs, but presumably as a ventilation fan this has been designed for a big volume flux over a small pressure differential. Do you know whether if we replaced the fan with something like a propellor, would it potentially support a greater pressure differential?

The pictures are with the engine running, so the craft is hovering, but about 5 cm.

- Keith
In the pics, it is hovering, but not very high and it doesn't support much weight.
Just a question about the theory: provided the skirt provides a good seal, is the pressure to support hovering by 2 inches significantly different to the pressure to hover 4 inches?

As I mentioned to Gaz, the tip clearance is ~5 mm, but there is definitely a back flow around the rim of the fan.

The engine weighs 43 kg, and the rest of the craft can't be much more than 40 kg. The base is an 8x4 ft (2.4x1.2 m) board and the skirt is attached 25cm from the edge.

Good to know you edited the book! Thanks for that information then.

I think you might be right about not enough hover height… We kinda of guessed at that and chose what seemed like some moderate dimensions. The cross section of the skirt is this:


It's a pretty primitive design! I don't know whether it's got too big a GC area and that is causing excessive drag?

Thanks again for any advice!

Keith Oakley
4-Aug-12, 03:51 PM
This ought to work - its just a question of working out why it doesn't!!

The fan tip/duct clearance of 5mm or less should be fine - many racing craft run fine at 10mm but I'm surprised you're measuring a reverse flow

You could have got away with 2 sheets of 4mm ply rather than 9mm but the excess weight there is small and not the problem.

The cushion area (8 * 4 ft = 32sqft) is about half that of most single seat craft thus the cushion pressure is potentially higher. Its going to need 2 engines to trim level so that gives a basic weight of 2 *43kg +40kg hull = 277lbs. Over cushion area of 32sqft that gives cushion pressure of 8.6lbs/cuft. With say a 150lb human payload that is a total weight of 427lbs and thus a cushion pressure of 13.3lbs/sqft - highish but there are plenty of craft running at that pressure.

The skirt design suggests you are aiming for a 1.5:1 bag/cushion pressure ratio which is fine but your target hoverheight is 9cm and you're achieving 5cm so the skirt is flattening out increasing your ground contact (GC) area and thus drag. Have you got the size and no of the feed holes in the inner bag wall correct to achieve a 1.5:1 ratio? Temporarily closing some off might be informative. Testing pressure with a water manometer at different points in the airflow (behind fan, in bag, in cushion) might also give us clues.

I'm left wondering about the fan. As you say it may be designed to shift volume rather than achieve a pressure and with the 1.5:1 ratio you need the fan to deliver 13.3*1.5 = 20lbs/sqft. Although the nominal pressure at 5cm is the same as at 9cm hoverheight if the fan is 'straining' to deliver 20 then the flow may be too low at that pressure. Moving to a prop isn't the answer, generally you need more 'solidity' to deliver pressure so more blades the better. I guess your big investment is in the engine and fan so changing that isn't optimal. As its only a test rig you might want to try increasing the craft size (using say 3mm ply) to reduce the cushion pressure.

Just as a check I assume the engine is actually revving to its nominated speed?

I bet others have got better ideas! Anyone used such a fan for lift?


4-Aug-12, 08:03 PM
My 2 pennorth is that the lift fan is so obstructed that it's efficiency is questionable.
It's hard to actually see the fan.
I would certaily investigate that area.


5-Aug-12, 06:37 AM
All helpful stuff, told you far more knowledgeable people would be along.

Adding to Keith and Kips advise, most of the 'useful air' comes from near the blade tips (the last few cm's), try to make any transitions as smooth and as big as possible so as not to slow or ad turbulence to the air you are shifting. You mention an 8cm gap between the two sheets, are the separators restricting air flow anywhere? Are the holes in the top sheet as big as possible to aid flow. if you can feed the air directly into the bag it MAY help.

If you are in the UK then one of the few places you will get the thin ply Keith mentions is Robins timber in Bristol (http://www.robbins.co.uk/) They are very helpfull and they have a delivery service.

Scuba Kev
6-Aug-12, 01:20 AM
Again I'm sure someone will come along who knows more about a bag skirt design than me but my view is that the holes in the skirt are too big and the fact that there is no contact strip around the bottom to hold the air under the craft to give a constant cushion. Keith Smallwood or Jamie Lewendon may be along to give some pointers

6-Aug-12, 07:40 AM
Usually the bag skirt needs more pressure i.e. more blade were a finger skirt needs more flow. Could be the problem of these fans not keeping enough pressure in the skirt.

7-Aug-12, 05:12 AM
My guess is the lift fan is constantly stalling (producing inadequate flow, over-revving maybe) because the footprint (cushion) pressure is too high for the pressure ability of the fan because the fan pitch is set for thrust. The most effective way to bring the cushion pressure closer to the fan's ability is to double the footprint or even 2.5 times it.

Do you have some way to measure the pitch of the fan blades ? fireman's fans will likely be maximised for flow which is great for thrust, whereas lift fans need to be both for pressure plus with the right amount of flow to sustain the throughput, lift fans tend to be around 30 deg and thrust fans around 45 although some integrated craft can run fans somewhere in between in the high 30's ~40.

If you can measure the pitch, design the craft around what you have rather than trying to make the fans work in a type of design they are possibly less suited to.

You have almost enough gear there to build a very nice two thrust fan 'eccles' or 'simple cyclone' type craft if you increase the footprint that will be best improvement you can ever make, and use your weight at the front to counterbalance. Maybe first try an integrated air flow bag skirt and if still not working as it should, go to a vertical mower engine separate lift fan around 30 deg and get the best advantages of separate lift.

Also take a good look at the Hovery hovercraft, it might change your thinking a lot :)



Paul Fitz
7-Aug-12, 07:22 PM
I suspect that you have identified the problem. If there is very significant 'spill' from the tips of the blades it is most likely that the fan is not producing sufficient pressure to overcome the system resistance. Ie. the required cushion pressure. Most fans will have some spill but when working near their most efficient duty, it will be minimal.

The calculator you used gives a reasonable duty,although In my experience, the power requirement given is too low, as it appears to assume a very high efficiency. You need to match the fan to this duty. Some fans will produce low volumes of air at high pressure and others High volumes at low pressures.

The fans you are using appear to be cooling blowers which are likely to be of the latter type. The fan blades appear to be Aluminium ?? in which case the fan is probably running at about 1500 rpm. (Aluminium fans are NOT recommended on hovercraft).

If you can give us more information about the fan perhaps we can offer more precise advice.

From the photos, I have estimated your cushion area at 29 sq ft (2.69 M2) and the required fan duty at

0.47 Mtr3/sec @ 453Pa -- 83 Kg weight
0.56 Mtr3/sec @ 656Pa -- 120 Kg weight (assuming the second fan on board and a 55kg passenger.

Sorry you will have to convert to imperial. I stopped using it 40 years ago :o)


8-Aug-12, 08:17 PM
Thanks for all the tips guys! All really appreciated.

I think you hit the nail on the head by pointing out the core of the problem is too high a lift pressure, and the fan being inefficient to pump across the pressure differential.

We're added another 4x4 ft board on the back, and extended the length of the skirt. The extension is very lightweight. Performance has been significantly improved, and it will now just about work on short grass, and well on concrete. The balance is very sensitive though, and it's quite easy to cause a side to knock the ground.

I think we need to aim to increase the width as well, just to further reduce the pressure required, but this is a less straightforward alteration.

Yamah, what do you mean by an 'integrated flow bag skirt.' I think the next thing to play around with is the arrangement and size of the bag holes, as Scube Kev points out. Also, what form would a 'contact strip' take? Just a curtain of fabric hanging down at the ground contact point?

Unfortunately, I think converting to a finger skirt would require a scratch re-build.

Thanks again for the comments!

9-Aug-12, 06:16 AM
Hi Alexmac,

"Yamah, what do you mean by an 'integrated flow bag skirt."

easy :- integrated :- fan does both thrust and lift, or in your case 2 fans do both thrust and lift and you already know what a bag skirt is.

Look at the Vortex Venom craft and maybe adapt that principle to two fans at the back like the eccles. Two fans at the back used to be quite common a while ago and it worked fine, you have two fans so consider that format as a possibility. IF you decide to develop that further later on, you can put a vertical shaft motor mower engine at the front with a >lift< fan on it and utilise the two fans at the back just for thrust which is probably what they do best anyway.

UH hovercraft have some smallish and some way too smallish entry level craft which utilise small horsepower like you have, albeit in a single motor/fan unit and the fan does both lift and thrust from the rear and they have a bag skirt. Typically the UH craft have a low hover height but they use their minimal hp very well so taking note of how they work will probably help you. look up UH-12 build and you will see the underneath stuff.

Another craft that goes very well on minimal hp is the sevtec scout family, and even the now discontinued fold up predecessor The Fan-Tastic-1 that could plane out of deep water on 4.5hp, that was basically a shallow rectangular plywood box so it wouldn't sink and the decks and most else was fabric and thin frame work for weight savings. http://www.sevteckits.com/catalog/index.php?main_page=page&id=21 scroll down, there are three or four pics and a sketch. One pic is climbing a steep river bank albeit with an 8 hp power plant. This is a sadly neglected design and good for a back yard project.

imho the either the Venom/eccles cross and/or Universal Hovercraft principles would be the easiest for you to copy from with what you have, laid out like an eccles, save weight wherever possible, if you think it needs a 4x2 then a 2 x 1 will probably do if you are smart about it, Same with the 9mm ply, 3mm would be heaps better, this is a low flying aircraft after all.

Have Fun


9-Aug-12, 06:45 AM
3mm ply bonded to polystyrene would make a very strong & light platform hull.

10-Aug-12, 11:11 AM
Yaham - the hovery hovercraft is incredible!

Paul Fitz
10-Aug-12, 11:23 PM
Alex,your craft is a simple experimental craft. It will never be truly practical for racing, cruising or pleasure on anything other than a football pitch or carpark, so keep it simple and learn from it. Forget about a contact strip on the bag and do NOT try using both fans together in an integrated setup, it simply would not work. Each fan would produce slightly different pressure leading to one passing air back through the other lowering both fans performance.

Read the articles on the downloads page of this site, they may help your understanding of the principles. carry on as you are going. Once you have lowered the cushion pressure and you have adjusted the balance with the load on board I am sure you will have some success. Balance is important. with the driver on board the balance point should be at the centre or just forward of the centre of the cushion. This will make it slightly rear heavy which will compensate for the thrust and stop the nose being pushed down under power. Don't over-do it, you will not have a lot of thrust.

Again, if your fans are Aluminium, consider changing them for the Nylon type. Aluminium fans can be lethal if they break.

Good luck and have fun.

Paul Fitz