Model Hovercraft project. Help finishing and releaseing to this awesome site :)

[RoboWanabe]

Hello.

I have been working on a hovercraft project on and off for some time now and i managed to get the circuit working perfectly and then i blew a few components which id like help finding the problem? But other than that its finished and i just need to work on the actual hovercraft its self i.e. base and curtain for the air cushion. So first i will layout my plan and the problems ive encounterd and once its finished it can be used by every one

Attachments
1) Ciruit diagram
2) My bits and pices
3) My design

plan
So first the main part is the circuit for the hovercraft which i have attached. You will see that i have used a psoc microcontroller for the brain of the craft but this can be changed to a micro of your choice with the same capablities. I am then using two fans. One for the center fan pushing air into the cushion and one attached to a servo to control the movement of the craft. To control these devices you can use a pwm signal from the microcontroller.

To communicate with the craft i am using an ifra red reciver module but i would like to discuss changeing this to Radio controlled for greater distance?

Problems so far?
1) I have broke the two voltage regulators used to power the psoc and the servo. One has stopped outputting completly and the other has a reduced voltage at 2.33v
2) i have blown one of the MOSFETS used to control the fans. saw a pooof of smoke :O

i would like to point out that i am using heat sinks on the MOSFETS and voltage regulators.
I could do with help solveing these problems?

So thats all i can think of including at moment but if there is anything else that you dont understand please ask or any improvements please share.

Thank you

 

[Smartie]

I've looked up the specs of the mosfets and the regulators and they check out fine...
Maybe you've got loose wires that may have shorted causing too much current to flow through?

 

[jpanhalt]

Google returns nil for P22NF037. Alldatasheet.com returns the STP22NF03L. That mosfet requires a Vgs of 10V to turn on fully.

Can you link to the datasheet for the actual mosfet you used?

John

 

[Smartie]

that's the same datasheet i looked up. If you look ar the Rds(on) you'll see it has "VGS = 5 V, ID = 11 A, Typ = 0.045, Max=0.06" That's the on resistance where as with a 10V Vgs, the typ only goes down to 0.038. which is still low enough to let plenty of current though.

If you're unsure, about the specs on the mosfet, hook it up to a power supply, with a Pot on the gate, that's also connected to the 12V rail and ground to vary the voltage of the gate, (maybe put a load on the mosfet to prevent it from burning out) and measure the current going through the mosfet as you increase the VGS voltage then measure the voltage drop (VDS) to calculate RDS

 

[RoboWanabe]

Hello thanks for the reply

Here are the data sheets and you might be write about shorting the ciruit, i am using strip board and my soldering is bad on a good day haha but i just wanted to make sure ive coverd all angles befor i buy more and blow them again

i was wondering what your opinons where on me not using capacitors on the input and output of the servo bith the regulator? is that not a good idea? and can anyone explain the purpose of them im not too sure?

Can anyone sujest any improvements on the circuit or have an idea of using Radio control for communication?

Thank you.

 

[RoboWanabe]

Sorry this never got sent with last message lol

Oooops sorry it did lol

 

[Smartie]

the capacitors help smooth out any voltage spikes that may occur from the motors running, i'm sure the servo would already have internal capacitors and other protection. as for optimising the circuit, there's not much you could do there. and as for radio comunication, you could try heading to a hobby store and buying a transmitter, receiver pair (simple PWM output from receiver) if you want a classic way to control your hover craft. Or you could have a go at building your own remote. Maybe read this article and see if that interests you?
https://hackaday.com/2012/05/15/universal-20-channel-project-controller-from-a-ps2-controller

 

[Smartie]

If you go with building your own remote option, there are a ton of different radio modules from SparkFun (https://www.sparkfun.com/categories/16?page=all) that you could use

 

[RoboWanabe]

Thank you so much

Ive had a look at your links and i am very intrested i think i might give it ago it dosent mention the reciver module tho? would i just buy a reciver and then hack the controller to work with it?

OK so the acutal design of the hover craft is ok i just need to go buy some more components solder them better and then hopefully it will work. So i will go and do that and look t this radio controller

Thanks for your help again and if you can just keep an eye on this thread because im sure i will have other questions

Thanks agin.

 

[Smartie]

with XBee's, they're transceivers, bi-direction communication. so just buy two modules and they'll talk to each other happily

 

[Pommie]

I can't see the flyback diodes on the mosfet board. Unless it's the things on the right. Have you fitted the flyback diodes?

Mike.

 

[RoboWanabe]

Smartie: I get it now thank you. Some times things just arnt as complicated as you think theyd be i will get back to you on my progress

Prommie: If by flyback diode you mean to stop the back EMF from the motor then i think you just need to put it in parrallel with the motor or is that something completely diffrent lol sorry still learning.

 

[Pommie]

Yes, the flyback diodes are across the motor. Not having them will cause the mosfets to burn out.

Mike.

 

[RoboWanabe]

Ok thank you Mike.

 

[RoboWanabe]

Hello again

I have checked my circuit board and i dnt believe i shorted any components so i dont think this was the problem.

Im thinking maybe its because ive solderd and desolders the components about 3 or 4 times now. Do you think i could of melted the components material inside ( i am terrible at soldering) for e.g. the mosfet i know is just 2 types of material snadwiched together? do you think i could of fused the materials together by keeping the iron on too long thus destroying the deplition zone in the mosfet causeing it to be constantly on?

This is just a wild thought if anybody can tell me if i know what im going on about and if i could be correct i would much appreciate it

Thank you

 

[Boncuk]

Hi RoboWanabe,

I guess you've saddled your horse from the wrong side.

A good aerodynamic design should be paramount over electrical and electronic considerations:

Using a ducted fan for the air cushion is not all you need. The air cushion needs to be strong enough to lift the craft over water level without the help of a curtain. The curtain is just used to keep the air underneath the vessel as long as possible to compensate for water waves. The fan pressure side must end in a diffusor which reduces air speed and increases air pressure, hence lift.

A good ratio of duct diameter vs. diffusor diameter should be 1:1.4. Example: fan duct diameter 100mm, diffusor diameter at its end 140mm.

The second problem you'll encounter is (almost) uncontrollable direction of the vessel due to torque of the lift drive. You can use two lift drives at opposite rpm with close to equal rpm to zero torque.

That way the lift drives can be held relatively small and enable the vessel for a good payload.

Having the lift free of torque you just need one more drive for forward speed and direction control by pivoting the drive motor including the propeller. Depending on the mechanical works you put in you can use the directional drive for reverse as well.

Google for really impressing Hover craft crossing the English Channel daily.

Boncuk

 

[RoboWanabe]

Hi boncuk some useful information there thank you. ive been trying to get my head around it but im finding it hard to get any usefull material do you have any sites to sujest and could you explain what a diffusor and the duct is please? im thinking now that i was taking it seriously and need to plan the design better i thought onece the air cushon was full the pressure would then billow out and create a kind air cushion and if the weight of the craft is controlled to a center point it would just stabilise and not move? is that not correct. thank you for your help

im still unsure about what the problem was with blowing the mosfets and voltage regulators, im pretty certain they wernt shorted in any way? could some one please help me with this as this i would like to solve this problem befor i buy new components?

Thanks for your help

 

[Boncuk]

Hi Robo,

I've made up a little sketch of basic axial fans and related pressures. I omitted the tail cone for the motor which increases the effectiveness of the diffusor.

The principle of a diffusor is as follows: If the fan duct diameter is increased the air speed inside the duct (diffusor) will decrease and hence increase fan pressure.

For a hovercraft you will need pressure instead of volume flow. That's why you should plan for a short diffusor - long enough to build up pressure slowly.

For your control circuit please post the schematic for the entire fan speed controller and used components.

Regards

Boncuk

P.S. My best regards to Lady Godiva

 

[RoboWanabe]

Hi boncuk.

If when your talking about fan duct you meant the imediate casing around the fan then im not sure i can change the dimension of that as its fixed. And diffusor is the output of the fan ( kind of like a nozel or funnel on the output) ?? im still strugling to find some reading material on this. Cheers for the diagram as soon as im not busy im going to get working on the design again.

The spped control is performed by PWM from the psoc which activates the mosfets on the output pins which controls the amount of power deliverd to the load? You can see it on the schematic ive put up?

I hope you can be patient with me lol

Thank you.

 

[Oznog]

First guess is you are using a transistor with a high gate voltage with only 5v drive, but I need the spec sheet.
I read this as a "p22nf037" transistor, but Google finds no hits for this.

Frankly this will work MUCH better with a brushless ducted fan and a brushless controller. These are dirt cheap now and a brushless motor is far more reliable, more efficient, and more powerful. Quieter, too. Talking to a brushless controller from a micro is not difficult. The controller will implement the power stage and that offloads the difficulty of you trying to design power electronics.