Yet another wireless power transmission thread.

[sbstn]

Hi everyone, this is my first post on the forums

OK so I want to design a microwave power transmitter and I need a few opinions before I make any commitments. I haven't made many complex circuits before, but I am going to do my homework if this is viable and break it into chunks.

Maybe this will be a lot harder to achieve than I think, but bear in mind the system doesn't need to be very efficient.

I need to beam maybe 50W max across a distance of 15 centimeters. The rectenna would have to be 5cm across but would be more or less static.

So my initial questions are -
Are these totally unrealistic specifications? Has anyone actually played with microwave transmission at all? How simple could I make this design?

Thanks guys ( I did do a search ;) )

 

[berkeman]

Hi everyone, this is my first post on the forums

OK so I want to design a microwave power transmitter and I need a few opinions before I make any commitments. I haven't made many complex circuits before, but I am going to do my homework if this is viable and break it into chunks.

Maybe this will be a lot harder to achieve than I think, but bear in mind the system doesn't need to be very efficient.

I need to beam maybe 50W max across a distance of 15 centimeters. The rectenna would have to be 5cm across but would be more or less static.

So my initial questions are -
Are these totally unrealistic specifications? Has anyone actually played with microwave transmission at all? How simple could I make this design?

Thanks guys ( I did do a search ;) )

Welcome to the PF.

Can you outline the safety considerations for such an experiment? What would be the safety concerns, and how are you addressing them?

Can you outline the FCC regulatory concerns for such an experiment? How are you addressing these legal regulatory concerns?

Do you have a faculty advisor for this project?

 

[sbstn]

The primary consideration would be RF shock from touching the antenna, and RF heating. I would be in the range of 10-20Ghz if possible, so the heating effect would be less of a danger though obviously I will take precautions.

And it would be OFCOM in my case, not the FCC. If I were to use 10Ghz I would be on amateur radio grounds though I would still need a license.

If i were to operate everything in a faraday cage could I get away from licensing issues ,as well as add a level of safety?

Nope no advisor, though I'll be attending UCL in a few months for something unrelated. Maybe I can find someone to help me out.

 

[berkeman]

The primary consideration would be RF shock from touching the antenna, and RF heating. I would be in the range of 10-20Ghz if possible, so the heating effect would be less of a danger though obviously I will take precautions.

And it would be OFCOM in my case, not the FCC. If I were to use 10Ghz I would be on amateur radio grounds though I would still need a license.

If i were to operate everything in a faraday cage could I get away from licensing issues ,as well as add a level of safety?

Nope no advisor, though I'll be attending UCL in a few months for something unrelated. Maybe I can find someone to help me out.

That's a good start. Yes, you should be doing this in a faraday cage, with RF absorbing foam in the target area. You need to be sure that you do not let RF interference out of your test area, or you could have issues with your OFCOM agency.

It definitely would be good to have an experienced local Mentor to help you with this project. What is your end goal for this project?

 

[sbstn]

Short term goal is to transmit as close to 50W as I can across 15cm.

After that I want to get the receiving circuit as small as possible to integrate into a quadrotor. I'll do as you advise and everything will be well protected.

In terms of what I have talked about (ignoring the quadrotor), do you think this is viable? I would love to use a much safer form of wireless transfer but they all seem to need large coils...

 

[mdjensen22]

http://www.witricity.com/pages/technology.html
This one uses resonant magnetic coupling. The technology is pretty interesting and some of the applications are very cool.

One avenue that I would also recommend spending some time is the patent listings for some of these inventions. There is a surprisingly large amount of technical data in some of of the applications.

 

[sbstn]

I've seen his TED talk where he uses a coil half the size of a cellphone, which is interesting but I wrote the idea off as too complex. I guess the idea is based on tuned LC circuits?
Not sure how much more complex the circuit would need to be than that (I chose microwave transmission because it seemed more straightforward, am I wrong?)
I'll have a look at the patents like you suggested.

 

[berkeman]

Short term goal is to transmit as close to 50W as I can across 15cm.

After that I want to get the receiving circuit as small as possible to integrate into a quadrotor. I'll do as you advise and everything will be well protected.

In terms of what I have talked about (ignoring the quadrotor), do you think this is viable? I would love to use a much safer form of wireless transfer but they all seem to need large coils...

How do the 15cm distance and the quadrotor concepts relate? Are you going to power a hovering quadrotor from below or something?

If you want to transfer 50W across 15cm, just use a powerful stream of air...

 

[sbstn]

Yeah that's right. There will be a pedestal with the transmitter 15cm below and the thing hovering at eye level. It's sort of a high-tech show piece...

I sort of need the thing to stay as still in the air as possible; I want to hook up a tiny microcontroller to pull it off (that much I can do). I think would be equally as difficult to get the aerodynamics just right.

Besides it would be pretty fun to get working don't you think?

edit: You seem doubtful?

 

[berkeman]

Yeah that's right. There will be a pedestal with the transmitter 15cm below and the thing hovering at eye level. It's sort of a high-tech show piece...

I sort of need the thing to stay as still in the air as possible; I want to hook up a tiny microcontroller to pull it off (that much I can do). I think would be equally as difficult to get the aerodynamics just right.

Besides it would be pretty fun to get working don't you think?

edit: You seem doubtful?

And how do you propose to make a clear see-through Faraday cage?

 

[sbstn]

A dark room and lighting the object from the inside would make it pretty visible despite the small hole size needed for the stuff. I would love to give the inductive coupling idea a try as it would free it all up but that seems too hopeful.

You sound very against this whole thing, level with me berkeman

 

[berkeman]

A dark room and lighting the object from the inside would make it pretty visible despite the small hole size needed for the stuff. I would love to give the inductive coupling idea a try as it would free it all up but that seems too hopeful.

You sound very against this whole thing, level with me berkeman

If you can do it safely and legally, more power to you. We just get a lot of fast-and-loose DIY folks who stop by the PF, and have no idea of the dangers and RF interference issues for some of their project ideas.

As a HAM radio person, I'm part of the groups that law enforcement calls when they have interference issues with their radios (or have been notified of interference with other licensed systems). We call them T-hunts (transmitter hunts)...

Stay creative, but please always keep safety in mind.

 

[sbstn]

No of course, and I think as I am displaying it at a university I would soon be stopped from becoming a liability.

If I were to cage off a large area, and then put perspex to stop people getting too close - would that be safe enough?

How directional could I make the antenna?

 

[berkeman]

No of course, and I think as I am displaying it at a university I would soon be stopped from becoming a liability.

If I were to cage off a large area, and then put perspex to stop people getting too close - would that be safe enough?

How directional could I make the antenna?

I still like the air-powered idea. Shoot a high-velocity stream of air up vertically out of a hole 12" to 18" wide, and make a wind-powered quadcopter that gets its power from its autorotating rotors. It starts on a "landing stand" over the air duct, and after the airstream starts and the quadcopter is powered-up, it can release itself from the stand, and rise up. Then it can hover, or turn slowly around, and maybe even do a little maneuvering in the airstream. That would be impressive.

(and a whole lot easier to make safe and legal)

 

[sbstn]

If your this unsure I guess the university will be doubly so, thanks for letting me down easy though.

How would it be powered through the airstream I'm not sure I understand?

 

[berkeman]

If your this unsure I guess the university will be doubly so, thanks for letting me down easy though.

How would it be powered through the airstream I'm not sure I understand?

You would have generators at each of the 4 rotor positions. The airstream turns the rotors, which generates some power for your on-board electronics and control servos. You control the pitch of the rotors individually to control your lift and other flight aspects. You can control the speed of the rotors by controlling extra electrical resistance for each of the 4 power circuits... Sounds like a fun project...

 

[sbstn]

That does seem like a much more simple way you've convinced me. (No servos needed for a quadrotor by the way.)

I'm still going to try r.f. transmission on the side though, I promise I will be careful with interference. Except now it won't be in the public space so you have certainly saved some lives.

 

[berkeman]

That does seem like a much more simple way you've convinced me. (No servos needed for a quadrotor by the way.)

I'm still going to try r.f. transmission on the side though, I promise I will be careful with interference. Except now it won't be in the public space so you have certainly saved some lives.

The servos I had in mind were to control the pitch of the rotor blades. I think you'll need that to control the lift of the autorotation, but I certainly don't know much about aerodynamics.

 

[sbstn]

Two rotors spin the other way to cancel out the angular velocity. Everything is controlled via rotor speed.

 

[Studiot]

UCL, not Imperial?

Don't know much about the background to your project, but Augusta-Westland have just announced an expansionist push and may be interested.

If it's a science fair thing
WBeaty here is a specialist in such matters.

Microwave ovens without a turntable transmit greater power than 50 watts by rotating the antenna.

See also 'Toshiba DeltaWave'

go well

 

[sbstn]

Yeah UCL

Hah I feel like this is far too amateur to be dealing with AW (I won't be doing anything engineering based at uni)

Not a science far, a component for a larger project. I'll try and PM WBeaty.

Working with a magnetron seems over the top - they generate a dangerous frequency of radiation in terms of dielectric heating.

And thinking about the wind turbine idea, it just doesn't seem viable at all. The motors themselves weigh only a few grams. berkeman if your still reading this; if I were to modify a wifi transmitter would I need to make the rectenna at least 7.5cm long?

 

[berkeman]

Yeah UCL

Hah I feel like this is far too amateur to be dealing with AW (I won't be doing anything engineering based at uni)

Not a science far, a component for a larger project. I'll try and PM WBeaty.

Working with a magnetron seems over the top - they generate a dangerous frequency of radiation in terms of dielectric heating.

And thinking about the wind turbine idea, it just doesn't seem viable at all. The motors themselves weigh only a few grams. berkeman if your still reading this; if I were to modify a wifi transmitter would I need to make the rectenna at least 7.5cm long?

You are limited for how much power you can transmit in the WiFi bands, so don't be thinking of transmitting many watts there. And yes, the size of the antenna(s) would need to be determined by the wavelength of the radiation.

And for the wind-powered version, you don't need motors on the quadcopter, just 4 little generators connected to the rotors. You tap the power out of the windstream to run the on-board electronics.

 

[sbstn]

OK thankyou for the help, I'll have to do some experimenting. I'll let you guys know what happens.

 

[es1]

Not sure how much more complex the circuit would need to be than that (I chose microwave transmission because it seemed more straightforward, am I wrong?)
I'll have a look at the patents like you suggested.

I think that means microwave power transmission is not a requirement. If so then I think the resonant magnetic coupling experiment is much simpler and more straight forward, especially if the load basically DC.

Pretty much all you need to do is built two LC tanks. Make the L by wrapping wire around shoe box or something. Use a C with a low ESR that can take high voltages. If you go for a resonant frequency of a MHz or so then the L and C can be pretty small.

Then put the two LC tanks in proximity of each other. The resonant frequency of the system and the load the secondary puts on the primary changes with the power delivery to the output so don't bother trying to match them too closely.

You can drive the primary LC with a half bridge sourced by a DC bench supply whose input is driven by any pulse generator (like a 555 and a potentiometer). Run 50% duty cycle and just tune the frequency until you hit the resonant point of the system.

A diode rectifier can by used get the power out of the secondary LC. Put two Cs in series to divide the Vp input to the rectifier if you have to.

Transient response is not so great so put enough C at the output of the rectifier to smooth it out.

In college I built something like this in about a day that could move 10W into a resistor (so DC) over about 20cm with ~40% efficiency.

You can probably get everything you need from digikey for ~$25 (assuming you already have access to basically lab equipment).

 

[wbeaty]

Short term goal is to transmit as close to 50W as I can across 15cm.

After that I want to get the receiving circuit as small as possible to integrate into a quadrotor. I'll do as you advise and everything will be well protected.

I've devoted much thought to the topic, :) but only ever experimented using Tesla Coils lighting up fluorescent bulbs, or running low-volt solar-cell motors remotely using capacitive coupling and a diode bridge to create DC.

The straightforward but hazardous experiment would be to use a microwave oven. See two such projects on youtube:

Microwave powered RC helicopters:
<< links removed by berkeman >>

A safer version would be to use a big desktop Tesla coil, high voltage, and capacitive coupling. To avoid scragging any electronics, start out with just a DC motor and rotor as your "helicopter." Perhaps use a half-volt solar cell motor at first, then go to standard motors as you optimize your setup.

I'd place a Tesla Coil main terminal below my helicopter. To keep destructive arcs from leaping, either seal the TC toroid terminal within a layer of heavy insulation such as a close-fitting oil-filled plastic tank, or put it inside a well-sealed thick plexiglas box, or just keep the helicopter high enough above the toroid that big arcs aren't triggered. (But it all should work better if the helicopter assembly could be very close to the main terminal.)

For your capacitive pickup antenna, make two thin metal foil parallel plates many inches diameter with several inches between them. (Perhaps the edges need a bead of silicone RTV caulk to prevent arcs.) Connect thin wires to the foils and run them up to the motor. Pass the circuit through a diode bridge to convert the high-freq AC into DC for the motor. Normal diode bridges won't handle the 200KHz TC frequency, so you'll have to build your own bridge using four fast-recover Shottky diodes such as 1N5819, or perhaps use a CDBHD2100-G bridge from digikey.com The bigger the spacing between the two plates, the higher the received voltage. The closer the plates come to the TC terminal, the higher the voltage. A TC with higher frequency should work better, or perhaps even find on eBay a 1940s home diathermy quack-medical machine (which is basically a CW continuous-wave vacuum-tube Tesla Coil at industrial heating frequency band near 27MHz.)

A perhaps higher-powered version would use invisible ion-cloud conductive coupling instead of plates. Do the same as above, but instead of the foil plates, provide a grid of downward-pointing needles below the helicopter (all needles connected electrically.) A second independent grid of needles pointed upwards goes above the helicopter. Run a wire to each needle grid, pass it through a high-freq diode bridge, and run your motor. The needle grid arrays could be quite large, yet shouldn't interfere with aerodynamics. And they might create a glowing violet nimbus when viewed in reduced light! (Also... instead of a complicated needle-grid, perhaps a simple tuft of carbon-fiber cloth from Tap Plastics or a fiberglas supply company would work just as well.)

In both of these systems the diode bridge removes about a volt from the received signal. If the output volts from the two wires is too small to run your motor, it should be possible to step it up by many times by using a low-weight small ferrite transformer from a switching power supply, or even build your own hand-made air-core pancake transformer.

The following one won't help you, but... Nikola Tesla in the Colombian Exposition provided spinning copper disks which were placed many yards from the 2-phase "Egg of Columbus" toroidial transformer he was using to create his rotating b-field. He was forming an "induction motor," but with a central stator coil, and many rotors yards distant. He cheated: the disks were on jewel bearings and suspended within sealed glass globes pumped down to vacuum! Hold such a globe adjacent the coil to crank up the RPMs, then remove it to any place in the room, and it keeps spinning forever (since it probably required far less than milliwatts to maintain the rotation once started.) I only found one reference to this old Tesla physics demo, and I've never heard of anyone reproducing it since the original at the 1893 Chicago World's Fair. Most physics departments no longer maintain professional glass blowers on staff!

 

[sbstn]

This is amazing, both of you :D I think I'm going to slightly change my goal and shoot for 10W, and try instead to make my copter more efficient.

I'm going to try the inductive coupling first to see if I can get enough power to meet my needs, that way I won't have to worry about shielding and giving people cataracts and whatnot. If not I'll experiment with your Tesla coil advice.
I'll be sure to post back by build.
Bye for now guys, thanks a bunch.

 

[wbeaty]

I just heard about a neat little schematic which keeps your transistors ice cold:

Zero-volt Switching RF driver
http://www.instructables.com/id/ZVS-Driver/step3/Schematics/

I'd just get rid of the flyback transformer and replace it with a big multi-turn loop antenna with a center tap connection. Use very heavy wire or the outer braid of coax as low-resistance conductor. Choose the value of that one capacitor to set the operating freq. (If it cap gets hot, perhaps need to use several series/parallel capacitors.)Ah, also I found the reference: the USA official RF power equipment frequency is 27.12MHz

http://www.fda.gov/ICECI/Inspections/InspectionGuides/ucm071626.htm

 

[sbstn]

This is very useful - presumably I could also use this to drive a resonance transformer?

If I replace the transformer with an antenna, is there any kind of setup that would negate the need for heavy shielding? Lower gain, more directionality?

 

[sophiecentaur]

Although it may not fit your original brief, it could be easier to have your position control 'on the ground' rather than in the 'vehicle'.
If it's just an 'illusion' you're after, it could be an acceptable solution.

 

[sbstn]

Can you explain that a little more?
All I need is a PID system (hopefully I can use SMD componnents) to keep it still. I feel like it would be more bulky if I added radio gear rather than have it more or less autonomous don't you think?
If you're thinking of a more physical solution, that would be grand..