Wireless energy via IR / Projects / Forums

Forums

4hv.org :: Forums :: Projects

« Previous topic | Next topic »

Wireless energy via IR

1 2 next

Move Thread

LAN_403

Conundrum

Mon Jan 14 2013, 06:42PM

Registered Member #96 Joined: Thu Feb 09 2006, 05:37PM
Location: CI, Earth
Posts: 4061

Hi all,
This is an ongoing experimental project to determine how much useful power can be sent approximately 30 feet from an upstairs window frame to my electric bike.

So far I've sourced two 3W IR LEDs, that are 940nm which is very close to the peak efficiency of PSC's according to the manufacturer's graphs.
The plan is to use a pair of fisheye lenses sourced from broken projectors for the two diodes, space them about 1 foot apart on their own heatsinks with silicone on all the delicate wiring and cool them passively with convection.
By pulsing them at approx 200ms on, 800ms off and alternating the drives there is a chance that 5.5W CW system power can usefully be achieved without adversely affecting the lifetime.

On the Rx end I plan to use 19% efficient "seconds" PCS cells connected as a series bank outputting 45V O/C at between 300 and 500mA.
Then charging up a capacitor and discharging it into a test load (ie 3 Pb-acid cells) as this seems to be the most efficient method according to several online sources.

As the "dump" power is essentially constant cycle to cycle, a simple Rx pulse counter should be enough to determine power transfer efficiency.

Comments?

Sulaiman

Mon Jan 14 2013, 07:57PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

I've been 'playing' with 1W 940nm and 3W 850nm eBay/Chinese leds
and lenses (mostly 127mm dia. glass from old bench light/magnifiers)
for optical communications;
I doubt that pulsing the leds will be of any benefit for wireless power transfer because;
1) optical output power increases linearly with led CURRENT but as the maximum current is approached the optical watts per amp reduces
2) as current increases so does voltage/electrical power/heat due to internal resistance of diodes
3) optical efficiency decreases with increasing die temperature.
The nett result is that for communications there IS benefit in pwm etc.
BUT for power transfer CW will be the most efficient AND most power output.

Beam divergence of the led output will be quite large unless a fairly large diameter/area lens is used in conjunction with a small high power lens near the emitter
(mine are +/- 60 degree)
Power reception also needs a fairly large lens, glass lenses are good,
Fresnel lenses are bigger/cheaper.

P.S. Hacking a microwave oven into a transmitter would give greater power
AND act as an anti-theft device !

P.P.S. What is a PCS cell ?

klugesmith

Mon Jan 14 2013, 10:11PM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

Both of you,
please figure where your IR projects stand with respect to eye safety limits.
The hazard is painless retinal injury to someone in the near-IR beam,
who happens to look in the general direction of the source without protection by visual reflex.

It's easy to find numbers and discussion of the safety limits. For example,

Conundrum

Tue Jan 15 2013, 05:45AM

Registered Member #96 Joined: Thu Feb 09 2006, 05:37PM
Location: CI, Earth
Posts: 4061

My understanding is that the IR output of these diodes is substantially lower than a heat lamp which is known not to cause damage.

Also they are used extensively as security "invisible" lamps and are sold as such, they even do 100W IR emitters and TTBOMK there have been no issues with either these or standard multidiode units..

One way I could reduce the risk is to add power feedback and beam steering, using a 0.3mW IR laser (eye safe, used in optical mice) shining from the centre of the solar cell array with a modulated beam.
If anything gets in the way of the beam or there is some anomaly it drops power down to 5% until the obstruction is cleared.

Sulaiman

Tue Jan 15 2013, 07:52PM

Registered Member #162 Joined: Mon Feb 13 2006, 10:25AM
Location: United Kingdom
Posts: 3140

I'm not concerned because the mode that I'm experimenting with is
back-scatter from the sky/clouds with significant transmit beam divergence
so no danger to anyone other than a careless operator (me).

klugesmith

Tue Jan 15 2013, 09:03PM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

My point was that if it's safe to stare into the IR source from up close, it would not be easy to collect substantial power from it far away. Perhaps OK with a wide but low-divergence beam. Transmitter could use a large-area collimating lens or mirror; receiver could use a large PV array or large, well-aimed optical concentrator.

How wide is wide enough for eye safety?
Let's play with radiometric units, starting with numerical values from the OP.
Suppose we want to get 1 watt of electricity from 20% efficient PV array spanning a whole square meter (or at the focus of a 1+ m^2 optical concentrator).
Irradiance at receiver must be 5 W/m^2. That's about 1/2 percent of noon sunlight. If all that power were pure white light, the illuminance would be 1250 lux. That's about 1.3% of noon sunlight, a very bright level for indoor task lighting.

A source 30 feet away must have radiant intensity of 418 watts/steradian (105,000 candlepower if pure white light).
If the source were 8 cm in diameter, then from 30 feet away it would match the apparent diameter of the sun. If uniformly bright, its radiance would be 84 kW/steradian/m^2 (200 times less than that of the sun's disk). If pure white light, the source luminance would be 21 million cd/m^2 (1.3% of sun).

But it's all 940 nm, invisible to the eye. Whether 0.5 or 1.3 percent of sun, I bet that's well over regulatory and common-sense limits, for places where people could look at the source with unprotected, dark-adapted eyes. Putting some very intense visible lights next to the IR source might reduce the practical (and regulatory) hazard.

Respectfully,
Rich

p.s. Yes, there are IRED arrays for IR "illumination" (strictly speaking, irradiation) of large areas. (Filtered incandescent lamps can do the same at lower cost.) But just as with visible light, the irradiance levels to support vision and photography don't support charging electric vehicles with PV arrays of practical size.

[edit] Care to share the diameter and beam angle (or datasheet) for your 3W IRED's? We can figure the viewing distances to match the irradiance and the source radiance in my example. Then you could contemplate looking into the beam, in the dark, at those distances.

Fiddy

Wed Jan 16 2013, 01:49AM

Registered Member #8817 Joined: Mon Dec 17 2012, 05:16AM
Location: Australia
Posts: 110

i have a 1.7W (3W diode) 808nm multimode handheld laser that would be better, it has horrible divergence for a diode laser but allot better than a led.

little video if divergence under NV:

Conundrum

Wed Jan 16 2013, 07:47AM

Registered Member #96 Joined: Thu Feb 09 2006, 05:37PM
Location: CI, Earth
Posts: 4061

120 degree angle without the separate lens,

It appears that as usual they have avoided regulatory hassle by selling the unit as a "general purpose security light source" so without the lens the beam should be harmless at about 4 feet based on divergence and inverse square law.

Shrad

Wed Jan 16 2013, 08:14AM

Registered Member #3215 Joined: Sun Sept 19 2010, 08:42PM
Location:
Posts: 780

there are 940nm laser diodes available

the best would be a fiber pigtailed 830nm laser diode as a fiber makes a gaussian beam with those single emitter laser diodes (plus they have integrated TEC)

Conundrum

Mon Jan 21 2013, 07:27AM

Registered Member #96 Joined: Thu Feb 09 2006, 05:37PM
Location: CI, Earth
Posts: 4061

Tried it with a 10W solar cell and an unknown wattage IR "security" panel, that has a few dud LEDs.

14V at 3.8mA, 3 feet away.

Either this security light is fried big time (can see some red glow but the LEDs are "patchy" ) or the solar cell isn't as sensitive to 780nm as 940nm.

1 2 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob