RF energy harvesting project

Are you measuring that 3.4v across a 50 ohm resistor that is terminating the output of your antenna? If you are, you have harvested almost 1/4 watt. Hmm, UHF, what are you measuring your voltage with? Post your schematic somewhere. Mike

Hopefully some of your project conclusions will include the fact that such energy harvesting has rather limited utility in the real world due to the

*incredibly* low efficiency with which it can be performed.

But that doesn't mean you can't have a lot of fun while you're at it. :-)

Are you measuring the 3.4V with some high-impedance device? (E.g., FET probe on a scope or spectrum analyzer?) In that case you're correct that the problem is that 50ohms is significantly loading the antenna, and if you're only getting ~500mV you are losing a lot of power in your Schottky diodes.

As Mike suggested, please post a (link to a) schematic.

In the meanwhile, you might be encouraged to know that that are various companies that manage to stay in business producing energy harvesting receivers. Check out these guys:

-- one of the founders was doing similar experiments to what you are now (albeit at 915MHz) back in 2004 and managed to make a master's level thesis out of it (which truly amazes me, but hey, more power to him): .

Jim Thompson: If you're reading this, check out that thesis and tell me how many chortles you get out of it -- especially the notes about SPICE around pages 24/25: "This software is too limited for this project." Um hmm...!

---Joel

The diode is in *parallel* with the antenna?

First, delete the parallel diode and use one series diode. Right now, you appear to be shorting out every other half cycle.

--Winston

[snip]

Almost didn't. It's the sort of title that normally gets deleted automatically. I just happened to see my name :-)

His "Master's" circuit actually bears some resemblance to circuits I've used in RFID tag chips... Schottky's with CMOS devices used as capacitors... but with active behavior (shunt regulation)... powered by a STRONG RF reader maybe 30' away MAX ;-)

PSpice simulates it just fine.

I guess I know what university to NOT recommend :-( ...Jim Thompson

--
| James E.Thompson, CTO                            |    mens     |
| Analog Innovations, Inc.                         |     et      |
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    |
| Phoenix, Arizona  85048    Skype: Contacts Only  |             |
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  |
| E-mail Icon at http://www.analog-innovations.com |    1962     |
             
Obama: A reincarnation of Nixon, narcissistically posing in
       politically-correct black-face, but with fewer scruples.

Post a schematic?

How do you know the antenna is 50 ohms? What sort of antenna is it?

Why don't you start by matching impedances, then? Better, since the Schottkey diode is nonlinear, why don't you try various load resistances experimentally until you find the one that gives the most power out of the antenna.

Because you're going to lose a bunch of voltage at the diode, you may want to use an antenna with a higher impedance -- look up "folded dipole". Or consider a step-up transformer into the diode.

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" was written for you.
See details at http://www.wescottdesign.com/actfes/actfes.html

You might look into some of the high end crystal radios. just replace the crystal with your diode and the earphones with your load. Here's so efficiency improvements,

But I think this is starting at the end rather than the beginning. Ben did some good studies!

-- MikeK

I was impressed with the number of crystal radio sites out there.

Just two, of the dozens....

The workmanship on a lot of these is excellent:

mike

A simple RF probe typically consists of an input capacitor connected to a diode cathode (anode grounded) feeding a mA or uA meter. Some RC filtering between the diode and the meter may help in preventing spurious radiation from the meter leads.

Try a transformer: hi Z input (from antenna) to lo Z output (load).

rahulponna had written this in response to

: Hi Mike,

I am a Master's student and as a part of my final year project, I am working on the same thing as you, viz. RF energy harvesting and using the harvested energy to drive a wireless sensor network. I just started working on this project, I am still in the literature survey part. I wondering if you could share any good links/documents you would have come across. I have lot of doubts as to how to go about it myself. Have just some untested ideas. If you can share any material you found really useful/insightful it would be really great.

Thanks & Regards, Rahul

MikeF wrote:

-------------------------------------

##-----------------------------------------------## Delivered via

Electronics Enthusiasts' Community of the Net Web and RSS access to your favorite newsgroup - sci.electronics.design - 397015 messages and counting! ##-----------------------------------------------##

Use back diodes.

What kind of antenna did that transmitter use ? Radiation pattern ?

What did you use as the pick-up antenna ? Radiation pattern ?

At what distance and orientation from the transmit antenna ?

What kind of load impedance did you use to measure that 3.4 Vpp ?

Was the measurement made in the Tx antenna near or far field ?

While in the far field, the free space impedance is assumed to be 377 ohms (120 pi ohms) and antennas are generally designed to work correctly in that environment.

However, the assumption of 377 ohm free space impedance is not valid in the near field, thus, an extra antenna designed to have a 50 ohm impedance and hence expecting a 50 ohm load, might not work well in the near field. For instance the impedance of a Yagi antenna changes radically when the dimensions and distances of parasitic elements are changed.

3.4 Vpp =1.7 Vp= 1.2 Vrms.

If this was a high impedance measurement, so in the best case after a _half_ wave rectifier, only 1.7 V would be available. For the maximum power transfer theorem with equal load and source impedances, 0.85 V would be available. However, in the loaded condition, the rectified voltage would not reach full peak voltage all the time, so about half of the open circuit Rms voltage would be more reasonable or 0.6 V. If you got 0.5 Vdc, this is in fact a quite good value.

A nominally "50 ohm" antenna placed in the near field of a transmitter antenna might not give the best power transfer when loaded with 50 ohms.