Hi,

Firstly the reason I have the constraints of not using diodes is that the AC input voltage is very low (200mV) and even low voltage Schottkey diodes have a forward drop of (230mV). And regarding external power supply, this circuit is aimed at energy harvesting, so the aim is to do everything with only the harvested energy.

Any ideas?

 

Not really, if you could turn on/off a FET or MOSFET at the right times you might get something, but then the pesky problem of how to power the transistors comes up. They require some juice after all.

If you are trying to harvest radio signals, good luck. There just isn't that much power available, and the faster you have to switch electronic switches such as transistors the more power they consume.

 

Yes I am looking at harvesting energy from radio signals, specifically cell phone band (950MHz).
Yeah with MOSFET the two main issues are how to control them and also switching them at such high frequencies and the losses.
Apparently a company called "powercast" has managed to do it.

 

You can use a power p-MOSFET as a diode.

Essentially the gate is connected to ground, and this puts the p-MOSFET into conduction, where it drops only a few mV. When the body diode conducts, it initially drops 700mV, but then the p-MOSFET conducts, dropping only a few mV due to the Rds(on).

Here's the application note:
http://www.maxim-ic.com/app-notes/index.mvp/id/2012

Don't know if it will work; I am wondering if the circuit requires bootstrapping (i.e. voltage higher than 700mV.)

Oh, and you could consider using a voltage multiplier, though again this depends on a diode, so I am wondering if this will work.

How does the 200mV/400mV AC range work on most multimeters? Perhaps that's a good place to start.

 

Electronic controlled switches for rectification are called synchronous rectificatiers - the technique works very well for very low loss rectification of substantial amounts of power where the tradeoff to generate the synch signal and the gate drive is reasonable.

Maybe you should firstly understand the energy contained in radio signals before getting your hopes up. uV multiplied by uA using ends up with pW.

The only time I've heard of people extracting power from electromagnetic fields is when they put a loop under a HV line (and got caught), oh and the guy who opened up a metallic tape measure right next to an inductrial RF induction heater (and got scortched). There have been many who researched the topic on government funding - one I recall harvested solar power in space and then aimed to microwave it down to huge arrays of antennas on the ground, and they needed hf rectification.

Ciao, Tim

 

We've had to disappoint folks before on this subject, and don't let commercials be your guide, they lie in many cases.

Radio energy, unless you are right next to the transmitter, is usually in microwatts or less (much less). There isn't any power to harvest.

 

Who said he needed "high power"?
Some wireless sensors can operate with as little as 2 microwatts
http://www.eetimes.com/electrical-e...ults-with-two-Wireless-Power-Transfer-Systems

 

Who said he needed "high power"?
Some wireless sensors can operate with as little as 2 microwatts
http://www.eetimes.com/electrical-e...ults-with-two-Wireless-Power-Transfer-Systems

He did, reread post #3.

You want to sense RF energy that is a receiver. Harvesting is something else. It implies getting enough energy for work.

I can receive with crystal radio, I can drive a sensitive earpiece with it. There are some nifty non electronic amplifiers out there using flames (for example). All amplifiers require an external energy source however.

RFID would qualify. The antenna and coupling is the thing, RF is measured in watts, if you are getting such low voltage you need a different antenna system and front end (high impedance).

As I've said though, this has not been the first thread on the subject, nor the second. They left unsatisfied, I suspect a lot of the technology is very proprietary.

 

What about stepping the 200mV up with a transformer then rectifying that?

 

Harvesting is something else. It implies getting enough energy for work.

Running a microcontroller at 1 uW can be considered work. Some applications are really low-power and if they get environmental power at 10 uW and run, I consider that to be energy harvesting. This fall, there is an "Energy harvesting and wireless sensor networks" conference here in Boston, and many of the applications are extremely low power sensor nodes.

I've been working for the last couple years harvesting energy from mud, produced by microbes. It may be 100 uW or less in some cases, but it can be useful.

 

Rahup, don't be fooled by the datasheets for the diodes.

Some Schottky diodes can rectify some current at far lower than their Vf on the datasheet. For example, a BAT41 diode can pass 1 uA at 0.1V. I've tested that out. That may help you. Also, there are some diodes that have essentially zero voltage drop. The tradeoff is they are very low power and very leaky. But it may work for your purpose.

If you're relying on low forward voltage, you may also see great variability according to temperature. When hotter, the diode will have lower Vf. That's something to look out for.

 

You might also be interested in rectennas and electron ballistics -- using shapes to rectify power, instead of semiconductor diodes. For example, check out the pictures on this page.

And PowerCast *can* gain enough energy to light an LED, but they are broadcasting their own RF source of power. Maybe 5 or 10 feet from the source, they can light an LED, but further it becomes less power.

Are you interested in harvesting environmental radio waves, or radio waves from your own source?

 

@sage : Yes the ballistic rectifiers you pointed out is the ideal solution I am looking at. But there is no way I can get them, can I?
I am interested in harvesting radio waves beamed by the cell phone towers.
I am in fact trying to test the characteristics of the diode HSMS-285C, by Avago technologies. It is mentioned to be useful for energy harvesting applications. BAT41 looks like a high power diode, innit? but your point is that it is quite leaky at low power levels and that could be advantageous in my case. Will have to look further into that. Do you know of any other very low power zero bias diodes?
Yes, you are right, even few uW of power harvested can be useful. Also, if you do attend the conference, keep me posted on any cool stuff. Thanks for your help and time btw.

 

Hey Sage.. do have any more inputs you have for the things I asked above?

 

Hey Sage.. do have any more inputs you have for the things I asked above?

Hi again,

Yes, my results on the BAT41 diode are from my testing of forward current, in which I found 1 uA passing at 0.1V forward. I don't think that is leakage, but rather it's that a small amount can pass forward at lower voltages that the specified Vf. I don't recall leakage current on the BAT41.

I also did find several other "zero-bias" diodes and ordered a few, some months ago, but I was not able to find the part numbers so easily. But here is one:
http://www.skyworksinc.com/uploads/documents/AD_DiodesSiliconSchottkyHighPerformance.pdf