How to drive small inductive loads up to RF?

[horror-vacui]

Hi Everybody,

Could you help me giving some keywords, tips and hints for a measurement equipment to drive a wide range of slightly (=not a motor) inductive load with a sine wave for a wide range of frequencies?

I wish to investigate wireless power transfer with a small integrated RX coil. For energy harvesting a given voltage level is needed in the RX coil, which can be increased by increasing the frequency. I aim for a tiny RX coil, with even just sub-mW transferred power. A higher than typical frequency will help to increase the received voltage.

What affordable test equipment could drive a TX coil in a wide freq range from sub-MHz up to 100s of MHz? Waveform generators are not that fast, and I need only a sinewave (some AM modulation would be nice though, but it is not required). RF signal generators tends to be pricey, and provide more than I need, but in features, and in precision as well. I though of some SDR solutions with an additional power amplifier, but the wide range of non-resistive load values are usually disliked by RF power amplifiers. There could be either high current or high voltage levels at the TX coil depending on the frequency and the coil. Are you aware of a class of amplifiers which can deal with these requirements? I would be happy with two instruments for different frequency ranges as well. In inductively coupled wireless power transfer the current drive is more important to build up the magnetic field, but typical instruments have a limited current driving capability related to the often assumed 50 ohm load.

I am happy to build something by myself, but I would rather trade the time of designing and debugging for the cost of an instrument. Nevertheless the cost of the instrument should be also affordable: I am sure, I can not consider anything above $1000€ at the moment. Do you have any ideas or suggestions, which path to take? Any keywords I should search for?

[Prying]

while im not sure what test equipment is avalible for this but cant this be done with a software defined radio IC/package?

[TimFox]

Have you looked for used "RF Signal Generators" from eBay, etc?
They usually have an internal 50 ohm source resistor, and will tolerate almost any (passive) load after the output connector.

[jonpaul]

Bonjour: Nikola TESLA (NOT the car!) investigated this at Colorado Springs in 1880s, with a 50 KW several megavolt tower at ~ 20 kHz resonance.

The physics of wireless power transfer and aircore transformer coupling is very well studied and documented.

The magnetic field falls off so rapidly that the efficiency is extremely low unless the coils are next to each other.

Just take any audio oscillator or function gen and wind a few coils.

Cost should be ~ $/EU 20..50 not 1000!


Bon Chance


Jon

[David Hess]

Could you help me giving some keywords, tips and hints for a measurement equipment to drive a wide range of slightly (=not a motor) inductive load with a sine wave for a wide range of frequencies?

I wish to investigate wireless power transfer with a small integrated RX coil. For energy harvesting a given voltage level is needed in the RX coil, which can be increased by increasing the frequency. I aim for a tiny RX coil, with even just sub-mW transferred power. A higher than typical frequency will help to increase the received voltage.

Wireless power transfer coils are resonated with a capacitor to achieve the needed circulating current and strong magnetic field for longer range.  They are driven close to their resonant frequency by a switching instead of linear stage.  A resonant switching regulator controller could be used and these are the types of circuit which should be studied.

For experimentation purposes at lower power, I might use a transconductance (current) output stage which will have a big advantage in speed for higher frequencies, but that is not how real inductive power transfer systems work.  In practice though a 50 ohm series terminated current feedback amplifier will be just as good and easier to implement.  There are some high power ones intended for DSL applications available.

[horror-vacui]

while im not sure what test equipment is avalible for this but cant this be done with a software defined radio IC/package?

Radios have limited output power, usually in the range of 1-1x mW, and they output amplifier assumes a 50 ohm load. This was one of my ideas, but then I would need boradband power amplifiers which could drive up to an ampere at low frequency, and they are still functional at 100s of MHz. The problem is that in the 100MHz range the parasitic capacitances will reduce the signal. That's why all RF stuff have matching network between stages. Such matching networks will rely on resonance and thus they are narrowband. Transformer coupling could alleviate this, but 1) the IC and PCB parasitics will possibly still limit the bandwidth, and 2) low-power signal transformers with a mention-worthy turn ratio for impedance matching are only available well below 100MHz, and well, they are low power 3) and now I've just realized, that transformer coupling still won't provide broadband matching, because the load is frequency dependent...

[horror-vacui]

Have you looked for used "RF Signal Generators" from eBay, etc?
They usually have an internal 50 ohm source resistor, and will tolerate almost any (passive) load after the output connector.

Yes, I did, but they are not fast enough, and their output current capability is reduced. The latter means, that they are optimized for 50ohm. Siglent 2000x wavegens had maximum output voltage on 50Ohm of +-5V and +-200mA, while they will go only up to 120MHz. I wish to go higher both in volts, amps and hertzes... The signal generators I found on ebay are also slower than my initial spec, and I do not think the cheaper ones could provide the required output current.

[Bud]

I do not think what you want is feasible. Especially given the "tiny coil" requirement.

[geggi1]

A former colleague of did some testing with a audio amplifier.
I believe it was class D or E amp.
The amplifiers are pretty robuste and wold most likely be able to handle your tests as long as you dont drive it to hard.

[horror-vacui]

The physics of wireless power transfer and aircore transformer coupling is very well studied and documented.

Indeed, but all calculations use serious approximations. I was to check their validity for my use case.
I probably will not contribute to the science, but I hope to verify calculations, provide a fast confirmations of the validity of my design ideas and also give me more confidence than hand calculation or some EM simulations.

The magnetic field falls off so rapidly that the efficiency is extremely low unless the coils are next to each other.

For a circular conductor it can be calculated that it falls off with the third power of the distance along the axis of the conductor after the radius of the conductor. It also indicates that for big circles, and small distances, the efficiency is constant (if leakage inductances further from the axis are neglected)

Just take any audio oscillator or function gen and wind a few coils.

Cost should be ~ $/EU 20..50 not 1000!

There are unfortunately no instruments below 100€ at the moment, the wave gens have a current driving limit, and often also a voltage limit. I am fortunate enough to be able to spend a few hundreds for an instrument, if I can reuse it later.

[horror-vacui]

A former colleague of did some testing with a audio amplifier.
I believe it was class D or E amp.
The amplifiers are pretty robuste and wold most likely be able to handle your tests as long as you dont drive it to hard.

That's true, but I am sure that I would need a higher frequency, than the audio range.

[dmendesf]

You need a power RF amplifier. You can test some used for cable TV: BGY888 for main frequencies (50-900MHz) or R2005240P12 for reverse frequencies (5-45MHz). There are others.

[horror-vacui]

I do not think what you want is feasible. Especially given the "tiny coil" requirement.

That's why I want to check it out. My calculations indicate that it could work, now I just want to test those calculations. The received voltage is proportional to the "effective area" x "frequency" product, so there is a chance for power transfer with a small coil.
I am not sure how well it works, but EM micro announced an passive UHF RFID tag chip with integrated coil packaged into a 1.5 x 1.5 mm chip. That is tiny as well! They claim 10mm readout distance.

[T3sl4co1l]

Easy.

Don't.

Resonate with a cap. That's what everyone does, and with very good reason!

The fundamental problem is the very poor coupling between coils.  Which means reactance dominates over real power transfer, by a factor (Q factor) of... 2, 5, 10 or more -- roughly, the inverse of the coupling factor.

With judicious use of ferrites (pole pieces and shielding), you can get a Q factor under 1, but only when things are very closely spaced, and the alignment must be pretty close as well.  Basically, you get something like a split-bobbin transformer in that case, and any air gap or misalignment counts as air gap between windings -- at expense of coupling.  Works for something like a cordless phone or toothbrush in a cradle, not so much for random sized items on a charging pad.

Resonance reduces the control problem to a matter of frequency, amplitude and/or phase control.  Whereas direct drive simply draws whatever current the reactance does (again, because reactance is dominant), resonant can draw much more power (from the inverter), because load resistance acts to dampen resonance, and its absence therefore requires a control strategy.

In exchange, you have a lot of time to solve the control problem: rather than the per-cycle control that a lot of switchmode controllers employ, you can spend several cycles working out what frequency to choose, or amplitude to modulate to, and merely a peak current fault (latched) will do to protect the inverter.  This is because bandwidth is also determined by Q factor, and a high Q means low bandwidth, means current changes less, from cycle to cycle.

Tim

[horror-vacui]

Resonate with a cap. That's what everyone does, and with very good reason!

That will be the plan for the project, but first I wish to do some experiments. The resonance can be used to increase the current in the TX coil, which in turn increases the magnetic field, for the same voltage drive (equivalent with reducing the impedance), but it generates the same magnetic field as a non-resonant coil with the same current.

While it is great to have a less reactive load, it "only" makes the life of the driving source harder. Of course if there is nothing out there, I will use different caps for every frequency in my measurements.

The fundamental problem is the very poor coupling between coils.

In itself it is not a real problem for me. Efficiency is not one of my concerns. The only issue I see, is that the inductive load makes it harder to drive.

any air gap or misalignment counts as air gap between windings -- at expense of coupling. 

I am aware of this, and this is also something I would like to have first hand experience with.

[T3sl4co1l]

Whelp... doing it at literally every frequency, you're going to need a highly capable amplifier.  Not a driver, there's no such thing as an inverter over that wide a range.

And at that, one that can handle severe mismatch, since you're very specifically and intentionally running an unmatched load: mostly reactive.

Not sure what you're going to accomplish with <mW; coil losses and diode drops may consume that already.  If you need a few watts to build something more representative, expect a 20-100W amp -- no small expenditure.  Plus the signal generator to run it, and maybe a preamp.

The more reasonable approach is to merely notice impedance is proportional to frequency, and then do the experiment at just one frequency and scale from there accordingly.  Accounting for stray effects as needed.

Also not sure about the "energy harvesting" angle; there's very little with any kind of field strength around it (and with good reason!).  If you're not intentionally crafting a transmitter pad sort of thing to go with it (or using any of the available ones), your receiver is going to be quite energy-starved no matter how un-tiny its coil is.  I'm probably misunderstanding your intent here, so if you could explain things in a bit more detail that may be helpful.

Tim

[nctnico]

I'd start by researching which frequency bands are actually allowed to use magnetic coupling. Or put differently: which frequency bands are being used for NFC and what are their emission limits. At higher frequencies you'll likely run into issues with skin effect and losses in the magnetic core material which will limit the amount of power you can transfer. Personally I would concentrate on using frequencies between 100kHz and 300kHz.