How does negative impedance cobverter enable impedance matching ?

There. See what I mean? The same vague question. We're certainly not going to give a full course on impedance matching here. A little searching on the web will get you plenty of that, anyway.

Now, what specific problem are you trying to solve, or what point are you missing?

Jeroen Belleman

There is a specific goup attempting to build better crystal receivers for LF/MF broadcasts. It seems to me that the OP in in this group. I have a feeling that their electronics is kind of similar to the most active audio enthusiasts with golden speaker cables.

--

-TV

All you need is a SiC mirror >:-} ...Jim Thompson

--
| James E.Thompson                                 |    mens     | 
| Analog Innovations                               |     et      | 
| Analog/Mixed-Signal ASIC's and Discrete Systems  |    manus    | 
| STV, Queen Creek, AZ 85142    Skype: skypeanalog |             | 
| Voice:(480)460-2350  Fax: Available upon request |  Brass Rat  | 
| E-mail Icon at http://www.analog-innovations.com |    1962     | 

     Thinking outside the box... producing elegant solutions, 
              by understanding what nature is hiding. 

"It is not in doing what you like, but in liking what you do that 
is the secret of happiness."  -James Barrie

If you are able to use active elements, perhaps you have to rethink the whole point of impedance matching.

Pere

Never was, and never will be an audio enthusiast. First learned about negative impedance converters in EE 411 class, about 16 years ago, at UT Austin. Recently, got hold of a recently declassified US Army study about impedance matching small antennas using negative impedance converters as as a key component. Lots of mathematical manipulations which hide the underlying physics. Hence my question. BTW, where do you get golden audio cables ? Who buys them ? I can't with my engineer's salary, but possibly some Arab oil shiekh, listening to music while deflowering the newest members of his harem ?

Here is a figure from a paper (authors and title seen on top) that sums up the principle:

There's plenty of information out there, it just takes some time and effort to sieve through it. If it's mostly behind paywalls for you, that's a different story.

Cheers, Nikolai

The left side of Fig. 1 is really silly. The slanting dotted line connecting the two curves is bizarre. The line should be vertical and at the point where XL = XC. In any practical circuit, the series resistance mainly in the inductor will prevent the net impedance from going to zero.

Electrically small antennas are usually tuned to resonance using a series loading coil. However, the radiation resistance, which indicates the power actually radiated, can be quite low.

The right side is equally bizarre. It is hard to find a negative capacitor that retains a constant value for all frequencies.

With blunders like these, it is hard to imagine the paper having much value.

ing

here

r r

ue.

Well, the figure is clearly just a simplified representation of the princip le. There's nothing like series resistance or radiation resistance consider ed.

r

Some other authors (Tade, Gardner, Hall) claim they have built one that's g ood up to 1.5 GHz. In the Sussman-Fort&Rudish paper they claim up to 20 dB improvement over 20-120 MHz for a receiver, and they provide the circuit (L invill NIC) in its gory details.

Without being an expert in the field, it doesn't look completely useless to me (-:

Cheers, Nikolai

At which point do you start to introduce reality? If you start mangled, it will just be more confusing later. Why not start accurate so your readers don't have to unlearn stuff later.

It would probably be better to start with the Radiotron Designer's Handbook from 1954:

The authors had no qualms about sticking to accurate details.

20dB improvement in what?

Can you provide links?

But how much of the information will you have to unlearn later?

The y-axis is reactance. I'm not sure about the RHS, I never made an NIC.

George H.

George Herold wrote:

This article is about antennas. The key problem is radiation resistance. You want to keep it as high as posible, which is not possible in short antennas.

Antennas are reciprocal. This means if an antenna is a poor radiator, it will also be a poor receiver.

Any paper about antennas that ignores radiation resistance is of little value.

There are a large number of papers on Non-Foster reactance. They are largely publish-or-perish papers where the author flings a bunch of equations around until he gets some kind of answer, then claims that is the result he was looking for.

The papers are mostly worthless, with some notable exceptions. The problem is anything you put between an antenna and a low noise receiver will degrade the noise figure, so you already have a poor performance antenna, and you make it worse. This is why these circuits are not used in commercial applications.

Here are some examples. Sorry for the wrap.

3/M07237477.pdf le.pdf format.pdf 77.pdf 73.pdf 5_Non- Foster_impedance_matching_of_an_electrically_small_loop_antenna_for_biomedi cal_telemetry/links/5a5ac5bd0f7e9b5fb388b9bf/Non-Foster-impedance-matching- of-an-electrically-small-loop-antenna-for-biomedical-telemetry.pdf? origin=publication_detail cal_telemetry &t=hy&ia=web t.pdf _Stephen_E_Sussman-Fort 25feb2013m_ppt.pdf engineer.aspx Converter-Paper

There are different pedagogical approaches. The OP already mentioned he got swamped with details. If you want to understand a basic technical principle it makes sense to strip it form the clutter, doesn't it?

Now, a discussion on the technical benefits, pros and cons of the principle is something different.

Thanks, that's a valuable reference. Also some classical texts by Fano and Bode for that matter.

SNR, presumably due to the better matching with respect a fully passive matching network.

Of course:

I really find that hard to believe. You have a low signal level from a small antenna, and you put a noisy circuit between it and the receiver. Noisy circuits do not improve the SNR. 20 dB is impossible. Someone is faking something.

Not all papers are honest and truthful. Use your common sense before you believe everything you read.

big snip, Ouch, sorry. I don't care too much about antennas at the moment, (well except for pick-up and the reciprocity theorem, as you say.) I was only pointing out that the graph didn't include resistance.

Are you one of those genius types who can't make mistakes? I'm just guessing, and sorry if I'm wrong, but I've had colleagues like that. pita, but can be godsends at times.

George H.

It also omits XL = XC.

No, just curious. I was wondering if this was some sort of miracle cure for short antennas. But the more I dig into it, the more flaws I find.

Measurement technique, test setup, instrumentation, interpretation and reporting are usually blatantly wrong. There are no real on-the-air comparisons of performance. And there are no references to actual use in commercial products.

None of the articles I found discuss the ESR of the negative capacitance generated. And you cannot treat an antenna as a fixed capacitance over a broad frequency range.

The HF bands are mostly dead since the internet took over. But back in the days, when Radio Moscow was active, if you tried to put an OPA657 at the input of a 40 meter receiver, you would be shot.

e

s r d

or

e

Perhaps you are right to be sceptical. The technique is not a silver bullet that solves all kinds of problems. But I have actually seen it in commerci al RFID tags. They often use very sub-optimal antennas, so anything that sq ueezes a little more range is welcome. Some tags have tunable matching, acc essible through a bunch of registers.

Actually, why do you assume that the NIC circuit should be noisy?

In that paper where the figure came from, they are considering the ESR, or rather the conductance and Q factor of the negative C. It doesn't look too terrible either. They don't go into details on noise, but they claim their transistor NIC circuit to be "low-noise", compared to op-amp solutions they 've tried. I guess that was not their main point to begin with, otherwise t hey wouldn't have used a receiver with 8dB NF LNA. Unless I'm also missing some detail in the picture...

Cheers, Nikolai

Again, do you have links?

Do you have any comparisons of the range between the bare antennas and the NIC versions? Or are the NIC versions simply marketing fluff to try to gain customers.

An obvious application would be using an automobile FM antenna on the AM broadcast band. If there was any advantange, there would be a plethora of cheap ICs available to do the job. There are none. Most RFID tags are passive. There is not enough energy received to power a NIC. If you cannot turn on the NIC, you cannot power the RFID. Chicken or egg problem.

Utterly obvious. You have a low level signal from a poor antenna. You put a regenerative amplifier between it and the receiver. Transistors generate noise. Regenerative circuits generate more noise. What else can you expect?

Look at the datasheets for low noise transistors, with noise figures below

1 dB. They will use some kind of passive matching network between the source and the transistor. None of them use a NIC. See the datasheet for the ATF-38143. This has a 0.4dB noise figure at 1.9 GHz. The matching circuit shown in Fig 5 is passive.

You cannot simulate a short antenna using a 50 ohm signal generator in series with a small capacitor. Free space impedance is 377 ohms, and short antennas are not simple capacitors over a broad frequency range.

None of the articles talk about radiation resistance. You want to keep this as high as possible. This is impossible with short antennas. So you have a poor performing antenna, and you put a noisy circuit between it and the receiver. You lose. This is why nobody uses it.

n e

in

a

a

t?

w

p t

is

a

Sorry, I can't provide any links or further info on those RFID tags. It was years ago and I actually signed a NDA. They were mass-produced though. And you can power a fair amount of circuitry even in a passive tag with the gi ven limit of 2W. Some of them had built-in sensors and whatnot.

With the risk of annoying everybody by quoting yet one more time the same p aper where the figure came from - they DO talk about radiation resistance. Particularly when they talk about transmitting.

Anyway, this thread has drifted very far from its origin. I'm off for now.

Cheers, Nikolai

See

IV. CONCLUSIONS

In this paper a critical investigation of the usability of negative impedance converters for antenna matching networks was made. As in a variety of other publications before the advantage of non-Foster elements for antenna matching regarding the signal level only seems to be obvious.

As those non-Foster elements are realized with transistorized negative impedance converters the noise behavior at the output of the matching network has to be taken into account under all circumstances. Therefore a noise analysis of the NIC-element was made and a system comparison between a NIC-matching network and a conventional active system with short monopole antenna for automotive application was performed.

With equal gain of both systems the NIC-system provides a significant higher output noise voltage resulting from its unfavorable noise impedance and provides a severe degradation of the SNR at the output. This shows that for the efficient design of small active antennas with non-Foster matching networks realized with NICs a careful noise investigation has always to be performed

The thread is NIC converters in impedance matching. They don't work.