Discussion in 'Electronic Design' started by anc, Sep 17, 2003.

1. ### ancGuest

I think there may be an error with the value of C1 in this circuit:-

The value is shown as .22uF, I did some calculations and believe it should
be .22pF

I estimated lead inductance around 7nH of 0.5 inch of lead for a capacitor
and worked out resonant frequency using 1/(2pi LC^0.5) estimated a
resonance at just over 4GHz.

I am not sure about the small signal equivalent circuit at these microwave
frequencies, whether each component lead will introduce inductance or the
effect on the input of the opamp.

Having said all this, I made the above circuit using a .22u (220n) capacitor
and nothing was received from any road camera in the UK, only an annoying
squawk from a mobile phone (960-980MHz).

If the microwave energy from a road radar camera is high enough, I would
imagine that even a "badly tuned" resonant circuit may recieve something.
Radar cameras in the UK use 10G and about 24GHz, has anyone else built or

My estimation of 7 nH arises from this equation using r =0.05 and l=1.2

Lac = 2L[ln(2L/r) - 1.00]

from the following page:

http://www.ee.scu.edu/eefac/healy/indwire.html

2. ### George R. GonzalezGuest

Er, Um, hmmm....

This doesnt look like any radar detector from this planet!

Radar detectors usually have some kind of energy collector,
either a waveguide, or a plastic prism, or both.

This circuit seems to depend on the microwave collection ability of a small
loop of wire, the leads of that .22 uf capacitor.

Now any loop is going to grab some signal from the ether, but this doesnt
look like a very well optimized collector....

Next glitch, there's no detector! Usually you'd expect some sort of
square-law device, such as a good old point-contact 1N21 diode, or newer
hot-carrier equivalent.

Instead this circuit seems to depend on the not too well documented
capabilities of a 741 op-amp input stage to demodulate microwave signals.

Just as a rough guess, this whole circuit is some 60db below optimum, I'd be
happy if it could detect a nearby cell-phone. Asking it to detect
microwaves from more than a few inches away is asking a bit much IMHO.

3. ### John LarkinGuest

This circuit makes no sense at all. A capacitor as a microwave
detector? Another example of why R-E magazine is dead.

I think they also did a similar circuit, using a capacitor as a
gravity-wave detector.

John

4. ### Scott StephensGuest

Of course it doesn't. Some piezoceramics, dielectrics, are lossy at
microwave frequencies and pyro-electric. The RF loss causes heating,
which causes a DC shift, which is detected by the crappy 'ol 741. So
that capacitor may be acting like a bolometer (?) or a common
passive-infrared sensor. A ceramic cap may do a better job of detecting
microwaves, especially if it happens to be bulk-resonant and lossy.

Or perhaps its the quantum super-symmetric pair breaking of spin
polarized gage invariant chiral Weyl-tensor asymmetries? Because its
supposed to detect gravity waves too.

Scott

--
**********************************

DIY Piezo-Gyro, PCB Drill Bot & More Soon!

http://home.comcast.net/~scottxs/

**********************************

5. ### Jim ThompsonGuest

[snip]
ROTFLMAO!

...Jim Thompson

6. ### John LarkinGuest

Which polarity?

John

7. ### Scott StephensGuest

I don't know. If it is a polarized electret, IIRC heat de-polarizes it.
In Sensor Magazine I remember seeing an article about a (IIRC) PVDF RF
detector, in which IR was flashed on it to reset it, like a
chopper-stabilized amp uses to measure tiny DC values. I didn't find it
in a quick search of sensormag web site.

The dielectric in effect acted like the permeable core in a flux-gate
magnetometer. But instead of the permeable ferro-magnetic material being
switched by a saturating magnetic flux, the ferro-electric material is
switched by an IR flux. That's what I remember anyways.

Scott

--
**********************************

DIY Piezo-Gyro, PCB Drill Bot & More Soon!

http://home.comcast.net/~scottxs/

**********************************

8. ### Scott StephensGuest

Stop it! I meant the Ricci tensor. Everyone knows the Ricci tensor is
about gravity. Ah hell, ya know what I mean, down cha?

Scott

--
**********************************

DIY Piezo-Gyro, PCB Drill Bot & More Soon!

http://home.comcast.net/~scottxs/

**********************************

9. ### N. ThorntonGuest

Hi. So you dont think it has the capacity to do that? Could I induce
you to use a coil for the app? I'd resist the idea of using a resistor
tho. As for using a transistor - that was just a transient idea.

Seriously tho, if speeders build non-working detectors then try them
out, they get caught - just what we want.

Regards, NT

10. ### ancGuest

I think the science of this circuit is good. A high strength microwave
field i.e. from a radar camera would induce a tiny current through the input
capacitor. That is the underlying principle.
My first tests, although grossly mistuned did respond to a mobile phone and
also around my microwave oven door seal. Whether this was just
electro-magnetic interference, I cannot say or prove for certain, but any
electromagnetic wave will induce a current into a straight piece of wire...
if the input circuit of the detector is broadly tuned to 10GHz, I would
expect a small current to be induced and amplified.
The circuit is not good in terms of RF design, but as a pure detector, I
believe this would work. My calaculations suggested the capacitor should be
in pF not uf though.

11. ### John ToddGuest

I doubt there is anything in the op amp itself that would respond
to a GHz signal. You may get accidental detection of a very strong local
signal (cellphone) from _any_ part of the circuit.
Paralleling C1 with a GHz-responding diode may get you started: the
op-amp will have DC input.

12. ### Dave VanHornGuest

i had severe problems on a project once, because the internal diodes in my
opamp were recitfying rf emitted by a nearby cordless phone transmitter.

what they hadn't told us, was that the transmitter used six watt pulses at
120 hz, of 920 mhz energy. those pulses ended up a a very loud audio buzz.

the phone also gulped current at 2a during the pulses, which was where i
originally thought the problem was.

no amount of shielding or bypassing helped, i even tried layered carbon
fiber/copper shields.

i ended up having to use a cmos, rail-to-rail, precision /aka slow as snail
snot in january/ opamp to get rid of the problem.

the antenna was very close to my circuit, but also i was trying very hard
/not/ to pick up the signals. the board was planed, all signal tracks were
as low impedance as practical, and absolutely as short as possible, shielded
cable, etc etc.

13. ### ancGuest

The opamp itself never works at microwave frequencies, it doesn't have to.
It only amplifies the "current pulse" induced by received microwave energy
on C1. It is almost the equivalent of a crystal radio with a short antenna,
except that the tuned circuit is a sole capacitor, with leads creating the
inductance. I don't know if the tuned circuit response will be wide or
narrow, or indeed the power of the signal from the radar cameras, but your
suggestion is a good one and I might try it out.

14. ### Walter HarleyGuest

That "current pulse" is AC, right? That means that for a fraction of a ns
it's a current pulse in one direction, and then for equally long it's a
current pulse in the other direction. The two average out to zero. And
they average out a lot faster than any component in a 741 can respond - the
diodes and transistors in that 741 aren't even fast enough to rectify the AC
so as to get a DC shift. (Put differently: take a look at the input
capacitance of the 741; now compare it to your desired sub-pF detector
capacitor.) Basically, the 741 is pretty much transparent to microwaves,
I'm thinking.

And by the way, you mentioned a "high strength microwave field i.e. from a
radar camera." Umm, I don't know the specifics but I'm thinking that if
they're aiming that thing at me, it's not very high energy. And if you want
to detect it from far enough away that you can actually respond with your
brake pedal before it's too late - that is, from at least a few hundred
yards - then it's very weak indeed.

Consider this: It is in the cop's interest for the radar signal to be as
weak as possible while still permitting detection - after all, they don't
want to blanket the area with radar signals that might give them away. The
radar camera has high-quality, purpose-built, sophisticated detectors. So,
the radar camera can put out a signal that is just barely strong enough for
its very good detectors to detect a bounce off your big metal object. Now,
you want to detect that signal from about the same distance (the cop has to
detect the bounce, so double the distance; but you want to detect far enough
in advance to slam on the brakes; so let's call it even). So it's a pretty
safe bet that your detector needs to be about as good as the cop's, to
compete.

15. ### George R. GonzalezGuest

Er, no, the bubble-gum they suggest for mounting the parts has been shown to
have
non-unitary fugacity in it's chromodynamic Hamiltonian.

16. ### Chris1Guest

I remember building this "gravity wave" detector back in '86 when I got
that magazine. The radar detector was the same, but with a smaller
capacitor. Neither circuit did anything, as far as I could tell. I'm
surprised someone bothered to post it on a web page.

Chris

17. ### Ben PopeGuest

Really? That surprises you?

But people will stick ANYTHING on a web page!

Ben

18. ### Ben PopeGuest

I hear those in the know use Nokia mobiles

They only do it once or twice though.

Ben

19. ### Walter HarleyGuest

Yep. Way too #\$%#\$ many circuits out there (many of them actually published
in books and datasheets, not just on web pages!) that could never have
worked. I love the "bad circuits" sections of AoE, drawn from real life
examples of not-a-chance circuits.

Too many people don't seem to understand that wires have nonzero resistance,
a 4.7k resistor might be 4.6k, a 9v battery is not a precision voltage
reference, and an opamp is only *close* to being an ideal component. They
think that if a schematic works in SPICE then it'll work on a breadboard.
Or, it's just "so obvious that there's no point in testing it."

Useful advice I learned in engineering school: "constants aren't; variables
won't."

20. ### Ben PopeGuest

Yeah, it's all very nice considering that an op-amp has "infinite input
resistance" and for many circuits you can consider this true - but knowing
the conditions under which a model is relevant is probably more important
than knowing the model.

Many people don't understand the term infinity I guess. It's not "the
biggest number you can think of... plus one"

I think it's probably more like: "A number so big that increasing it results
in no significant change". Pick that apart as you please
I like that.

Ben