EEVblog Electronics Community Forum
EEVblog => EEVblog Specific => Topic started by: EEVblog on July 30, 2014, 11:01:51 pm
-
Did you know you can use your frequency counter to detect gravity? You've likely done it before and you didn't even know it!
Dave demonstrates the phenomenon of 2g-tipover on quartz crystal oscillators in an Agilent 53131A frequency counter.
Related videos:
How a rubidium frequency standard works: https://www.youtube.com/watch?v=I55uLRRvLCU (https://www.youtube.com/watch?v=I55uLRRvLCU)
FE-5680A Rubidium Standard Teardown: https://www.youtube.com/watch?v=FRdGsSu5Nec (https://www.youtube.com/watch?v=FRdGsSu5Nec)
EEVblog #646 - Gravity Detection Using A Frequency Counter! (https://www.youtube.com/watch?v=zILwgQhjC_Q#ws)
-
gravity force stress/strain does cause some effects - for General Relativity effects at human motion scale you need lots more digits than cheap Rubidium
-
you can pretty much use any piece of test equipment to detect gravity.
stand up on a surface. hold instrument in both hands with arms stretched in front of you.
let go of the instrument
if there is gravity the instrument will be attracted by it and move towards the origin point
no electricity needed.
observation : the more expensive or fragile the equipment is , the faster it reacts ...
-
Measuring gravity using 6 cesium standards and a minivan: http://leapsecond.com/great2005/tour/ (http://leapsecond.com/great2005/tour/)
-
yes I was thinking more like arms length muscle powered motion
there's the round the world flight test that has calculable GR effect http://en.wikipedia.org/wiki/Hafele%E2%80%93Keating_experiment (http://en.wikipedia.org/wiki/Hafele%E2%80%93Keating_experiment)
and the past few decades have seen custom lab clocks reduce the needed distance changes to see GR
Optical Clocks and Relativity
Observers in relative motion or at different gravitational potentials measure disparate clock rates. These predictions of relativity have previously been observed with atomic clocks at high velocities and with large changes in elevation. We observed time dilation from relative speeds of less than 10 meters per second by comparing two optical atomic clocks connected by a 75-meter length of optical fiber. We can now also detect time dilation due to a change in height near Earth’s surface of less than 1 meter. This technique may be extended to the field of geodesy, with applications in geophysics and hydrology as well as in space-based tests of fundamental physics.
Science 24 September 2010:
Vol. 329 no. 5999 pp. 1630-1633
DOI:10.1126/science.1192720
-
there's the round the world flight test that has calculable GR effect http://en.wikipedia.org/wiki/Hafele%E2%80%93Keating_experiment (http://en.wikipedia.org/wiki/Hafele%E2%80%93Keating_experiment)
That photo in there is mine, I have done a video on that bit of kit.
-
Fascinating - I never thought about gravity's affect on a crystal - very interesting!
-
I think that same physics is behind some of the gyroscopes used from 1960 to 1980s (before military and space switched to fiber optics gyros). I have seen at least one so called Cylindrical Resonator Gyroscopes which is according to Wikipedia article on Vibrating Structure Gyroscope used in space crafts. http://en.wikipedia.org/wiki/Vibrating_structure_gyroscope (http://en.wikipedia.org/wiki/Vibrating_structure_gyroscope). I think these were used in submarines in 80s for dead reckoning as well.
-
Dave,
ever thought about a video about basic/budget/secondhand frequency generators in real world applications? What gear to choose, and the pitfalls you can encounter? What options do you really need?
-
That's interesting, didn't expect that gravity changes the frequency. But I wonder how the frequency counter measures mHz. Looks like in the video the gate time is one second, so this would be 10e6 cycles per second for 10 MHz. Even 100 mHz more would be still 10e6 (measured) cycles per second. Does it measure fractional cycles within the gate time, too?
-
I have no idea how this unit works, but there is a simple way using combined period/cycle counting:
The start of the measurement is synchronized with the signal edge. Then the frequency is counted for the the given gate time. In parallel a faster internal clock is counted. When the gate time is over the measurement continues until the next edge of the input signal. To get the final frequency value you have to divide the number of counted cycles by the measured period.
Lets say the internal clock is 100MHz and the gate time is 1s. This gives a resolution of about 8 digits. The resolution increases with internal time measurement accuracy so it can be increased without using longer gate times.
-
Not that it would be any better than always keeping your test gear the right way up, but just considering the theory for laughs -- could this problem be ameliorated to any extent by connecting two crystals in electrical parallel and physically mounted in opposite positions?
-
Lets say the internal clock is 100MHz and the gate time is 1s. This gives a resolution of about 8 digits. The resolution increases with internal time measurement accuracy so it can be increased without using longer gate times.
Thanks, this makes sense. So for mHz resolution for a 10 MHz input signal, the internal clock has to run in the GHz range? I guess jitter etc. is really challenging for such a stable result as seen in the video.
-
There are other ways of high resolution time measurement without the need of super high clock signal:
http://en.wikipedia.org/wiki/Time-to-digital_converter (http://en.wikipedia.org/wiki/Time-to-digital_converter)
Jitter on the input frequency is basically the limitation of the achievable resolution for a given gate time.
-
That's interesting, didn't expect that gravity changes the frequency. But I wonder how the frequency counter measures mHz. Looks like in the video the gate time is one second, so this would be 10e6 cycles per second for 10 MHz. Even 100 mHz more would be still 10e6 (measured) cycles per second. Does it measure fractional cycles within the gate time, too?
The HP counters are reciprocal counters, therefore the basic resolution is constant for all frequencies, depending only on Gate Time and XTAL frequency.
That would be 1e-7 per second.
Additionally, these counters use time interpolation methods, so that the original 100ns is further interpolated by analogue or digital techniques.
This counter series allows 100ps interpolation, that means 1e10 counts per second gate time.
Also, they implement continuous counting and time stamping, which may further increase resolution.
The famous HP 5370B used a digital interpolation scheme and achieved 20ps T.I. resolution, i.e. nearly 1e11 counts per second.
Frank
PS: uploaded AN-200, as maybe by tomorrow, it will not be available @ agilent any more.
-
Given the effect that we have observed in this video a certain degree of care needs to be taken when considering PCB orientation. It would for example be a bad idea if a frequency counter needed to be on one side when adjusting the calibration.
Now, if anybody has some material on high resolution counters I might have some reading for the weekend. I already know about ordinary and reciprocal counters but once these get beyond a ten second gate things start to get inconvenient.
-
@German_EE:
Handbücher und Application Notes der üblichen Verdächtigen (HP, Agilent, ...), Time-nuts Maillist - sollte genug Zeit wegmachen ! ;)
-
Just to be clear on this, reciprocal counters return the same number of digits for a given gate time. Prescaling the input does not reduce the number of digits returned. On many the gate time may be set to an arbitrary period.
Interpolating counters return extra digits beyond that which the clock alone will provide. There are all kinds of methods to achieve this. Digital storage oscilloscopes which support equivalent time sampling do the same thing by interpolating the time between the trigger and sample clock.
This HP counter does both. An HP5314 does neither. An HP5315 or HP5316 is a reciprocal counter without interpolation. Many or most later HP counters do both.
-
Why did Dave feel the need to remove the cover from the Agilent frequency counter? Anything to increase the number of viewers. I lust after his equipment.
-
Why did Dave feel the need to remove the cover from the Agilent frequency counter? Anything to increase the number of viewers.
You'll see in the next video. I originally did this as an aside in part of this next video, but thought it would be interesting enough to warrant it's own video, so released it first.
-
PS: uploaded AN-200, as maybe by tomorrow, it will not be available @ agilent any more.
Nice, thanks. Maybe discussing the details how such advanced counters work would be a good idea for a "fundamental friday".
-
They also respond to pressure. Can be used as a vacuum gauge and of course are used bare in semiconductor fabs as a method for determining how much material has been deposited onto a chip, or how much you have grown onto the surface.
I do not have any method to measure the change easily, though I probably can take a crystal and mount it into a housing with a screw on lid along with a schraeder fitting to expose it to vacuum and pressure with assorted gases. Just have to make the single feedthrough needed for the one lead of the crystal, and ground the other to the housing.
-
Is there a way to 'counteract' the effect of gravity/acceleration? Like using a second crystal oscillator placed 'upside down' respect to the first one, so that the two variations cancel out?
-
"Is there a way to 'counteract' the effect of gravity/acceleration? Like using a second crystal oscillator placed 'upside down' respect to the first one, so that the two variations cancel out?"
Good question (although maybe I don't understand the physics). Isn't turning the crystal upside down (180 degree change) the same as switching the leads? So, if you had an oscillator that would periodically switch the crystal leads and average the results, wouldn't that solve the 'problem'? It wouldn't fix the 90 degree change, though.
I'm fascinated by this, but I will readily admit that it is like arguing how many angels will fit on the head of a pin.
-
Part of the change do to acceleration has to do with the crystal's mounting arrangements and how they apply stress to the crystal. Some designs are better than others and this is another specification included with high precision crystals.
-
Yep, here I found one of several papers explaining the gravitational effect:
http://www.ieee-uffc.org/frequency-control/learning/filler_paper.html (http://www.ieee-uffc.org/frequency-control/learning/filler_paper.html)
It confirms, that this effect is caused by interaction between the crystal plate and the mounting.
Frank
-
I think that 1 bpm change is not really significant to talk about, while we have thermal EMF, aging, temperature coefficient, mechanical stress, michrophonic effects etc... on ther components which are several order of magnitudes bigger than this.
Also, I never really understood why we can get crystals in hermetic packages for 0,1 USD, while a resistor or a reference voltage is 10+.
-
I think that 1 bpm change is not really significant to talk about, while we have thermal EMF, aging, temperature coefficient, mechanical stress, michrophonic effects etc... on ther components which are several order of magnitudes bigger than this.
I think you mean ppb (parts per billion), not bpm (beats per minute). :-)
Also, I never really understood why we can get crystals in hermetic packages for 0,1 USD, while a resistor or a reference voltage is 10+.
Supply v demand / economies of scale? Most consumer electronics get by just fine with the plethora of 0.001 USD 1% resistors and 0.1 USD 5% regulators out there, so regulators (I'm considering references and regulators to be close relatives here) and resistors with higher specs simply have very little demand. On the other hand, quartz-grade oscillators are essential for just about any kind of Wi-Fi, bluetooth, USB (sort of), anything that keeps track of the time of day, etc etc.
So even though a high-end resistor might only cost 0.05 USD in parts and materials, the demand is so low that they must charge a huge amount more to cover factory time, intial engineering outlay and capital costs, and so on.
-
I think that 1 bpm change is not really significant to talk about, while we have thermal EMF, aging, temperature coefficient, mechanical stress, michrophonic effects etc... on ther components which are several order of magnitudes bigger than this.
I think you mean ppb (parts per billion), not bpm (beats per minute). :-)
Also, I never really understood why we can get crystals in hermetic packages for 0,1 USD, while a resistor or a reference voltage is 10+.
Supply v demand / economies of scale? Most consumer electronics get by just fine with the plethora of 0.001 USD 1% resistors and 0.1 USD 5% regulators out there, so regulators (I'm considering references and regulators to be close relatives here) and resistors with higher specs simply have very little demand. On the other hand, quartz-grade oscillators are essential for just about any kind of Wi-Fi, bluetooth, USB (sort of), anything that keeps track of the time of day, etc etc.
So even though a high-end resistor might only cost 0.05 USD in parts and materials, the demand is so low that they must charge a huge amount more to cover factory time, intial engineering outlay and capital costs, and so on.
It was more of a ranting than a real quesiton. ;)
-
Is there a way to 'counteract' the effect of gravity/acceleration? Like using a second crystal oscillator placed 'upside down' respect to the first one, so that the two variations cancel out?
Isn't turning the crystal upside down (180 degree change) the same as switching the leads? So, if you had an oscillator that would periodically switch the crystal leads and average the results, wouldn't that solve the 'problem'? It wouldn't fix the 90 degree change, though.
No? The natural frequency of the crystal is a property of the crystal itself, independent of whether the leads are wired to anything (capacitative loading aside), let alone which way around they're wired. It's a reversible component like a resistor or capacitor. I believe using two crystals in opposite physical orientations probably would indeed cancel out errors if the entire arrangement is flipped entirely upside down, but I agree it's unlikely it would handle all orientations perfectly -- that would require assuming that the frequency shift was precisely a linear function of the gravity along a single axis and that axis alone.
(Stating the obvious; I suspect that in the real world, it's better to buy a crystal that inherently has required immunity to gravity.)
-
if you would go a level higher in the building and it then changes a digit, isnt that gravity detection ?
this is more a tilt detection.
-
if you would go a level higher in the building and it then changes a digit, isnt that gravity detection ?
this is more a tilt detection.
Weak answer: False dichotomy. It's the direction and magnitude of the force of gravity on the crystal that causes the effect, so it's perfectly fine to call it a gravity effect.
Strong answer: If you take a pendulum/mercury switch (a classic 'tilt detector') into space, it'll sit in some undefined position and be useless. However, the frequency generator will see the absence of the 1 g and deterministically output a change accordingly. Thus, just like a single-axis accelerometer*, it's detecting the force of gravity, not just the direction, so it's much more a gravity** detector than a simple tilt detector.
Dave, you should build a centrifuge and stick the frequency counter inside!!
* It is a single-axis accelerometer.
** According to some kind of relativity, gravity and acceleration are equivalent/indistinguishable.
-
if you would go a level higher in the building and it then changes a digit, isnt that gravity detection ?
this is more a tilt detection.
I don't know how high your building is but assume you have 100m
Earth radius 6371km
6371000² vs 6371100²
or about 31ppm of 1g
2g can change it a few counts, so it's safe to say that 0.000031g won't make a noticeable difference.
P.S. gravity is a vector not just a number.
-
yes now it makes more sense to me.
learning everyday ;D