A ballistic galvanometer is very slow; it measures integral of current, i.e. charge.
A transimpedance amp might work, but so might a zero-ed active capacitor divider (capacitor negative feedback becomes a hold capacitor). You could reverse the polarity and run the experiment twice, to get rid of the inevitable offset voltage.
It just has to integrate for a half second during the deceleration, then hold for long enough to get a reading. Relays for that kind of timing are easily arranged.
A load seems unnecessary now (a century later). Just feed through a switch directly into (-) input of an op amp, capacitor in feedback. That gives you circa zero ohms, of course. The capacitor holds the charge, so output voltage is equal to Q/C, with the switch ( relay?) open it can hold that value indefinitely. Before applying the brake, a short across the hold capacitior (another switch: probably a relay is best) will ensure that it starts off discharged.
There's still the coil resistance, of course.
Tolman was getting a nanocoulomb; a few hundred pF of capacitance will give a nicely measurable voltage. His notes mention that the rotating apparatus works best if made of wood; I wonder if a toy-top-drive, i.e. a pull string, would be the appropriate motor? Probably a small voltmeter can be incorporated into the rotor, with batteries and such.
I think the expected signal is about 200 nV.
The physics creates voltage. Why would anyone want to measure current?
This will be a very difficult thing to measure.
-- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
The voltage is in the range of errors in op amps, the current is far higher than errors in op amps.
Also, the 'braking' results in a pulse; one wants to integrate the pulse, not look at its dynamics, and current is easily integrated, onto a capacitor.
Yeah! I think this is a fun experiment. Matt, if you want to see Jan's circuit, you need to find a little triangle up in the RH corner and select 'show original'.
GH
Yeah, Matt, you've got a ~250 ohm source impedance, you might as well take advantage of that.
100 years ago current was the most sensitive thing to measure, it's easy to get stuck in the past. :^)George H. (I think I suggested to Matt that if he wanted to measure current he should use a TIA... so this may be my mistake.... A TIA needs a 'good' current source, with high input impedance.)
I ti
The source looks like about 200 nV, 250 ohms, and some inductance. It is not a current source. Calling it voltage or calling it current doesn't matter, but the physics computes voltage.
That TIA probably won't work. A TIA doesn't magically make noise go away.
I'd go for a good diffamp and use its CMRR to reject some noise. There will be a lot of noise.
Why not digitize it and process mathematically? Why not look at dynamics? That would be a good reality check. An integral is just one number, and there are lots of good and bad ways to get there.
The fixed coil takes out ambient mag fields, so it's got to be processed too.
-- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
And extremely difficult. The people who have done this successfully seem to have really worked at it.
I found another version that oscillated the coil rotationally on a torsion bar and used a lock-in. That was hard too.
-- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
Exactly.
-- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
On Wednesday, June 5, 2019 at 10:42:43 AM UTC-4, snipped-for-privacy@gmail.com wr ote:
tron motion upon braking the coil.
oil outside the rotating coil to measure the signal from the rotating coil? Elaborations appreciated!
the Food Truck when it came to Ithaca College. If you look at our setup, w e have something very similar: we have a second, stationary, coil directly below the rotating coil, and this coil is counter-wound. This definitely h elps to reduce the 60 Hz noise (if there is other noise, it is drowned out by 60 Hz noise). At the time I thought our methods were essentially identi cal. Do you think a larger stationary coil surrounding the rotating coil w ould work better than a stationary coil of the same diameter beneath the ro tating coil?
our voltage signal over 17 ms, so we get one data point for every 60 Hz cyc le. This does reduce our noise. I planned to try to remove the 60 Hz nois e via FFT and then fiter out the high frequencies, but I wanted to try and find an experimental solution first.
copper wire 8 inches from the rotation axis, and we spin it up to 6000 or
7000 rpm. It has a lot of angular momentum! Half a second is the best we' ve been able to do. And I am afraid we've worked pretty hard to get it spi nning down even that quickly. As for the split coil, we have essentially d one that with the rotating coil and the stationary coil. They can't both b e rotating, or the signal we are hoping for cancels out (the electrons in t he two coils are counter-rotating, so you get two pulses of current in oppo site directions).and is the electronic analog of the ballistic galvanometer. But it was har d to get the op-amps to be stable and still be able to integrate over a who le second. So I figured we have much faster electronics now, so to get the total charge I can just integrate the current, so that's why we are trying a TIA now.
Matt, I'm still kinda confused about the coils (signal and compensating) and the rotating contacts.. or whatever?
You've got the amp and batteries rotating on top of the signal coil? Where is the compensating (60 Hz cancellation) coil? And how do you contact it?
I our Earth's field thing, the two coils are in series. I don't think the size matters that much... except the compensation coil adds R and L to the input.
George H.
That combination gives you a voltage gain (10e9/200) of 5,000,000
The input noise is 5.1 nV/sqrt(Hz)
So we should see 25mv/sqrt(Hz)
1GOhm and 10pF = 10ms / 100HzSo a minimum of 250mV of white noise (above 10Hz)?
Given the 1/f noise is 250nVp-p, say 40nV rms; shouldn't we see something like 2V rms of 1/f noise at the output?
Where am I going wrong?
-- Mike Perkins Video Solutions Ltd www.videosolutions.ltd.uk
On a sunny day (Wed, 05 Jun 2019 15:56:17 -0700) it happened John Larkin wrote in :
= 0.000180
Is not that what a galvanometer does, measure current? Beta can be as high as 900 for some trannies.
Maybe in 1916 when the paper was published...
Collector current For an Rb of about 100 Ohm:
-> 900 * (200E-9 / 100) = 1.8e-06 is 1.8uA
1.8 uA in a 500 k collector resistor:-> (900 * (200E-9 / 100)) * 500e3 = 0.9 Volt
Is not Linux wcalc cool :-)
Fits nice in my storage scope ADC range using PIC 1.024V internal reference... Or any multimeter of this age,
You can always use a darlington to get an other factor 900 ... Zin does not go up the same ratio.
Using high impedances is often asking for interference trouble.
:-)
Math right?
On a sunny day (Wed, 5 Jun 2019 15:22:07 -0700 (PDT)) it happened whit3rd wrote in :
Yes, I find it really nice to see how they did those things with the available tools at that time. I have used galvanometers, although that is a really long time ago. Very nice and clever instruments.
That would require, though, instrumenting also the braking acceleration, as a function of time, which is excess to needs. Simpler to just note the initial velocity then stop it in a sub-second, and record the integrated current, the charge.
One measured rate, and a known 'full-stop' final condition, is sufficient information to give the integral of the acceleration. So, it relates to integral of current from the (known) source resistance. There's no reason to call it a 'current source' to make the result meaningful, if we use an op amp's pseudoground (-) node as a lower-impedance-than-the-wire current receiver.
In this experiment, both the original and modern versions, the real signal is typically overwhelmed by artifacts. It would be easy to frob around until you see the integral that you expect, declare victory, and publish.
And antenna.
-- John Larkin Highland Technology, Inc lunatic fringe electronics
There are papers online of modern versions. It's still really hard.
You have computed a voltage gain of 4 million in a single transistor. A reasonably biased single-transistor amp might have a gain around
200.If I were doing this, I'd consider using a good NPN differential pair, across the ends of the coil, running open-loop gain with the collectors going into a cheap diffamp. Mount all that on the spinning coil and bring out a big signal.
I think there are some really low-noise integrated diffamps too. ADI has some around 1 nv/rthz, but 1/f really matters. Getting a clean signal out of the spinning coil is really the hard part; the electronics isn't difficult.
-- John Larkin Highland Technology, Inc lunatic fringe electronics
On a sunny day (Thu, 06 Jun 2019 06:38:07 -0700) it happened John Larkin wrote in :
Yes trannies are good!
Bad spwcimen?
Yes mounting it on the weel is perhaps a good idea, optical link to receiver?
Right, my circuit is simple :-)
I found this,
Which doesn't talk much about electronics.. but does talk about getting leads to the rotating coil.
I'm still confused how Matt is making connections. It seems like the local B-field is a problem too... If that's the case then doing measurements in a city /building can be made more difficult by all the moving pieces of iron... Cars, trucks, elevators, etc. Which means the local B-field changes in magnitude and direction by 1-2%. Maybe things are better late at night. (when all good data is taken in my experience. :^)
George H.
Right, that's the torsion-bar angular oscillator. That makes an enormous amount of sense. Nothing spins. The leads don't wind up, no slip rings, the signal is symmetric/AC, and you can signal-average for days, essentially over millions of experiments.
-- John Larkin Highland Technology, Inc lunatic fringe electronics
Try it.
-- John Larkin Highland Technology, Inc lunatic fringe electronics
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.