Science/Engineering/Technical forums

I'm getting pretty straight lines, and we've had some nice swings in the weather (including the "weather" in the basement). So, now the waiting starts. I've been monitoring SIDs by observing WWVB (60 kHz) and I'm surprised to see a large increase in the daytime signal as winter approaches. It makes sense, but the magnitude of the change surprises me. I'm now getting daytime and nighttime signal amplitudes to fit on a single, linear graph; today looks like about 30 to 1, max to min, whereas during the summer, the graph would immediately peg at sunset and drop quite low during the day. It isn't relevant to the experiment, but interesting.

So the ionosphere is reacting. If I guess right, you're one skip away from WWV? I'm two, but haven't been monitoring it. If the lambda thing is a predictor, and the time long, it could be the prediction signal is long over, too.

I'm still running the experiment, but I did move it to another room last weekend. There haven't been many flares and I've been getting a lot of straight lines on the charts. I was watching it closely after the move to make sure it was stable at the new location when I saw a dip. Several hours later there was a flare 15:00 on 12/14. Interestingly, I was able to "predict" two more flares on 12/15 and 12/16 just by adding six hours to the times of dips in the chart (dips lasting several hours and dipping more than 0.05%. Those two predictions were dead-on, the second being within minutes. In reality, I think these were coincidences, but they're intriguing. These flares were pretty small, for one thing, and the 6-hour delay is just something I've latched onto for no particular reason.

The plot looks a little noisier lately, possibly due to the move, or possibly due to more "activity" of some sort. I think I need to patiently wait for a much larger event; these "signals" are a bit too close to the random wanderings of the plot. Heck, as far as I know, every little wiggle is caused by this effect, but I doubt it! I'll keep plotting along.

Negative results don't really say much, but it is interesting that I got one of the flattest plots I've had today (yesterday, UTC) and the solar x-ray activity is really flat, too. So, I suppose I should be able to "predict" no flare for the next several hours. : )

We'll see. I'd love for you to post (or link) some of the other data you've shown me. Pretty impressive, though the mechanism that would cause a dip on your device that also seems to predict a flare later is past me (at the moment, could make perfect sense with some interpretations -- example, dip in a 'field' that normally holds those back). Could be the excitement of the chase, or...well, you know. One thing I've found -- raw data is very often more valuable to have logged because a better understanding may appear later and it's good to have that stuff (with appropriate caveats on experimental changes/adjustments) later on when you know more about what you are looking for/at. Because if a new idea about the "why" or "what" comes about, there will be this gold-mine of data to test it against. Dash it against the rocks if it's a wrong idea, or....maybe we learn how to change data to information. (I know you don't have to hear this -- but there are lurkers who do)

Always save the raw...and whatever else you can. Interpretation might change, but data is data. I had that pounded into me at some point, and I'm glad of it, the idea has proven itself in many contexts.

This started as a lark, and, frankly, I expected to have an easier time getting down to well below 0.05% stability. Ha, that wasn't the case. I'm maintaining just about that stability over a day, so I should be able to see the larger events without much trouble. Anyway, I'm really looking for more smoke from the end of the gun before I go to much more trouble. This experiment doesn't even have a reasonable theory behind it, as you mentioned. If I become more convinced that this actually works, I'll set up a more sophisticated experiment with automatic logging and a better chamber or two. Right now it feels a little like reception of "scalar" waves or other similar nonsense, and I don't think it warrants much more attention just yet. I don't want to end up on late-night radio! : )

As I type, I see that the plot has made a little dip right at 4:00 UTC, so let's see if a flare is forthcoming, maybe in the morning just before sunrise. It's a small dip, and could simply be wander, but that's always the case. I might as well call the amplitude; I'll put it at C5.0. : ) Ok, that's just a guess. The one big flare I've monitored was preceded by a multi-hour dip (see the link below) but these little dips only last a couple of hours. They're all about the same amplitude, under 0.1%. That also makes me a little suspicious that it's just system wander.

I did post a lot of graphs here: http://www.techlib.com/newprojects.htm . The first image shows a long dip preceding a large flare.

The little dip I saw last night was pretty typical but there was very little solar activity today, at least on this side of the sun. In fact, the GOES satellite plot looks about as flat as I've seen(http://www.swpc.noaa.gov/ftpdir/plots/xray/20101222_xray.gif). So much for reliable predictions. I suppose there's no guarantee that the phenomenon, if there even is one, reliably produces flares.

Charles
Let it run
In terms of the universe your presence doing this is minute --all comes to those who wait--can not remember who said this but?????????????

Merry Christmas

After reading that same same news article last year, I also decided to try reproducing the results by experiment. I've spent the last several months getting equipment and researching stuff - I'm currently building the apparatus.

I've just recently discovered this board - I'd like to describe my proposed setup and get feedback and suggestions.

My experiment measures differently from Charles' ... I view this as a good thing.

I've obtained Uranium ore samples from Grafton, NH (Uraninite). The samples are well characterized and about 300myo, so there should be no appreciable decline in radioactivity for the one- to two-year duration of the experiment.

I plan to measure 3 GM tubes: One fully open measuring beta+gamma, one with a beta shield measuring gamma only, and one some distance from the sample measuring background radiation.

The samples decay at about 10,000CPM using the Radium series, which has a lot of Beta and gamma. GM tubes don't measure alpha.

At 10,000CPM, a variation of 0.3% translates to 30 counts. Averaging over 100 minutes should reduce the random variations enough to see differences at this scale.

I'm intending to timestamp and record every click with a resolution of about 30 nanoseconds. With interrupt latency and for various reasons, I'm expecting an overall accuracy of 1 microsecond.

Over 2 years, this will generate a 50GB database of individual clicks which I plan to make publicly available. The data can be analyzed to look for correlations ex-post-facto with anything that comes to mind.

The technical details are a bit of a challenge, but I think I've got everything figured out.

Any suggestions?

Sounds good. Be sure and get halogen-quenched tubes, or they won't last through this experiment. The old Russian tubes on ebay are hydrocarbon quenched and go bad in a couple billion counts, which you might hit on this.

Many geiger tubes DO count alphas, in fact, all of the ones I am using. If they have a thin Be or mica window, they'll do alphas. Has to be close to the source.

Time stamping individual counts may not be required -- you might get just as interesting data with some coincidence logic and looking at that.

Use the leading edge to trip your count/interrupt -- pulse widths on geigers are wide and variable.

!0k cpm is getting to a range where you'll drop a fair number of counts in the tube's deadtime )50-150 uS). You might want to use a little less stuff....pulses come at very random intervals and many are very close together, even at fairly low count rates. You can check this out on a scope before deciding to start a run.

Make sure they all see the same type of "sky". There will still be cosmic ray effects, and showers due to mass above the tubes interacting with primaries. I get significantly lower counts outdoors vs indoors here because of roof generated showers -- moreso on faster detectors like scintillator/phototubes, but also some with the slower geiger counters.

And, welcome to the board -- why not put up a little ditty on the "announce yourself" forum so we know who we're talking with?