Klaynos

If you cut out the tree area you could use the contrast between bird and sky and count the number of continuous areas. It's far from perfect. https://xkcd.com/1425/

The code is cleverer than that so it would only tweet when the situation is frost likely every few hours. Generating graphs is pretty easy. I wanted to avoid any new apps especially ones that would do things frequently as most of the year and even most of the winter there is no chance of a frost. I did consider a wireless link with a receiver which would change led colours based on the situation but decided graphs would be nicer.

I have a bit of MATLAB and a thingspeak app that is supposed to put up Twitter messages when frost is likely. It's never done it though. I'm not sure if it's a setting or code problem but I've never actually investigated other than noteing it didn't work. Prefer checking the graphs.

I have something running on a raspberry pi to do that. I don't use it though. I use a free service from thingspeak. They've got done nice tools and you can run MATLAB code on the data as it arrives etc... It's pretty nicely done.

It's funny you should say that. This wasn't going to be the second one I posted but... Project 2 - the meteorology instruments I wanted to answer the question of "do I need to drive my wife's car?" I have a Stevenson screen (this is news, last winter with the first installation I had only an unvented clear plastic box). I have 3 one wire temperature sensors, one in the screen, one on grass and one on concrete. In the screen I also have a dht22 for humidity (and another air temperature). When the 3 one wire temperature probes are all together they give readings that are one reporting level of each other (0.06 K). The 3 temperatures and humidity let me calculate a few/frost point temperature and see how different surfaces are likely to frost over. The sensors plug into a wemos D1 mini which is an esp8265 breakout board. This is a microprocessor with onboard WiFi and a PCB antenna. The data is sent via http to a web service which collects the data. On different parameter thresholds the system should tweet but this has never properly worked. The frequency of measurements depends on the values measured from every 12 hours when it's warm and dry to every 5 minutes. Between measurements it's in a super deep sleep to save power. It's powered from a 3.7 V lithium ion battery. I have a modified (else on I loose energy to back currents in the solar panel) charge controller with a 5 V solar panel plugged in. This is my most complete and actually used for purpose project. Between November last year until now it has had the solar and controller added, the screen added and some general upkeep but the code is pretty much as was. I'm not monitoring the energy in the battery or from the solar panel but I'm considering it in the future.

If they're identical how could you tell? (Similar problem to the one electron universe...)

! Moderator Note We are not equipped to help you. Sorry. Please please find your country on this list and phone the number in many countries this will be free. You're not alone people can help. http://www.befrienders.org

The WYSIWYG editor is trying to be too damn clever for its own good. I miss bbcode. I don't really understand their decision, people are becoming more technical not less!

I was aware of this. But where's the fun and challenge?

An Introduction to Error Analysis: The Study of Uncertainties in Physical Measurements by John Robert Taylor. I think this should be forced reading to everyone doing a numeric subject or any kind of experiment design or analysis.

Our experience is that if it's not achievable people just post in the other sections. I'm not sure if we've talked about (or if it's possible) to relate it to reputation though. That's an interesting idea.

I'm hoping to get some input, ideas and views on my ongoing projects. I'm also hoping that some of this might inspire some similar threads in this area. If more than one of these becomes a talking point I'll (or I'll ask another mod so I'm not involved) to split the topics. Project 1 - the parking sensor (On mobile so no diagrams I'm afraid) The idea was to use a set of ultrasonic distance sensors to perfectly park my car in my garage every time. My car only just fitted. The plan was to have a forward facing sensor, and angled sensor just inside the door to ensure initial line up. Then sideways pointed sensors every few inches to maintain distance from the edge of the tyre. Finally sensors at the end to get the distance I need to drive in correctly. A set of around 5 sensors would be attached to an arduino micro which would control a set of LEDs red (too close), amber (too far) and green. It's all be powered off of a 12v battery. I started by buying a few sensors and building a calibration rig. I'd get a distance readout in nearly real time for any sensor plugged in. There was then set distances at which a block could be placed. I ran this calibration for a set of sensors and plotted curves for each. I was surprised by how good they were between 5 cm and 50 cm, consistent and linear. So I needed to set up my system so that was the distances I'd be using. I then built a proof of concept with 5 sensors and a breadboard led set up. All held in place with gaffa tape. It worked! Great. Unfortunately by this point I'd planned phase 2 which would use a servo at the door with a sensor to track the front of the car in and then I got too good at parking without the kit and replaced my car with something that won't fit any more... Oh well...

I've seen a few posts recently about the fraction of science threads relative to politics and religion threads. I would agree with those posts but... Does anyone have any ideas to how we can shift the balance? Over the years we've tried closing those areas and tribe a hard line. It results in those threads filling up the other fora. Currently I believe you need a certain number of posts to post in those areas which certainly has reduced the rubbish. I really want to know your ideas and thoughts.

[offtopic]Which part of Somerset? Devonian here. [/offtopic]

If I were you I'd plot each parallel individually as lines between points. All on the same plot. Probably also on the same plot I'd include a box plot for each time. I might do this on the same plot but that might be too messy. Those are just the first steps. Next would depend on how they look. Caveat here is that I'm a physicist rather than a statistician or chemist.

Are you familiar with violin plots? https://en.m.wikipedia.org/wiki/Violin_plot http://www.sthda.com/english/wiki/ggplot2-violin-plot-quick-start-guide-r-software-and-data-visualization I can't add anything else to the last two replies to help.

That is a question with a massive number of answers. Almost all of which would be useless. Knowing how to answer would require us to have a few more details. What type of "project". Do you want funding or to whom do you want to "sell" your projects. What protection do you have on your ideas? If your talking academia then do you have an institution backing you from which you can apply for grants? What country are you in? These are the start of the questions that you'd need to answer for any meaningful advice.

What are the error characteristics of your measurement system? You have 6 data points. Are they all supposes to be a repeat measurement of the others? With the information you've posted and only 6 data points it's impossible to give any good advice. (with the possible exception of repeat the experiment some more)

A quick Google yields this http://onlinelibrary.wiley.com/doi/10.1002/grl.50236/full Which is slightly older but indicates that hindcasts are constantly being done as more is learnt.

I agree, many but not all new MRTs and the technology has been around for a long time (e.g. Docklands light railway, 1985). It's definitely an interesting area.

My thinking around this always goes to trains. It is far simpler to have driverless trains than pilotless planes. Why is it we don't have many driverless trains? The technology exists and has done for a while. My impression has always been it's mostly down to people feeling better with someone sat at the front notionally in charge. Now expand those feelings on trains to people's increased fears of flight. I don't think we'll have pilotless passenger planes very soon.

There are no neural nuclei. The each atomic nucleus is made of protons and neutrons. Yielding a positive charge. The op is just wrong.

Sorry. I'm going to correct myself. On reflection given the dependence here on the orbital time you need to be quite accurate. Therefore 5559 seconds is what you need. This give 432 N. Using a more precise orbital height will deal with the rest of the difference between that and Newtonian gravity.

The height and period are consistent with the iss. For the rest of the post I've not calculated it so wouldn't like to comment...

Whilst I agree with your first point to a degree. The seals do break and batteries do explode, you increase the probability of this by exposure to a strong vacuum. Also this is moving the goalposts I was replying to whether phones have sealed components (they do). Not whether this would be affected by a mild vacuum. You're second point I cannot accept. Take a submarine. They're fine to some really quite impressive pressure changes as described by swansont. But if you put it into space you have some very confused submariners who's air would be leaking out. Low pressure and high pressure designs are very different. Now this actually supports the argument that batteries are not too much of an issue as they are designed to stop the gas inside getting out, which is the same design requirement for vacuum operations. But again that isn't what I was replying to originally.