Spring / Summer 2018 and onwards software

[hamishmb]

Sounds good, I believe that was our original idea

Okay, so if no one objects or has anything else to suggest, I'll implement an overall log in that kind of format.

Hamish

[TerryJC]

I'm not sure that it was. As I recall it, the original idea was, as per your example, a record of status etc for each sensor, with a separate file for each.

What I've proposed is a single file (and eventually one set of database records) that contain all of measurements from all of the sensors, so that we could chart it as I've suggested. Of course my example still lacks some key data points, such as the status of the pumps and valves in the system. The addition of those could make the file quite unwieldy. (Not to mention the addition of meteorological data at some point in the future.)

Of course, if each data item has its own record in a database, then we can construct all sorts of reports to allow us to study the causes and effects of events around the system.

[Penri]

Hello

I like the format on both, they work for me. I know they are examples so actuators may have been missed but I would also like to see status information on those devices, pump on/off or set speed and valve position.

Hwyl

Penri

[hamishmb]

Update:

The software now supports logging of readings from the remote pis at the master pi end - easier to obtain readings.

Readings files are now also rotated every week, with filenames in the form of (for example):

"SUMP:M0-2018-07-09 14:59:47.836151.csv"

The filenames are kind of long, but this prevents duplication of file names.

[hamishmb]

There are also various other improvements, like centralised config in config.py, meaning fewer cmdline arguments. I'll install this new software on Tuesday, all going well.

Penri, do you have the stage butts parts we need to go ahead with installation now? I should be able to do that with you on Tuesday as well.

Hamish

[Penri]

I have the parts, the ones we'll need first are at WMT.

[hamishmb]

Good progress this morning

Here are the latest results. Nothing unusual in there as far as I can tell. Note that the 2.4GHz antenna on our temporary router doesn't tend to work when it overheats - we'll probably just have to live with it for now though.

Hamish

[Penri]

Hamish

The sump was overflowing late morning, float was above 600mm but the butts pump did not appear to be on.

I moved the depth sensor to make sure it wasn't being obstructed and pumped water out of the sump to the butts manually. I let the water come up naturally to above 700 but again the butts pump did not cut in. I tried the approach again then used the net to move the float, I could not get the pump to turn on.

As I left I've pumped most of the water from the sump into the butts, turned the butt stop cock to almost closed so only a small amount of water flows back into the sump.

I will go back in later to see how things are, if I do is it worth turning the sump Pi off and on to. see if that clears the problem?

Penri

[Penri]

Hello

Went back in just before closing time, water level around 400 mm so has added about 200mm from butts since I left it.

Wasn't too sure how to leave things so: first I opened the butts stop cock fully and watched water level in sump rise through 500, 600 and finally 700mm, butts pump did not cut in; therefore I closed the butts stop-cock fully with the sump level sensor reading 700mm and circulation pump on level 2.

I'll drop a note to that effect to Nick (in tomorrow) and Mike (in on Thursday).

Hamish if you do make it in and can re-establish normal operation please let Nick and Mike know, thanks.


Penri

[hamishmb]

Hi,

Oops, didn't see your post until after closing.

I've had a quick look at the software and run it through the syntax checker again to detect any glaring errors, but none have made themselves apparent. I have time tomorrow morning, so I'll come in early and look at the log, fix it, and hopefully leave early. We'll see what kind of mood the router's in tomorrow

This wasn't entirely unexpected - I made a lot of changes, many of which modified the internals of the software fundamentally to enable future works, so I was expecting a few things to go wrong. It'll be fixed promptly

Hamish