Spring / Summer April 2018 and onwards deployment(s)

[hamishmb]

New thread for the new readings.

When I took the readings, I realised we had old ones dating back to 2018_01_25! Did anyone know the system had been on for the last 3 months?

Anyway, these old readings are attached as well as the new ones. They're a huge dataset, but if someone wants to plot on a graph with some tool, they might tell us something useful.

All looks fine as far as I can tell.

Note:
The probe in the butts is sticky, so may have been reading 0 for a while, but the level was around 400mm when I came in today.

Hamish

[TerryJC]

Did anyone know the system had been on for the last 3 months?

I did know that the WiFi was up when I was testing the Webserver Antenna installation, so I guess it was implied.

Anyway, these old readings are attached as well as the new ones. They're a huge dataset, but if someone wants to plot on a graph with some tool, they might tell us something useful.

I'll have a look at them later today.

[TerryJC]

Attached are two LibreOffice Charts that show the Butts and Sump levels over the monitoring period. I've also attached the two CSV text files that I used to create them, so that you can import them into your preferred charting tool, if desired.

I'll you to decide how useful they are.

[Penri]

Hello

Inspired by what Terry has done I tried an analysis of the measurements taken from the 3rd to the 5th April, looking in particular at the time the water in the sump was taking to go from one level to another. It's currently in tabular form as I couldn't get a chart to show the information in a way that I felt was meaningful.

I've had a bash at relating what I see in the figures to what happened on those days, I also noted that the 500mm level does not appear in the log, I'll need to take a look at the probe.

Hwyl

Penri

[TerryJC]

At the moment the levels oscillate quite a bit and once the system is fully functional, we should be able to reduce that considerably. Obviously, when we have a Gate Valve at the Butts, we'll only return water to the Sump when it is beginning to show empty or the Butts are particularly full and we can accommodate some of the excess in the Sump. I suspect that a fair bit of fine tuning will be needed, but until we have a Gate Valve so we can control both directions, we will have to accept this 'hunting'.

Having said that, if I've understood your comments correctly there was insufficient water in the Butts to have much effect on the sump levels, so the oscillation in this case must be down to rainwater and (potentially) leaks.

In the spreadsheet you made the following comment:

I find the repeated durations rather interesting, is this to do with the control software being at in sync with an internal clock? I would have expected the analogue nature of water flow to have more variation than be accurate to steps of 30 seconds (despite my truncation of the ms). I don’t think its an issue, at this point just interesting.

I suspect that the cause of the apparent accuracy of the steps is to do with the hysteresis in the measurement system and coincidence. However, it does seem as if the changes are synced to the clock. Hamish, is there anything in the software that would account for this?

[Penri]

Hello Terry

From my reading of it there wasn't enough head in the butts to push water back into the sump until around 08:40 on the 4th, from them on everything works very nicely. I'm not averse to the hunting, at the very least it keeps things moving and I find the timing patterns interesting.

The rise in level, which started sometime between 23:00 on the 3rd and 08:00 on the 4th, was in all likelihood rain.

Agreed on the motorised valve, I'm making very slow progress on that I'm afraid.


Hwyl

Penri

[Penri]

Hello

Just added a little more information to the analysis so I don't forget it.



Hwyl

Penri

[hamishmb]

Hi,

Interesting! I think the oscillation is simply because the valve is open. I reckon the reason the readings happen with such common intervals between during the oscillation is that the system adjusts the reading intervals based on what's happening. For example:

If we're pumping, we get new readings every 15 secs, so we know exactly when to stop pumping.
But if the level is otherwise not near the edges, the interval is much lower, more like 60 or 90 secs (can't remember which).

Also, the water from the butts probably flows at a fairly constant rate, so we have to empty at regular intervals.

We shouldn't be losing water this quickly though, and we really need to fix this. Our water consumption is probably fairly ridiculous.

Hamish

[hamishmb]

I came in today to investigate the problem, and I also downloaded the new readings, which are attached here. It doesn't look like anything particularly dramatic has happened - we've just lost water gradually.

In the mean time before I come in to try and fix these problems during the week, I have re-routed the river so that almost all the water is now flowing the long way around, rather than the direct route along the walkway by the Lady Hannam building - we identified leaks in those areas before. If the leaks slow down now, then it means the leaks in that area have gotten worse, otherwise I will still patch them, but then continue diagnosing the system.

My plans are:

1) Pressure-wash and fully drain the two disabled sections so I can re-seal them. Drain the affected sections using the Karcher pump, and some random pipe I hope I'll find (!).

2) Seal using the underwater putty/concrete stuff Penri bought.

3) Rinse and drain the rinsed water - making sure to *NOT* let it into the rest of the river and poison wildlife.

4) Refill the river using the water I stored in the butts today when I drained it.

5) Use our pipe sealant to fix the small leaks by the Wendy House butts - lower priority cos they're small leaks.

Terry, if you come in on Tuesday, and you help me get all this done please? We won't get it all done on Tuesday, but if we can do most of it that would be good. Anyone have any ideas where I can grab a temporary pipe for the Karcher pump from?

Hamish

[Penri]

Hello

Saw the dam today, Hamish, looks solid but a little water creeping through. Had you considered part filling a robust plastic bag (rubble sack or similar) or two with sand and using them to dam the water, placed as you have the wooden blocks on top of the weir?

See you when I get back.


Hwyl

Penri