Summer / Autumn 2017 Prototype Deployment

[TerryJC]

NOTE: In line with the relocation of the SumpPi control assembly to the green box, the SSR assembly will be relocated there too, I'll re-arrange the mains wiring to suit.

I'm assuming that by this you mean the SSR itself and the switch?

[Penri]

Terry

Yes, its all contained in a small plastic box.


Hwyl

Penri

[TerryJC]

Yes, its all contained in a small plastic box.

Yes. I wondered if you might be relocating the 'innards' of that box and putting the switch on the outside of the Sump Pump Control Box.

[Penri]

No I decided to keep it simple and re-use what we had with minimal changes. Your suggestion worth noting for consideration in any future re-design.

Hwyl

Penri

[Penri]

Good morning

I'm going to be late getting to WMT tomorrow, I need to take a poorly dog down to the Vet at 09:00 then bring him home before coming along.

I have finished the magnetic probe and SSR cables plus concreted a foot onto the magnetic probe, which I hope will make it sit on the sump bottom, so intend taking those along to WMT before it closes today.

BTW the probe will need to be tethered at it upper end to make sure it doesn't fall over.


Hwyl

Penri

[Penri]

Hello

I've just been down to WMT to drop off the probe, SSR assembly and the probe connection cable. I also took the opportunity to drill a hole in the bottom of the green box nearer to the sump, through which I propose we route the various cables. The attached photographs show how and where I've left things, note that I've used tie-wraps to hold the magnetic float away from the concrete foot while it finishes curing, they need to removed during installation. The concrete in the "foot" has turned out to be sandier than I was expecting, please leave the plastic container in place as an when you lower it into the sump.

Things I need to do:
Size and attach wooden battens in the green box from which to hang/mount the SumpPi control box
Rearrange the mains wiring to provide sufficient supply points in the green box for the stuff being re-located there.

Hwyl

Penri

[TerryJC]

Hamish,

After our (eventual) success this morning, can you confirm that you only updated the two Pis to autostart the programs and didn't revert the delays to 5 minutes?

[hamishmb]

Hi,

Sorry, just got this. I did revert the delays, was I not meant to? XD

I can briefly pop in tomorrow and change if needed.

Hamish

[TerryJC]

Ah, that explains a lot, see my sitrep which I'll be posting shortly.

[TerryJC]

Sitrep

Over the past two days we have done quite a lot. Yesterday morning, we successfully tested the Hall-effect Probe with Sump Pi; the levels all came up correctly. We then tested the SSR but found that it wasn't switching on; this was diagnosed to insufficient drive voltage from the Interfaces Board. We proved this by temporarily driving the Manual switch from 19 V instead of 5 V. It was decided that the work to make this permanent for both manual and auto was too much to attempt it on site, so I took Sump PI home.

Yesterday afternoon I completed the conversion and this morning we reconvened at the WMT to try again. Initially, we had a problem due to my dumb configuration of the Ethernet and WiFi Adaptor, but once that was sorted we were able to talk to both Pis. Unfortunately, there was still a bit of a problem in that the SSR worked on auto, but not on manual, but that was ignored in favour of getting the system pumping water. Hamish did a bit of debugging of the software and shortened the delays from 5 minutes to 30 seconds so we could see the results of changes without falling asleep! (Or ruining Penri's back since he was simulating a full sump by holding the float at the top.) At that point we had some confidence that the software was doing a lot of what it should, but we'd only been able to test half the functionality.

This afternoon, we traced the SSR fault to a wire off on the cable between that Sump Pi and the SSR boxes. Once that was fixed we were able to observe the sump filling from the butts (by opening the valve) and the Butts Pump coming on when the level got greater than 700 mm. We then reduced the flow of the water from the butts and observed the Butts Pump switching off when the sump level fell below 700 mm. We didn't have time to test the float switch stopping the Butts Pump or the behaviour when the sump level got too low, so there's plenty left to do.

Things to Think About and Do

1. Modify the delays to make the reaction times a bit quicker, (we only need to store the measurements when they change, but we should keep looking).
2. Think about what hysteresis we want when the sump depth reaches the high level and then comes back down again (the same at the bottom).
3. Devise more tests to exercise the system as much as possible.
4. Analyse the results to help us decide what needs to be done to make the system work as we need it to.