Overall System Design Specification

[TerryJC]

All,

I have just posted Issue 0.2 of the River System Design Specification to the Development Website (see http://www.hadrian-way.co.uk/WMT_River_ ... ss_0.2.odt). Please review and comment on it.

Some points:

1. I have recorded all changes in the document from issue 0.1, but then hidden them to make it easier to read (Use View/Track Changes to turn them on again). Don't forget to accept all the changes (if you agree with them) before you make any changes of your own.
2. I've tried to incorporate all the things that came out in the brainstorm, but not Penri's Capacitance Probe because we still need to work out why the capacitance increases with time. If the experiments work, I'll add an entry plus his circuit, etc.
3. I haven't done anything in the Annexes yet. That will come much later when we've firmed up the design.

Please post updated docs here and I will upload them to the website.

[hamishmb]

Okay seems good, very thorough.

Note:
Do we need to keep remote devices below 35 degrees C? This might be an additional expense, especially since Pis can run at up to 82 C and snesors don't give out much heat (probably).

I may add some stuff, but I need to digest it first XD

[hamishmb]

Oh, should we also create a simulation of our project (perhaps as a video) that we can show to visitors for educational value?

[TerryJC]

Do we need to keep remote devices below 35 degrees C? This might be an additional expense, especially since Pis can run at up to 82 C and sensors don't give out much heat (probably).

Actually the range is 0 to 70 deg, see https://www.raspberrypi.org/help/faqs/# ... emperature.

Even so, I can see that the figure does seem a bit low and we can change it if we like. I used it because it is half the rated spec of the board. Also, as you say it is unlikely that the sensor electronics will generate much heat, but if we are driving a solenoid valve, then the MOSFET will dissipate some heat.

Please be aware though that in a totally enclosed environment we are likely to get hotspots, as we did with the bells control box. The case for that system is made of metal and has an internal heatsink, as specified in our document. With the bells I was able to overcome the problem with a 'stirrer' fan, but that may not work in a smaller case. Also, we may have to use a plastic case and/or the case may be exposed to sunlight, so we shouldn't assume that there won't be any problems

Another thing to be aware of is that the BCM chip isn't the only device on the RPi board. When the bells system was failing, I was able to monitor a core temperature of only 48 dec C, but the system was still resetting itself. There are three chips on the RPi that need heatsinks, but even with those fitted, we still had problems.

[TerryJC]

Oh, should we also create a simulation of our project (perhaps as a video) that we can show to visitors for educational value?

I think any ideas to enhance the educational value of the Project are welcome. I have to ask though what would the video show that the visitors can't go outside and see for themselves?

[TerryJC]

Another thing to be aware of is that the BCM chip isn't the only device on the RPi board. When the bells system was failing, I was able to monitor a core temperature of only 48 dec C, but the system was still resetting itself. There are three chips on the RPi that need heatsinks, but even with those fitted, we still had problems.

If we do start having problems with overheating, we could use one of these https://www.amazon.co.uk/100W-TEC-Therm ... B007TPCEKI.

[hamishmb]

Interesting, my RPi runs stable at 50-51 C at full load 24/7, but I guess that's the silicon lottery or whatever they call it, so fair enough
The pad looks pretty interesting.

It could for example show how we collect rainwater, where it runs, how the pumps will work, etc. Perhaps an animation in a similar style to Clive's diagrams is what I was thinking. It could show people things they can't see, like the level & flow sensors (depending on which ones we use).

[TerryJC]

Interesting, my RPi runs stable at 50-51 C at full load 24/7, but I guess that's the silicon lottery or whatever they call it, so fair enough

Actually, I think it's because the Bells Pi is only running at around 1-2% except when it's actually playing something, then it shoots up to 6%

It still fell over due to hotspots.

It could for example show how we collect rainwater, where it runs, how the pumps will work, etc. Perhaps an animation in a similar style to Clive's diagrams is what I was thinking. It could show people things they can't see, like the level & flow sensors (depending on which ones we use).

OK. I see where you are coming from.

[Penri]

We made good progress on the inter-barrel plumbing today and should complete it sufficient for preliminary leak testing on Thursday. Testing at full load will need to wait for the welded joints to go off completely, 24hrs recommended. I planned to put in 2 vertical pipes, fed from the bottom of barrel interlinking pipe, so we can conduct 2 depth measuring trials simultaneously. Is that going to be enough?

I would appreciate a chance to review the design specification face to face, would that be possible on Tuesday 30th at WMT at say 11:30?

Hopefully this first attempt at using the forum is ok, comments and guidance welcome.

[hamishmb]

Hi,

That sounds good to me Penri. With some luck I might have finished the monitoring software by Thursday, but no promises cos I'm a noob with electronics XD

A bit earlier would be nice, but I can stay later that 11:00 if needed, though ideally I wouldn't do it every week XD

Seems fine Penri