Great kick off session tonight.
We are going to pull this off.
Great kick off session tonight.
We are going to pull this off.
Itching to get started on this. Just waiting for the version 2 announcement. Any day now…
We will need one of these ourselves and I will buy the various bits if I can get some help to assemble it.
Looks like I’m going to need to learn how to solder, yikes
I’m starting to get sucked into the world of Arduino. Help!
I’m no expert on Arduino, @treb0r, but I’m happy to help as much as I can. I don’t think you’ll be short of help in that area.
Despite the crazy speeded up high pitched voice, these videos are a good way to understand what’s involved in building a food computer.
Still no word on version 2.0 of the Personal Food Computer. There is however plenty of activity on github, which seems to bode well.
I’ve yet to read ask the links in this thread, but have just reviewed the Food Computer main board BoM, (11 months old, V1 ?).
I wanted to see (i) What is monitored, (ii) What is potentially controlled, (iii) Rough idea of energy consumption.
I’ve only guessed at what items would be the largest power consumers. Assuming only the three items below, let’s say 350W max.
Energy consumption? I’ve no idea of the duty cycle of the heater and LEDs but made a wild guess at 40% ON time.
350 x 40% => 0.14KWh / Hr.
=> 3.36KWh / day.
=> 100KWh / month.
=> £15 / month (£0.15 / KWh).
Circulation / Vent Fans: 8W
Water Temperature + Conductance Sensor
Light Intensity Sensor
Air Temperature and Humidity Sensor
I looked for cheaper alternatives to some of the most expensive items, but couldn’t find much improvement on what was already chosen. As others have predicted, the CO2 sensor is the most expensive item, at $109.
Eagerly awaiting any news of the V2 Food Computer BoM
Our Personal Food Computer Lives!
Raspberry Pi 3 and Arduino Mega HW was used.
Pi had a wired Ethernet connection. Arduino was connect to Pi using USB for power and data.
PFC2 software Installation performed as per “http://wiki.openag.media.mit.edu/openag_brain/installing/installing_with_docker”.
Now I can access the web-based user-interface running on the Pi! I can export csv log files, attempt to recipes etc. Log files only have headings, (no readings as no sensors connected yet), and I haven’t worked out how to load the json recipe files yet.
This has worked on Chrome (Windows and Android), and Microsoft Edge. Internet Explorer 11 didn’t work.
I hit a couple of issues:
8GB uSD card was too small. Symptoms: Both containers always restarting (“docker ps” repots Restarting (1) for hours). Reviewing uSD partitions (“fdisk /dev/mmcblk0”, option p) showed over 12GB of partitions attempted to be created, but card was only 8GB. It now runs using a 64GB card.
I ran “docker -compose up -d” before the Arduino was plugged in. Errors messages were output, and only the Couch-DB container ran.
Nice! And paging @treb0r …
How did you get on with the BOM? I briefly took a look at this the other week when we were discussing it down at the space, but the part numbers appeared to be “internal” ones and the descriptions a bit vague. Have you got a list of actual manufacturer part numbers and/or specs?
Good old internet explorer, always keeps you on your toes.
Exciting stuff @paul123. I would love to have a play with what you have so far.
Let’s do what we can to figure out exactly what we need to do this and also how much it is going to cost.
As discussed, I’m happy to put some money in the kitty to get the ball rolling (faster).
I estimate the PFC2 BoM will come to £1388.22.
This is the sum of:
£750.54 (from incomplete OpenAG BoM as of 28/11/16: USD926.94);
£637.68 (My estimates for the remaining 70 other rows that have no costs in OpenAG BoM yet).
Our shared spreadsheet on Google Docs / Drive.
(Please let me know if you want edit-access.
If anyone cannot access this link, let me know and I’ll send the actual file.)
I’m also willing to donate £150 to the project. There was a lady at the AGM who was keen. Does anyone know her name so we can ping her?
I remember Giles was interested too. I’ll see if I can find his username to let him know of these updates.
I’ve never done a project like this before. The hardware for the box seems surprisingly expensive to me, at £382 for shelving, piping, perspex, frame-work and so on.
And their LED prices are unknown today - I’ve given a rough estimate of £126 for this sub-assembly.
Shall we see how fundraising goes then when we get commitments close to the target, make some purchasing decisions?
Hi Paul and Rob et al.,
Thanks for the info, I have had a look at the bill of materials looks pretty straight forward. Presumably the CO2, humidity, temperature and pH sensors all interface with the Pi using some kind of ADC? If these data could be processed by the Pi from another source then we may be able to use some existing sensors in the hack space. On the shelves on the left as you enter the main room there are a load of hard case boxes which have various detectors in (CO2, humidity, temperature) which I think all output via RS-232, they will need checking please feel free to have a look.
Unfortunately I will not be about this Wednesday, going to be pretty busy for the rest of this year but would very much like to help out when I can, next year will be a lot easier.
That’s great @Giles, I’ll have a “sensory rummage” tomorrow night thanks
The PFC2 sensors all interface digitally via an Arduino. I’m not sure if they use I2C our UART - I expect yours can be made to work with it.
I had some thoughts about the casing / box for the whole PFC2. I wonder if a fridge freezer can be repurposed for this task.
Integrated cooler, (and possibly temperature sensor).
White walls to reflect light.
Easy clean inside surfaces.
Cheap (for a broken one).
Electronics, nutrients etc can sit in freezer compartment.
Would look ok in a kitchen.
Integrated supports to rest LED assembly.
Can’t see inside, (mitigation: PFC2 already specifies two cameras. Fashion a transparent door inside the insulated door.)
Deviation from PFC2 design so standard recipes might give non-standard results.
I strongly believe that we should stick exactly to the published specification.
Using non standard components can be addressed further down the line when we are more confident about all of this…
We should scale this farm up to enterprise level
Check this out>>>>>>>