NEW TO OTHERNET - Information Request (Technical Project)

Syed · 2020-10-22T17:45:34.686Z

You should take a look at the work that @Martin has done.

S_H · 2020-10-22T17:49:03.102Z

I think the battery could be made bigger, but then the weight pay off could mean its tiresome to maneuver.
the ports of external charging and input for the solar panels will be covered by a rubber seal “plug” in my head i think this would be the best option, especially if it is a heavy duty plug with decent O-Rings.

but to answer your question, the parts that will be in direct contact would be the Monitor and the Keyboard, but bare in mind the monitor will be sealed with some kind of adhesive, and the keyboard covered in a rubber membrane.

One problem i can think of is if you damage the rubber membrane.

S_H · 2020-10-22T17:51:05.614Z

That is exactly what i was thinking! just without the extender.

i think you could make it pretty compact.

Syed · 2020-10-22T18:07:07.349Z

If you are starting this from scratch, you may not need the Arduino, as there are a few pins exposed on the Dreamcatcher which should allow you to control external hardware.

S_H · 2020-10-22T18:09:51.430Z

Does the Dreamcatcher allow for reprogramming?> I saw that the I/0 pins.

Is there a Schematic anywhere that outlines the IO ?

Syed · 2020-10-22T18:27:46.786Z

Dreamcatcher 3.05 SoC-GPIO Schematic.pdf (1.5 MB)

S_H · 2020-10-22T18:31:22.950Z

Thank you, does the software allow you to add extra code?

Syed · 2020-10-22T18:49:34.586Z

You have root access to Skylark, which is based on Buildroot, so you can do whatever you want. But it’s highly stripped down. Sorry, we don’t provide any support on custom development projects.

S_H · 2020-11-04T22:52:25.199Z

Schematix1328×729 46.6 KB

So this is what I have been thinking so far.

Can anyone spot any obvious problems with this?

Thanks for all your help!

maxboysdad · 2020-11-05T16:12:42.391Z

It looks good, but I question what do you need the “Digital Signal Video Decoder” for, as the RasPi has HDMI output? Just asking because it looks like you are pretty well covered otherwise. Jerry

S_H · 2020-11-05T16:47:04.285Z

The Video Decoder is just the driver board for the HD monitor, I have decided not to go with a normal panel design, so the monitor comes in 2 parts.

The screen itself, and the driver board that handles the signal processing and has access to 12v rail voltage.

Thanks, I was toying with designs today to make sure it would actually fit, although very basic. this is what i was hoping the internals to look like.

Schematic 4.PNG2337×826 404 KB

Schematic 3963×1068 64.8 KB

Schematic 21692×948 285 KB

Schematic 1.PNG2355×920 710 KB

maxboysdad · 2020-11-05T17:21:24.506Z

That looks like a pretty good layout. How will your LNB be mounted when operating?

S_H · 2020-11-05T17:41:59.974Z

I was hoping to make something that would auto track to the nearest satellite.

The thought was to have a X,Y axis servo holding the LNB, Have the system run on a Arduino Nano.
My hope was to get a GPS receiver and a compass module to send Longitude/ Latitude data to the Nano, and by using the magnetic compass, use that information to move the servos pointing the LNB in the correct position.

Another way I was thinking of is using a startup script from the host Raspberry Pi, that would run the servos through the max range, while doing this, the signal strength from the satellites would be used to almost auto focus on the specific point in the sky.

They both have various bad points to them. Its just how far do i want to be pushing it.

I want to make something that is efficient but also not complex, the least amount of complexity the better, less things to fail.

What do you think to the idea?

maxboysdad · 2020-11-05T17:48:14.760Z

I like the ideas you are putting forth. A bit advanced for me to take on, but it can work. I would send you to read @Martin Martin’s threads here:

Outernet for sailors (weather forecast grib files) Feature Requests

Unfortunately not yet. The part with the battery is still missing because the lowest layer just won’t fit into the hole. I need to make a new one with the CNC first. But there are no LEDs in my setup plus the transparent ball would not light up in the dark like a white one. I don’t have a lot of time at the moment to work on it, but I’ll make a video when its ready.

A lot of good ideas and he shows how he is building his unit into a geodesic designed weatherproof enclosure…

Martin · 2020-11-05T23:47:50.379Z

@maxboysdad Thanks for mentioning me here, also @Syed, but sorry I seemed to have missed your post. @S_H I like your project and your ideas! I also have a GPS included in my setup but I’m not using it in the code yet. I guess the best way would be a combination of both methods you mentioned. I am planning to use the GPS and compass/gyro to roughly point to the position of the satellite and do the fine-tuning by checking the signal strength. Also if you won’t use a GPS, since the satellites are geostationary it would be much faster only checking the sky towards the equator. Another issue I recently learned about is that the reception with the bare LNB seems to be not reliable enough in most areas to get a good reception. You should think of adding a kind of horn to your setup. Also have in mind, that on a boat, especially on a steel boat, the compass readings can be quite a bit off. My boat is made of steel and judging by my magnetic compass this seems to be quite an issue. I’d like to make a test with a magnetic sensor and my arduino to make a kind of deviation table. But anyway, cool project and I wish you best luck with your scholarship!

Mark_Phillips · 2020-11-06T19:30:07.698Z

You’ll need to make sure that the “inverter” has a built in solar charge controller otherwise you’ll get nothing from your solar array. Whilst solar panels claim to output (say) 5amps they only do that when they are in full sunlight and there is no significant load connected to them. The charge controller makes sure that the panel does not see a significant pull from the load.

Try to calculate your load including battery charging draw. This will dictate how large of a charge controller and panel arrangement you’ll need. A PWM charge controller is simple to design but a MPPT controller will yield more power for longer. Bear in mind that you are only getting full solar output for about 5 hours a day. You can increase this by aligning the panels optimally with the sun until it sets.

Mark_Phillips · 2020-11-06T19:38:36.261Z

Also,

I see you are using WiFi to connect from the RPi to the Dreamcatcher. I would do this as a wired ethernet link. This way your RPi always has access to the dreamcatcher using the most efficient and economical (ethernet crossover cables require no additional power) manner possible. If you need WiFi access to the dreamcatcher data then I would have the RPi act as a WiFi access point.

S_H · 2020-11-06T20:32:22.236Z

Can you do a direct link between the Dreamcatcher and the Pi ?
I looked at the schematic for the dreamcatcher, and I didn’t see any I/O that indicated this?

As for the inverter, I have not looked too deeply into the inverter yet, would you recommend a specific model for this type of thing?

Rough estimates show that it will be running at 4/5Ah with everything running for purpose with 2Ah Idle with the screen in standby mode. so my idea for the 18650’s is bundle a cell of 3 giving me 14ish volts, and then times that by 12, giving me around 40AH Capacity at 14V.

GPIO Dreamcatcher v3.05658×554 206 KB

S_H · 2020-11-06T20:35:01.876Z

Do you know where i can find the full schematics? for the dreamcatcher

Mark_Phillips · 2020-11-06T20:51:47.407Z

You can replace the WiFi adapter on the dreamcatcher with an ethernet adapter and then use a crossover cable.

Or as Syed has already stated, you have access to the root of the dreamcatcher. You could do some sort of network re-route via the IO pins (I2C maybe?) into the RPi. The serial port should be capable of megabaud speeds?