The future of affordable home fabrication: FABTotum (Part II) - Sturdy design

I was not intending to further review this product so early after my previous post, at least not before their indiegogo campaign would reach $135000 and publication of a new stretch goal. However, a comment from a backer bringing into question the sturdiness of the design ultimately triggered the publication of photos of the internals of an alpha prototype of the FABtotum in the FABtotum Developer's blog.

These two photos are my favorites (check the blog for other photos). There you can clearly see the strong metal structure. When they say it is designed to "...sustain way more punishment than you can possibly cause without hurting yourself", they really mean it. There is no plastic in the structure, just strong sturdy metal.

I really think that the guys from FABtotum started the Indiegogo campaign already with a very solid design and tons of already tested and well executed ideas. 
This can be felt from the fast reactions to goals and resolutive behavior when deciding to add a new feature.

They did their homework very well . They tested several known technical solutions, evaluating the key advantages and disadvantages and further developing them to improve the performance. One example of this is the belt design. Though the real system has not been disclosed yet, alpha versions featured close belt designs similar to CoreXY, but requiring less belt and less pulleys thus reducing stretching and torque lost due to friction and bending of the belt, as commented in the Indiegogo campaign.

While some integration might still be needed, I really can not wait to see what they have prepared for us in their further stretching goals... well they are already over $110000, so less than $25000 to go.

The future of affordable home fabrication: FABTotum (Part I)

Several technologies have been used at home for fabrication by hobbyist from hand tools to CNC milling machines. However, no other technology has changed fabrication at home as the 3D printers did.
 
3D printing is not a new topic at all. Industrial 3D printers are a technology from the 80's. However, lapse of relevant patents on FDM technology allowed the first DIY 3d printing efforts which lead to the first open source 3D printer in 2007. Open source licensing has definitely helped in democratizing this technology.

Most blogs refer to 2014 as the new revolution of 3D printer, as basic SLS/SLA patents lapse. While this might be true, specially if a successful open source effort is released (there are already some DIY efforts in youtube in this direction), I believe that 2014 will be the year of multipurpose devices allowing more to personal fabrication by combining 3D printer with CNC milling and other functions.

Though not the first multipurpose device, FABtotum has caught my attention as being an open source effort toward personal fabrication focused on reducing the manufacturing cost of the device. It integrates a 3D printer, a CNC milling (2.5D and 4-Axis) as well as a 3D scanner.

Their crowd-funding campaign is running on indiegogo. You can see all the features of FABtotum in their indiegogo campaign page as well as the campaign video.

You can get one as a perk assembled or DIY kit for around $1000 (around 800 Euros), shipping included (and you will because they are already overfunded by more than a 100% with 36 days still to go). That is specially excellent news for European hobbyist tired of paying custom's taxes, as this one comes from inside the UE.

Of course 800 Euros is in many eastern European countries two (or three) times a normal salary. However, other open source 3D printers available in Europe go to 1600 Euros+VAT+shipping. Multi-tool devices (personal fabricators) go well beyond that. From that point of view it seems a good deal, apart from the fact that this crowd-funding campaign will help create a new company and a new product.

But is there more to it? Definitely yes! These guys from FABtotum seem to take seriously their stretching goals. If they reach $135000 (which they will and pretty soon), they will include a heated bed in each FABtotum.

If you look at the comments section of the Indiegogo campaign, together with the teaser possibilities in the main page of the campaign, I will not be surprised if they end up adding precise paper cutting with a (laser?) diode, a Dremel holder for harder CNC works, a SD-Card slot for stand-alone working mode (without having to be connected to a computer) or even LAN connected 3D printer.


Additionally, they seem to be listening very well to their backers (their first request was a heated bed and they got it) and I will not be surprised if it ends up being one of the most complete affordable personal fabricators in the market.

The era of the personal fabricators has started!!!

Arduino and 433 Mhz weather station

A long ago I bought a 433 Mhz wireless weather station in LIDL. It came with one remote temperature sensor and the possibility to show up to three other sensors.

Lately I wanted to show other temperatures measured by an arduino device by using a cheap 433 transmitter to send it to the station. Unfortunately I found no libraries for this concrete weather station, so I decided to try to reverse engineer the protocol.
I started by connecting a DSO nano oscilloscope to the data line of the remote temperature sensor transmitter, and looking at the waveform in Audacity:

There I identified the two symbols, coding digital '0' and digital '1' (I did not know which one exactly). Furthermore, the sequence of 28 bits was repeated 8 times in a frame.
With this information I searched the internet for similar protocols and I found this (spanish) and this, which dramatically accelerated my reverse engineering process, as most information could be taken directly from those posts.
When trying to send from the arduino, I could replicate the frames sniffed with the DSO, but every try to send a different temperature failed. So I assumed so checksum had to be present.
In order to decode the frame I recorded several frames with different temperatures, some forcing the transmitter to send and others just letting it work. This allowed me to identify the changing bits that could be the checksum. I considered one-bit and multi-bit parity as well as CRC (I originally discarded checksum, as only 4 bits were available).

Failure after failure I decided to try the checksum as sum of bytes taking as a result either the most significant nibble or less significant nibble, which also failed. Then, I realized that I should try options with higher entropy, which just discarding half of a byte were not, so I decided to start focusing in nibble operations... and that was it. The first four bits were sum of nibbles minus one!!

So summarizing:
- 4 bits: sum of nibbles minus one
- 8 bits: thermometer identified
- 12 bits: temperature in two's complement
- 2 bits: channel (0-2)
- 1 bit: low battery indication
- 1 bit: forced send from transmitter (causes the receiver to beep).

I decided to make the code available in Github under GPL v3 license.

Special thanks go to +dephiox and +Alejandro Vargas, as they managed most parts of the protocol.

Unfortunately, the receiver code of Alejandro did not work for me (maybe because of too much electromagnetic noise). Lately, +Petr Stehlík  tested my code (look at his post) and has created a receiver code, which works for him and which he intends to upload to Github.