I am starting this project as a learning experience and to give my contribution to the open-source community I love and support. I am a computer technician with a personal server that I use for everything. I've always wanted to relieve some of the network related work from it (like media storage and primary stage backups) and a NAS would fit my needs. But purchasing one to fit my constantly expanding needs is expensive and building one from store bought parts is just as bad. Plus I've noticed a lot of "what if" situations that the commercial NASs I've found don't address.
My goal is to make this as open-source as possible, using current open-source material, modifying them to fit this project when needed, or developing my own hardware/software.
Files
tidr606.pdf
Reference Design for USB3 to SATA Controller
Adobe Portable Document Format -
192.82 kB -
03/09/2016 at 20:22
Finally got a confirmation about the C.H.I.P. being shipped, so it should be here in a week or so. Once it arrives, I'm going to put it through a few tests to see if it would be more suitable for this project, if not, I have the Pine to fall back on. I have tested the Pine in Ubuntu with two USB/SATA and it seems fast enough so far, but I have not gotten a chance to stress test the read/write or network. So far I've just used store bought equipment, but as I've stated, I wanted this to be a ground-up project, mainly for experience, but also I'm a firm believer in fixing things yourself and not wasting parts. This way, if something goes out, you can fix it instead of throwing it away. We live WAYYY too much in a throw-a-way society and personally I'm tired of the "Oh, the next generation will fix it, so I'll have my pie now" attitude. It has been going on way too long and nothing has been fixed; society, environment, economy, etc. and that pie is almost gone. Everybody expected somebody else to bake another one. Don't wanna get into it. So I'm doing my part the best way I can by this project.
Anyways, cross fingers it comes sooner and it'll do the job.
Before I go forward dissecting the Pine64, something hit me: Do I need that much power? 64-bits? Will it be used or needed? For a project like this, I'm thinking no and could go for a smaller form factor like a C.H.I.P., but that may be too little functionality. I should be getting the C.H.I.P. next month (that was a longer wait than the Pine64) so I'll see if it will do.
Just a small update: Today I received the Pine64 today. So now that I have a visual, I can read the tech documents and see the workings and figure out how it ticks and how it can integrate into the project.
Just a shout out to the Pine Team: They were great. As any startup, they had some setbacks, but considering they started as just two people when the funding started and it severely outpaced their expectations (hard to anticipate the popularity of an idea when just starting) they did a fine job. They kept their backers updated as much as possible and despite some not being happy (we live in the "I want it NOW!!!" world) they still kept going. For those thinking about crowdfunding or backing one, there will be setbacks. Things happen. Production runs fail, natural disasters happen, QC doesn't live up to par... all these could set a project back. As long as the project leaders keep you inform and provide transparency in the process, you should just stay calm, sit back, and don't immediately think it is a scam. You are not working with Apple, the corporation, you are working with Apple, the garage startup. Patience.
Now without firing it up yet, the design looks fairly good. PCB looks professionally done, components look solid to the naked eye (except there was one momentary button loose in the box. It fits in the power or reset position and obviously not soldered. Wonder if this was intentional, an after thought at the production facility, or an option. Gotta read more.) , and the separate and optional bt/wifi fit snuggly without question how it fit.
In other news, I was going to set up a custom USB microscope to take pictures of PCB layouts, components, and such, and while I have the majority of the parts, the zoom lens is expensive for now. I choose a zoom instead of a fixed so I could adjust the view to fit the part without changing lenses. That way I could get the smallest component clearly for part identification while still getting the majority of a PCB in view to show where it goes. Trying to make things as one armed as possible. I do have 25x jeweler's glasses coming, So I might be able to use the detachable opticals and temporarily use them on my iPhone for pics until I find a price point I can afford.
So if you are interested in the Pine64 (no I'm not getting anything out of this plug, I just liked the idea and so far the product) you can head over to pine64.com and look around. They have links to the hardware and software on their wiki and are updating at wiki.pine64.org.
Now time to do some work. Pics of the board will be coming soon.
Well, March is here and I should be getting a Pine A64 shipped sometime this month, so I will have "the brains" for testing this project. I'm still researching ways to get things together like schematics for a USB to SATA converter (TI has a schem of one, just looking into how difficult it would be to change the USB type) and I also took apart one of my 12 port USB2 hubs just to see how it ticks. Finally got both computers running (yay!) and bought a HD USB cam to take pictures and two bright USB lamps. All I need now for it is a zoom lens that will be able to see the small SMD parts clearly and my electronics/soldering microscope will be finished. I've also found the Allwinner A64 chip (3x what it was advertised on the web, but I bet $5 is for large bulk orders) and pinouts (or is it ball outs now?), so when it is time to design the custom PCB for the NAS, I'll be ready. The other components should be easier to get. I'll post the SATA adapter schems and A64 whitesheet later for everyone.
The first entry for the project. So let me explain my mind set. I love building things and am quite the do-it-yourselfer and I'm around computers so much and been so all my life I swear electricity is in my veins and I think binary. But that is an expensive hobby; things change constantly, new technology gets developed while old technology gets dumped and it can happen fast. Plus needs change over time. Buying outright from stores is not cost effective nor fits 100% of your needs and building your own systems may get you there, but adds more to the expense. Then came the open source community. Now you can design some of your own components or program your own programs to perfectly fit your needs and learn how things work in the process.
Well, my need is a headless server for storing my files. I went digital a few years back and love it. Streaming movies, televisions shows, music, turning my 65" tv into a temporary work station, life can't get any better! (Then something better comes out.) I take my blu-rays, rip them to the quality I like and prefer with the options that fit me (why do I need Russian subs and audio when I'm Texan?) and stream them to every tv and device in my house. TV shows I record on to the computer, edit out the commercials and watch them whenever I dang well please anywhere I dang well please. I get my CDs, rip them to FLAC, and I have full access anywhere. Problem is, that takes up space. A LOT of space. My "server" is custom and hand built, so it can handle a lot, but HDDs are only so big and the bigger they are, the costlier they are. Anyone who owns a computer knows you store a lot of crap on there. So add the above, plus programs, plus games, plus all the junk you've downloaded and forgot, but keep "just in case", plus photos and videos both the wife AND mother take on their phones of the kids, plus documents and god knows what else, and you have a storage nightmare. A "jack of all trades" system can't handle it. 6Tb is still too little. And backing that up? Good luck. With ISPs data capping you, do you know how long it takes to back up 6Tb in the cloud? Your drive will fail before it backs up completely and your ISP may shut you down for the month or charge you an arm and a leg for overage. Plus I had 2 hard drives die in the past year and one of them I didn't have a place to back it up or the time to move the essentials to the cloud, so all my music was lost. I had every CD since the invention of ripping I owned on there and some of them I can't replace because CDs are not indestructable, the moving trolls just happen to like your style of music and play them at the "they're gone for good!!!" party with the underwear gnomes and those gremlins that breaks stuff, and those "friends" forget to return them or disappear on ya.
So I discovered NAS. Great idea! A server that stores your files and is accessible from the network and its primary purpose is just storage. Now the cheapest I've found was $85 dollars and was only 4-bays. Not upgradable or expandable. And it only goes up from there. I could build my own, but I might as well just build another computer, just as expensive. Then came the microcontrollers like Rasberry Pi. Pretty powerful, considering, and cheap. It is cheaper than a CPU, motherboard, fans, and video card, but while they can be used as servers, they weren't build primarily for that use. There are limitations and to overcome them, it is hard and expensive, not invented yet, or just plain impossible with the technology used.
Well, that is what I hope this project addresses; overcoming those limitations. For the brains, I funded the Pine64 Kickstarter. It is a cheap and powerful microcontroller and will be open source. So I could add things needed and remove things that aren't to make it cheaper or to fit the needs of this project. I will also be designing USB 3.1 hubs, a 6-12 port internal one for hard drive use that will be switchable to extend drive life and a 4 port for external devices,...