Slashdot Mirror
Constructing A Low-Power 2U Wireless Rack-Box
adelayde writes "Recently we decided to build ourselves a custom rack-mountable box that we could use as a web and DNS caching proxy and which would offer flexible wireless networking facilities and have an uniterruptible power supply. The result was a 2U rack-box with dual wireless networks built upon a low-power Via EPIA MiniITX motherboard. The box has two wireless networks built in with external antenna connectors, locking switches on the front to avoid tampering, a battery to give at least 20 mins of autonomous operation, a low wattage power supply and most importantly lots of blue LEDs :)"
1. Sell these
2. Profit
No ??? needed!
Seriously...
--TheOrangeSquid Is it any wonder things seem so awry? We swim in a sea of confusion and don't have to think to survive
Yay LEDs, they're the wave of the future because they're all blue and such. It reminds me of those old Flash Gordon episodes where everything was polished chrome and looked like it could kill you if someone flicked a switch somewhere.
Oh yeah, and I already can't reach the site. *laugh*
----- Wtcher Dragon, UDIC
Let's see how long she lasts ...
...and most importantly lots of blue LEDs
You may have meant that as a joke, but blue LEDs suck quite a lot more power than red or green ones.
When you care about power consumption, rather than coolness, come back and ask again.
http://mirror.us.psand.net/plinth/
good thing it's not a chick that made this thing, or I could get in a lot of trouble for say this....
"Hey, Nice Rack!!"
Do you actually get decent performance out of mini-itx computers?
I've been toying with the idea of making a HTPC-ish box, but the performance questions stop me from buying anything.
Then how about a mirror?
http://hackish.org/~rufus/flakey.info/plinth/
Looking at the pics on the page, it looks like they could have easily built the thing inside a 1U chasis of they used the proper power supply and heatsink. All of the other parts should fit within 1U.
That being said, they could have simply used an ultraportable laptop with the screen unplugged and unnecessary parts removed/disabled.
You'd be amazed as to how little there really is inside a laptop. Think about it -- the drives and batteries take up about 75% of the chasis. Leave about another 10% for the power supply and heatsinking, and you've got a REALLY small PCB.
If space, not power, was their main concern, they could have also used one of the Shuttle cube boxes. They pack an incredibly strong punch for their size, and are usually on par with their desktop equivilants. Hell... they've even got an opteron box. The performance on the EPIA boards is horrific. What were they thinking designing a processor without a FPU? That being said, they're pretty cool because they're small, low-power, and widely availible (which laptop MBs strangely aren't). Still, they're pretty expensive considering that you're getting a PC which would have been considered pretty slow 4 years ago.
-- If you try to fail and succeed, which have you done? - Uli's moose
The regulator looks like it's there to power the "6V yellow led". Obviously these guys know nothing about electronics - the wireless bridge is powered from a "DC-DC converter" made from an LM317(!) - that's a linear supply, not a DC-DC converter. This supply is probably superfluous anyway - the wireless bridge it powers runs off a 7V supply, telling me it most likely has an internal regulator. They should have checked - might have saved themselves some work...
Incase it is needed: http://fluky.org/slashdot/flakey.info/plinth/index .html
Their site seems to be slowing quite fast:)
I'm a college degreed Electrical Engineer with over 30 years of experience, including teaching electronics at the college level for three years. That regulator circuit looks quite familiar. As long as the heatsink is sufficient for the heat dissipation in the LM317, there should be no problem. Since this is powering only the wireless bridge, the current drain at 7 volts should be modest and it only needs to drop 5 volts across the regulator IC. The total power dissipation spread across both of the 1/2 watt resistors is only 85 milliwatts so no trouble there either.
If you are referring to the capacitor voltage ratings, the only requirement there is that the voltage rating of the cap be more than the voltage actually applied to it. For example a 35 volt cap is just fine with 6 volts across it. It could even be a 1,000 volt rated cap with no ill effects.
"Do the Right Thing. It will gratify some people and astound the rest." - Mark Twain
"Do the Right Thing. It will gratify some people and astound the rest." - Mark Twain
Personally, I found their antenna designs much more interesting.
There has to be something special or unusual about this setup, but damn it all I can't see it.
Sure, they integrated a wireless bridge into it, but with all that room, it doesn't look like anything special... So where's the nifty part that makes this thing "News For Nerds"?
Looking for hardware (Currently need: Large Etch-a-Sketch) Have one? See my journal!
1. Go to ebay.
2. Find and buy cheap notebook.
3. Get into argument with seller over shipping.
4. Wait for notebook to arive.
5. Pick up hammer.
6. Open notebook.
7. Hit notebook screen with hammer until it comes off.
8. Stick some WiFi cards in notebook and put it on shelf.
"Have you ever thought about just turning off the TV, sitting down with your kids, and hitting them?"
Link: Soekris
42
I'm curious as to what their definition of "low power" is. Low, compared to a power-sucking P4 or Athlon, maybe, but probably not very low by low power standards.
I've just set up a similar system as an home file server (no wireless, though, and I've added a cheap DVDROM drive), and my box is sucking up around 55-60W, idle. That's measured via an actual wattmeter connected to the power cord, and not by multiplying V*A.
On second thought, maybe a soekris board and a 2.5" disk drive might have been a better solution (less RAM and CPU, which would probably be fine for an home fileserver, but the power usage would probably be in the 10-20W range).
How fast a CPU do you really need? Years ago I popped open a 3COM Ethernet bridge/router and found a Motorola 68020, running at 25 MHz if I remember correctly. It was fast enough to handle two fully loaded 10-megabit Ethernet segments.
Mea navis aericumbens anguillis abundat
He may have built this for remote locations
or heat filled rooms/closets. In those cases
you want decent space and decent airflow INSIDE
the box.
Laptops have neither, and tend to suffer heat
related problems easily enough.
Its good already that he went low power and
low thermal, and put it in a good solid spacious
chassis.
If dollars were a concern, going down to 1U
isn't bad, but no further for real applications.
So, he writes up some stuff about his new box, and then posts it on Slashdot to stress-test the thing? Clever, clever...
Hey, you try to find an open nick these days!
I find that UPS battery being too naked next to the powersupply... It rather be protected by itself in a cage because battery's can do strange things sometimes (I can tell :S)
..
...
if that thing explodes in your rack you can throw away the hard drive, maybe the mainboard and wireless bridge
so far for a rendundant/autonome system
although a beowolf cluster of blue leds? hmmmmz...
--- I am known for the ones who want to find me on the net. Is that a privacy risk or a privilege? One might wonder..
They used both a DC-DC supply and a LM317 converter. The DC-DC supply was a pre-made thing that takes in something like 12V and supplies what you need for an ATX motherboard, 12V, 5V, 3.3V, -3.3V, etc. Lots of low-power and small form factor computers use them, as they dissipate less heat and are smaller than normal AC-DC switching supplies.
Then they used a LM317 linear regulator to provide 7V from 12V for the wireless bridge. The bridge's power supply was rated for 1.4 amps. A 5V drop over 1.4 amps is 7 Watts wasted as heat. Not impressive in something that's supposed to be low power. With their 20 DegC/watt heatsink, that is a 140 deg C temp rise from ambient to the TO-220 case. Easily out of spec. The wireless bridge surely doesn't draw the full 1.4A in operation, that's what saves them from their regulator melting.
They probably should have seen if their bridge would have run from the 5V line or if the bridge's power supply could handle 12V. They probably didn't even need the 7V regulator.
If they did need one, it would have been much better to use a switching regulator that would be around 95% efficient instead of their 58% efficient linear regulator. TI powertrends makes integrated switching regulator modules that would work perfectly. I used one for to power a digital camera off a 12V deep cycle battery so it could take photos on a timer for several days straight.
I also have to wonder why they used locking switches instead of just getting a case with a locking front panel. I've got a rack of such 2U cases at work, they're not hard to come by.
The switch from the 2nd hard drive's power was unecessary too, you can turn the drive off with software.
Looking at this page, and a bunch of their other pages, I think cost was their main concern. They found a 2U box for only GBP29, while the 1U box mentioned in another post by jjshoe runs GBP149. The wireless access bridge "was just laying around".
Most of their projects are of the "cheap and green" variety. They have built pedal powered repeaters, solar powered satellite receivers and the like.
I'm a bit appalled by their lack of engineering knowlege on some of their projects. But I do admire their "slap it together cheap" attitude. They are not building reliable, production level kit, they are tweaking items they found in the spare parts bin and making useful one-off projects.
the AC
Hemos is like...sci-fi fans;he thinks technology is cool, but he hasn't bothered to understand the science it's based on
here
Debian GNU/Linux Sid
The milliamp rating of the part bears no direct relationship to the need for heatsinking. The heatsink requirements are related to the amount of power that the regulator IC needs to dissipate to stay within rated operating temperature limits. The power dissipation is given by Pd=(Vin-Vout)*I. Where I is the current drawn through the regulator. For example, in this case being discussed, the voltage drop is roughly 5 volts across the IC and if the current was 1 amp, then the power being dissipated would be 5 watts. Given, from the data sheet, that the junction to ambient thermal resistance is 50 degrees C per watt, the junction temp rise above ambient for 5 watts would be 250 degrees C which is far in excess of the junction maximum operating temp of 125 degrees C. So in that example a heatsink, which reduces the junction to ambient thermal resistance, is clearly a necessity. See LM317 data sheet
Don't feel bad, though, it's a common misconception that has caused a lot of burned fingers. You are right that the heatsink can't hurt, even if not required, since lower temperatures are ralated to lower MTBF (Mean TIme Between Failures.)
"Do the Right Thing. It will gratify some people and astound the rest." - Mark Twain
"Do the Right Thing. It will gratify some people and astound the rest." - Mark Twain
when I got a patent on them for my corporate employer in the 70's using all discrete parts - no IC's. It's pretty easy today to build a small one with relatively few parts, but you have to look out for the noise that they produce. Here's a link to an application note for one that would work in this case: LM2575 ap note (warning, 26 page pdf) I still like linear regulators for a lot of applications unless efficiency is paramount.
"Do the Right Thing. It will gratify some people and astound the rest." - Mark Twain
"Do the Right Thing. It will gratify some people and astound the rest." - Mark Twain