The KAOS lab recently bought a fleet of five Inspiron 11 3000 3138[PDF Warning] (Celeron N2815/4GB/500GB)laptops. They’re tiny little machines with 8 hour claimed battery lives, they’re pretty cute and sort of obstinate.
I borrowed one to play with, notes:
Continue reading
Some years ago the KAOS lab bought a number of Netgear WNR3500L routers to use as network glue. A pair of them have been in constant service since 2011 in the lab and machine room, and both of those died in the last two weeks. I’ve finally satisfactorily tracked down the culprit, and figure it’s worth a quick write up.
The WNR3500L is a bit of a false-promise machine; it was sold as an “open router” with excellent community firmware support, but it has some pretty extensive blobs that mean it really only runs versions of DD or Tomato with a small range of 2.6 kernels.
I initially just swapped the first one to die out for a spare (which happened to be flashed with Tomato instead of DD, it was an experiment when they were first set up), and reloaded most of the settings by hand (we failed to save or document many of the settings elsewhere. Our fault.) until I had time to investigate. When the second one died, it became urgent.
Conveniently, the WNR3500L has a 3.3V RS-232-like serial port (115200/8/n/1) under an easily removed panel, retained by one small torx screw/pin thing (Both are out in the picture while I was figuring out which side was which).
This leads to another victory for the Bus Pirate. Any 3.3V compatible UART adapter would work, but the bus pirate has much nicer cables for this sort of thing. In retrospect, those headers are correctly sized to just plug the female harness ends directly, but I used clips out of habit.
It turns out they were booting right up with only warnings, but the configuration was sufficiently garbled that neither the WLAN nor the Ethernet ports were coming up. I blew the first one’s settings away (note: at least on the old version these were running, reboot TWICE after clearing the NVRAM to get back to a consistent state: the first time it writes default values, the second time it boots cleanly). With the software-only nature of the problem established, I took more time to investigate the second one, at which point…
nvram show
[1310 lines excluded]
size: 32782 bytes (-14 left)
Well, There’s Your Problem.
Turns out the tiny little log fields for monthly traffic statistics filled up the nvram and were causing configuration corruption, and thus failures.
There are some forum threads and bug reports on the matter, and it was fixed about two years ago with a test to prevent writing past the end of the file, a default setting to automatically delete traff logs after 12 months, and various other enhancements. Now that we’re up and running again it isn’t desperate, but it looks like there is a DD release from mid-2013 with the fixes applied that still supports these things, and they are probably due to be re-flashed to keep this from happening again.
Lessons:
Other notes:
Posting up my notes from SC13 is another thing I didn’t get to during the end of the semester. Remedying now.
The main takeaway sequence from conversations on the floor is as such:
The other really exciting thing that it seems AMD is going to make it, and more. Their lean period finished when the payoff on the XBone/PS4 came in, and they have a VERY good plan for the next >2 years. It works with the premise above about single-core/die MIMD performance ending, and points in the HSA direction – this is the crazy parts with MMUs so a CPU and GPU can share memory without skew penalty and such. ARM and partners are also generally pointed that way, and have been for some time, though apparently AMD isn’t getting out of the x86 game, but it does look like they are getting out of the fat core game.
Continue reading
One of my previous posts about my Singer attracted an email conversation with another owner about the swing-arm mechanism on the no. 42 cabinet. Unfortunately, the end-of-semester insanity struck before the matter was settled, and I am still unsatisfied with what I’ve been able to figure out.
Now that I’ve had a bit of time, I pulled apart my mechanism and took photos, shared below. I’m reasonably certain that if the mechanism is complete and correct, the arm will automatically deploy when the leaf is lifted. Unfortunately, I’m also quite sure that the pieces I do have are inadequate to support that functionality, and I can only guess what the other bits might be.
The larger diameter end of the part I do have matches the diameter of the holes in the hinge, and the smaller-diameter end matches the hole through the swing arm and base.
My best guess is that there are several objects similar to the pictured pin, one of which protrudes below the table into the catch hole of the swing arm through the holes in the hinge mechanism, springloaded “up” such that it retracts when the leaf is out, and is depressed by something protruding from the hinge-hole in the leaf pinning the arm when closed. The pictures sent by the other owner show what looks like the end of a similar pin protruding into the bracket, but it does not extend any where near far enough to retain the arm.
Posted partly in the hopes that my pictures will help other folks with their cabinets, but also if anyone with a no. 42, especially if it has a working swing-arm mechanism or parts that are not pictured, sees this I’d love some more information about how they’re supposed to go together.
In another episode of fixing things for the Cameras as Computing Systems class I’m taking, I made a PKGBUILD that apparently correctly builds and installs chdk-ptp on Archlinux systems. Chdk-ptp is a tethered-control application for Canon cameras running CHDK, that hooks a variety of custom extensions to the ptp protocol. Their build system is a little lackluster, is missing things like an install rule, and requires a helper script be installed to do some path munging before running the binary, but the documentation is good enough to sort it out, and the program itself seems to work. My package depends on iup-all-bin from the AUR which also provides the cd dependency (not marked in its provides array, though there isn’t an official package to conflict with). I tried to use the built-from-source AUR packages for cd (Seriously, who thought that was a good name for a piece of software?) and iup, but iup was giving me a hassle and the chdk-ptp documentation suggests the binary distribution will be less trouble anyway.
The PKGBUILD format has changed a bit since I was last making my own- I like most of the changes in terms of clarity and modularity, but it does require a bit of re-learning. It also means I have a couple of pet packages that probably need attention.
The build is a little janky so despite it passing through namcap with only one expected warning, I don’t want to put it in the AUR until I’ve tested it enough to be reasonably sure it works as intended. I expect I will get around to that eventually.
I gave the discussion-starting talk for this week’s CS departmental BoF session. The topic this week was “Unlimited Internet bandwidth – Would it be a game changer or no big deal?” My opening talk was just to get people up to date on the current events to kick off a discussion, though it is very similar to asking “What will we do with the internet” in the early ’80s, and thus rather difficult to suggest up plausible cases that aren’t “The same thing but faster.”
Callout:
Gigabit to the home is rapidly becoming a reality. Programmable hundred gigabit networks are already being constructed. Cellular networks are improving so fast that people are dumping their wired connections in favor of wireless. In short, networks are becoming so fast that one can begin to imagine a world where bandwidth is essentially unlimited.
But how would the world change if we had (essentially) unlimited bandwidth to everywhere? Would it change anything? Don’t most apps already have all the bandwidth they need? Aren’t networks already fast enough to support “the Cloud”? Have we already max’d out on bandwidth? Are there any super-cool apps that could still be enabled by even faster networks? If so, what are they? Will they be truly radical, or just an incremental improvement?
Come and find out at our next CS Bof, Friday, Nov 1 at 3pm when we will debate whether there are apps that will benefit from even faster network speeds. If there was unlimited bandwidth, what new apps could *you* see emerging? Join us Friday to dream about the possibilities and give your opinion.
In a related note, I’ve made slides for several presentations recently in Beamer, including this one, and I’m pretty convinced I’m never voluntarily using Powerpoint or any obstinate WYSIWYG slide system again. Beamer is a superior tool for the job for every kind of presentation using slides I’ve ever run into.
There was a problem with the Lab’s Makergear M2 recently that is worth writing up.
Last weekend, we had it off site for a demo event for a group of highschoolers interested in engineering and medicine, which included assembling samples of these printable articulated hands that Dr. Dietz designed. (Semi-related: I’ve watched him do it, and continue to be mystified how he does things like this in OpenSCAD). We printed some models live, but on the last print of the trip it made a burning plastic smell. Not the usual, pleasing corn-syurp-y PLA-at-work smell, but acrid recently-deceased-electronics smell. We couldn’t locate the problem, and everything was working, so we assumed filament impurity and moved on.
The next day, the bed stopped heating. Probing around with a DMM indicated the board was fine but the wire harness was not (A thank you to the Ultimachine folks for putting proper test points on the outputs). A bit of googling turned up this thread implicating the quick release screw terminal block that connects the bed leads to the board. Sure enough, upon yanking the connector free, the removable wire-side portion was a melted mass.
The existing connector is a Molex 39533-2002 rear-entry, two lead “Eurostyle” connector. Interestingly, Makergear and/or Ultimachine (Makergear buys and installs RAMBO Boards) used straight-through 5.08mm (200mil, but usually talked about in metric) quick release connectors – screws on the top, wires enter from the right – for the other MOSFET outputs, but a rear-entry (screws on the left, wires enter from the top) for the bed heat. This seemed odd since the side of the little laser-cut enclosure the RAMBO board lives in obscures the screw heads, but the reasoning will become clear. When I was poking at it to see if I could figure out what went wrong, I noticed the threads on the melted side were all but stripped out, but I don’t know if that was a cause or a symptom.
Since it is a standard connector, the (apparently intact) board side will mate with any two-position 3953x part. We found an ebay listing for five 39533-2002 for about $8 shipped, which are compatible and straight-through. Sadly, ebay is usually cheaper and faster than dealing with an electronics vendor if you only need one thing. The replacement seems to work fine, but clearly the original was a rear-entry model to keep the wires from being pressed against the side of the case. The M2’s wire harness is definitely its weak point; two of three serious intervention-requiring stoppages have been related to the bundle headed up to the bed/Y assembly.
The general lesson is to keep an eye on your screw terminals, especially if they are carrying current. I recall replacing the (non-quick-release) bed heater screw terminals on Collexion’s Makerbot ToM after a meltdown not long ago, so there is clearly a general issue with running that much current through screw terminals (several of the common crimped connectors would likely be superior in most ways). Since it came up (and since it is intuition-defying and I’ve heard a lot of people who should know better get it wrong) I’ll close with a reminder that you should not tin your wires before placing them in screw connectors – I know I’ve read this in a standard, research-backed source from ISO or NASA or somesuch, but I can’t find it right now.
ADDENDUM: It cooked two more of the same connector, then we discovered that the SD card connector was a little lose, and apparently causing a ground problem. It hasn’t cooked another connector since, and the bed heats faster than it was, so strong evidence that that was the root cause.
While I’m writing up things I’ve done recently, I finished this year’s edition of the Jonathan Strahan edited The Best Science Fiction and Fantasy of the Year collection. As in previous years I’ll mention the high points.
Continue reading
I wrote about this machine once before a few years ago, but only brought it down to Lexington to work on it at the beginning of July. I’ve had a delightful time cleaning restoring it over the last couple weeks, and just wanted to post some pictures and musings. I did have it correctly identified before – it is a 1947 Singer 201-2, in good mechanical and OK cosmetic condition. The machine’s story from the family has settled on it being my great grandmother’s machine down the matrilineal line, but I don’t know if they were the original owner, or what exactly has happened to it over the last couple decades. My grandmother noted that she remembered her mother doing upholstery work on it, and my mother remembers using it as a child, and it was in my grandmother’s basement three years ago. Descriptions follow pictures below.
Continue reading
