So, while learning how to use the auto-bracketing on my new camera, I decided to shoot the Koi Pond my uncle has, and then try an HDR stack. After that I played with the colors a bit, and I think it came out pretty nice, but I hate that fence in the background…. either way though, take a look:

And I promise, I’m still working on that write up for the balloon!

So, recently I took another tour at Brookhaven National Laboratory, this one was of the RHIC (Relativistic Heavy Ion Collider) facility at the laboratory. In a nutshell, they take gold atoms (ka-ching!) and make them zoom around in circles at 99.995% the speed of light, and then, wait for it….. they collide them!

Bet you didn’t see that coming!

The collision causes a sudden burst of massive heat, which causes the gold matter to melt down to it’s most basic components (quarks, gluons, stuff like that) and gives scientists a view of what the universe might have been like in the first few instants after the big bang. It’s really, really, really cool!

This experiment is such a big deal, it is made up of parts all over the laboratory site.

Basically, the process start at TANDEMS, middle center of the above map, where the gold atom is stripped of most of its electrons (making it an ion) and accelerated part of the way, up the stream to EBIS, where it’s accelerated further, and flung in to AGS, where the remaining electrons are knocked off and it is further accelerated, before finally getting passed up to the RHIC, where it is accelerated up to 99.995% the speed of light, I say this again because that is really fricking fast! So fast in fact, you can’t measure it in speed anymore, but in terms of how much energy the whirling particles have. Can you tell I’m impressed?

While the particle (actually, a cloud of particles) whirls around the RHIC really, really fast, another particle (cloud) whirls around it in the opposite direction. This is handled inside two pipes that run the whole length of the RHIC tunnel under ground at the laboratory. Just for reference, the RHIC tunnel is underground, but the support structures and clear path above it make the whole site not only visible, but obvious from space. Check out the Google Maps snapshot below!

View Larger Map

And under that big ring, it looks like this:

So that is the RHIC in a (slightly jumbo) nutshell.

Also, I’m still working on my write-up on the High Altitude Balloon Project, but between work and my crazy social life (for once, that’s not a joke) I haven’t been able to finish it. It is taking a while because it will be quite in-depth and photo-filled. As an olive branch to my impatient fans out there (I think I see one of you, maybe two?) I’m going to share one of the photos I am working on in preparation for this write-up. It’s not ready, and I may not even use it…. but here goes:

The photos in this post are all from my new Canon Rebel T2i camera, which I purchased recently and am still learning how to use. Expect to see more photos here though, as I really enjoy using it, and soon enough I will be at Maker Faire New York 2011 with it!

So, as you’ve seen below, my friends and I have built and launched a high altitude balloon. This was a continuation of the stalled balloon project I mentioned last year. The balloon pack was completed, and launched successfully. A few minutes in to the flight, we lost all telemetry very abruptly from the balloon, and we were not able to locate the landing zone. While this was unfortunate, the project was still overall a success in my eyes. The system we designed functioned well in all preflight tests, and the leading theory is that the solar power system was entirely too effective, creating an over-voltage condition that fried the sensor pack, the GPS or the radio transmitter module, or some combination of those components.

More details can be found at the balloon project page which I will post a link to as soon as it is complete.

The other topic, today I visited Brookhaven National Laboratory for one of their ‘Summer Sundays‘ events, this one as the title says was the tour of the National Synchrotron Light Source, also knows as the NSLS, and it’s younger brother, the NSLS-II which is still under construction. The tour involved an open house at the construction site with several exhibits, and a tour of the parts of the actual storage ring structure that is complete, plus a close up look at one of the sextupole magnets used to guide the electron beam in it’s circular journey. I was accompanied by two of my friends on this journey, Tom and Matttew, and we were quite… inquisitive. I have to tip my virtual hat to the tour guides, they were very friendly and welcoming of our questions and answered them in great depth, while being pretty good about making it understandable for us. I snapped quite a few photos of that, and while most of them are sitting in post processing, I have attached a few here to this post as a preview.

If you’ve come to my website because you located a downed weather balloon package in upstate New York, please contact me immediately at admin@mattkrass.com. the pack was launched this past Sunday and has stopped transmitting its location to us. If returned there will be a reward.

More details to follow, for now, this ambiguous overly photoshopped preview!

(I got bored and had a camera?)

Make special note of the awesome arrangement of resistors in the bottom right area.

I just added another guide I threw together tonight on Potentiometers. The guide is specifically geared towards FIRST Robotics Teams, but if there is interest I can expand on it to be more generic.

It’s available here, and as always, questions, comments, etc are welcome!

This was fun. Took a bunch of photos with my PowerShot of my flight system setup on my disheveled desk with the room lights off, fixed aperture, and progressively slower shutters, and used Photoshops Merge to HDR Pro command with a Surreal preset to make this:

So, if you recall a couple weeks ago I was helping a friend try to revive his Fonera 2100 router. New Years Eve and Day consisted of our next burst of efforts in recovering the little router and I’m happy to report much progress, and irritated to report another brick wall.

To recap (sort of, since I never actually explained this) we got the JTAG interface working with a Fedora 13 live CD (Windows and my parallel port were having a bit of a marital snit and I was too cheap for counseling, the live CD worked fine) and we were able to access the flash on the device. In doing so we confirmed the bootloader was toast. So this time we decided to try and transplant the bootloader from another Fonera 2100 my friend had in active service. Accessing the MTD devices in the little tiny linux build it was running allowed us to make an image of the RedBoot bootloader and transfer it to my desktop computer. From there, we used the live CD and JTAG cable to write that to the appropriate portion of the flash memory on the router.

This. took. forever.

192KB of data. 2 hours, 1 minute, 35 seconds to write it all. This is how I rang in 2011

But it was worth it, when I got back to it New Years Afternoon (after sleeping in, then going for a rousing lunch with friends) it was ready to go. I hooked it up to my FTDI breakout board with a truly scary rig of alligator clips and jumper wires around the JTAG hookup, and booted it up. After a few moments we had RedBoot starting up, complaining about the royal disaster the flash was in, and insisting I regenerate its settings. At this point, the live CD crapped itself, and we switched back to Windows, no longer needing JTAG capabilities.

Once in Windows we used PuTTY to access the serial port and configured the bootloader to bring up the network interface on a static IP, and then we were able to telnet in to the router as well. At this point, we were celebrating, figuring we’d just TFTP a new firmware image in place, burn it to the flash and start it up.

Yeah, not so much. This is the brick wall we ran in to. For some reason, every version of DD-WRT or OpenWRT we tried on the Fon started up fine, only to slam headfirst in to a violent kernel crash and lock up. At this point, we believe we did something wrong when we recovered the bootloader and perhaps the memory locations we’re working with are not valid because of it. The next thing we’re planning on trying is figuring out the checksums on the written flash, and if that is not helpful, a complete reimage of the entire router, as a clone of the working one. This is a bit of a brute force approach, and itwill take a while, but it should definitely work, and confirm there was no hardware damage to this router.

So, just to summarize:

• We have restored the RedBoot bootloader to the router and gotten it working over serial and telnet.
• We have (presumably) successfully written full flash images to the router via RedBoot/TFTP, but they don’t work. At this point we’re not 100% sure it’s transferring correctly, but there doesn’t seem to be a problem with the transfer, we’ll get checksum data working soon to confirm.
• The routers network and serial hardware interfaces are completely operational as far as we can tell, and we used both extensively to try and restore the firmware.
• The problem with the kernel startup is prevalent in multiple builds of DD-WRT and OpenWRT, meaning they’re unlikely to be the (only) culprit

Next steps

• Figure out how to compute checksums and verify the images are all valid and writing properly
• If that checks out, clone the working router
• If not… figure out why!

Stay tuned. I’ve made it a personal mission to get this router working now!

When I was high school all those years ago…. (ok not that long ago) I used to be a big fan of Warhammer. My school had a gaming club that played weekly. I was so intrigued I went out and bought a starter set, which came with models for an Orcs & Goblins army, as well as an Empire (humans) army. I started painting them, then got busy with robotics and life and never finished. The local gaming place also closed up, so I had no incentive to finish the models because I had no where to bring them. Well I’ve decided since I’ll have money to invest in a hobby and I’ve been dying to get back to tabletop gaming, finishing these models is a priority again. It’s been a fun experience remembering how to paint. I was never that good to begin with, but I think I might be getting better this time, after so many years of carefully wielding probes and soldering irons around circuits. I’ve also gotten a bit more clever with color blending. Hopefully you’ll be seeing lots of my models from now on. For now here is a preview of an experimental unit I’m painting, trying out some things that I think are coming out pretty good, especially with regards to the mouth and the bloody horns.

Also, note I have resumed work on the guides section of this site, and I’ll have a new guide up shortly!

Sorry for the lapse, since graduation I’ve been getting buried under professional projects, and I’ve been picking up any work I can to get by while I searched for the perfect job.

Well for starters, I’ve found the job! I was made an offer by a local company near my home and I’ll be starting work there soon. This means more time for personal projects and website updates. The projects and guides section both have updates pending I just need to finish proofing them and they’ll be ready to go.

As far as personal projects, not many lately, but that should change. The only things on deck right now is an attempt to use JTAG, serial ports and some magic to transplant firmware from a working router to one that is completely and utterly bricked. I’m also helping my friend put together a little GPS Speedometer/Heading display for his Jetta. Actually, the router transplant is for the same friend… he’s really good at breaking stuff.

And to prove I’ve been doing stuff, here are some photos of the little router with its guts exposed and wired to my JTAG cable:

Also, check out my new logo…. yeah it sucks a bit. But I’m sure it’ll get better with a few more tries…