Oh, boy. You know what’s happening next weekend? The Midwest RepRap Festival. The greatest 3D printing festival on the planet is going down next Friday afternoon until Sunday afternoon in beautiful Goshen, Indiana. Why should you go? Check this one out. To recap from last year, E3D released a new extruder, open source filaments will be a thing, true color filament printing in CMYKW is awesome, and we got the world’s first look at the infinite build volume printer. This year, The Part Daddy, a 20-foot-tall delta bot will be there once again. It’s awesome and you should come.
We launched the 2018 Hackaday Prize this week. Why should you care? Because we’re giving away $200,000 in prizes. There are five challenges: the Open Hardware Design Challenge, Robotics Module, Power Harvesting, Human-Computer Interface, and Musical Instrument Challenge. That last one is something I’m especially interested in for one very specific reason. This is a guitorgan.
Building a computer soon? Buy your SSD now. Someone fell asleep on the e-stop at a Samsung fab, and now 3.5% of global NAND production for March has been lost.
Need to put an Arduino in the cloud? Here’s a shield for that. It’s a shield for SIMCom’s SIM7000-series module, providing LTE for a microcontroller. Why would you ever need this? Because 2G is dead, for various values of ‘dead’. 3G is eventually going to go the same way.
A bridge collapsed in Florida this week. A pedestrian walkway at Florida International University collapsed this week, killing several. The engineering efforts are still underway to determine the cause of the accident, but some guy from Canukistan posted a pair of informative videos discussing I-beams and pre-tensioned concrete. It’s going to be months until the fault (and responsibility) will be determined, but until then we have the best footage yet of this collapse. It’s dash cam footage from a truck that rolled up to the red light just before the collapse. This is one that’s going to go down in engineering history along with the Hyatt Regency collapse.
Need to test your app? Here’s a delta robot designed for phones. You would be shocked at how popular this robot is.
Really like the videos AvE has been doing about the bridge collapse. I know his delivery can put some people off, but he’s a sharp guy and can speak from experience on many subjects.
Was a little surprised to see some comments saying his videos were “too soon”, or capitalizing on a tragedy. But, you know. Internet.
Anyone know if I can enter the Hackaday Prize contest as a group with some of my friends?
yes.
“Building a computer soon? Buy your SSD now. Someone fell asleep on the e-stop at a Samsung fab, and now 3.5% of global NAND production for March has been lost.”
Didn’t see anything about someone falling asleep, but still there should be backups preventing such a failure.
I’ve seen backup systems fail, where over time the power usage creeps up, but no downtime is given test and if an upgrade is required for the legacy redundancy system design (maintenance downtime is treated by many as money lost). Operations never notified site services of incremental new equipment installs. So the live system works, the switch over UPS battery banks works, the backup generator system works (which takes 5 minutes cold start to be fully online), but the fully loaded live switching between the three is under rated (or over loaded) and melts. Oversight, or a failure of communication, or even worse of all not in my job description.
Oh and employees hear downtime and instantly power down 95% of their equipment, so a failover power test is not usually the same electrical load as in the middle of a typical work-shift. So even the testing during a downtime can work when they should fail. The only way is continual re-evaluation of everything and tracking of all changes no matter how trivial.
I’ve heard about a fab catching fire almost in the same way:
– Main power outage, switching to the backup sources for turning off all the machines safely,
– Evacuation of the clean room due to no HVAC and fumes extraction
– Restarting the system after the power outage, but failing to notice that a part of the cooling system doesn’t work
– Some nasty chemicals catch fire within 10-20 minutes, emergency stop and trigg
– The fumes, and the fire brigade pollutes all the clean room and do a fair amount of damage to the equipment
That led to 2 months of downtime and quite a huge loss…
Apart from being expensive, the backups themselves tend to fail sometimes and nothing apart from actually switching to them can’t really test their condition.
BUT – because that test could lead to a power outage, nobody does that. Instead they use artificial tests…
Also, an on-line UPS (the only kind that will resist failing when the power quickly goes out and back on several times) wastes several % of all the power flowing trough it, that tends to get VERY expensive just to run (maintenance costs are whole different monster) when you need to back up several megawatts of power.
The power consumption of the fab is ~100 MW. They probably do not have the capability to do anything more than shut down the machines, resulting in the loss of any wafers in those machines. Its actualy pretty suprising that there wasn’t any equipment damage,
Wow, $400 for 3servos and frame? I think should revise our price or go to tindie ;)
http://www.ebay.com/itm/173219842411