Newer posts are loading.
You are at the newest post.
Click here to check if anything new just came in.

September 05 2017

The US-Mexico Border Wall Actually Juts Into the Ocean—Where Surfers Staged a Glorious Protest

August 21 2017

Sorry, Atlantis: Eden’s Achin’ Organ Seeks Revenge

August 18 2017

LA artists take a symbolic surf across US-Mexico border

August 15 2017

Surfers on the US/Mexico Border are Making a Subtle Statement About the Absurdity of Borders
Echo Park Lake Becomes Home To Fiesta Perpetua

July 28 2017

Speculative Email Generator

July 27 2017

Sci-Fi Cyanotype

July 19 2017

Surf Border

June 26 2017

Sousveillance & Raspberry Pi

June 14 2017

TOG @ Coolest Projects

TOG will be at CoderDojo Coolest Projects Showcase 2017 this coming Saturday 17th June in the RDS. We’ll be showcasing some of our own projects and running some workshops on Bridge Building, Soldering and Lock Picking. We will be in the main hall – central area.

From the Coolest Projects website….. “Come to the RDS, Dublin on Saturday 17th June 2017 to see the coolest projects from CoderDojo kids, aged 7 – 17, around the world showcasing their vision and creativity using technology to make the world a better place. Tickets for the CoderDojo Coolest Projects Showcase are now available at http://tickets.coolestprojects.org/ CoderDojo Coolest Projects is a great day of tech exploration and fun for the whole family.”

Visitor project: Dominion storage solution
NYCR Members Kari Love and Matthew Borgatti teaching Soft Robotics and Bioinspiration at ITP Camp

June 13 2017

Sonder-OpenChaos am Samstag, 21. Mai: Revision Nachlese

Wie seit Jahren üblich traf sich die Demoszene zu Ostern wieder auf der Revision um sich in verschiedenen Wettbewerbskategorien zu messen. Wieder gibt es die besten Werke beim Sonder-OpenChaos zu sehen.

TOG Mid-Summer Open Social

Our June Open Social evening takes place on Saturday 17th May from 7PM. That’s about as close to the longest day of the year that we can manage! Dublin Maker is approaching fast too. Our members are starting to get a bit edgy. Drop in, poke them a bit, and ask them if they have their projects all ready to go 🙂

If you’ve never been to our Social, you’ll find it a great alternative Saturday evening in town. Come in and look around. If you’ve never been in before, we’ll give you the grand tour of the space. Talk to members and visitors about projects or things you’d like to do. If you like what you see, ask about joining as a full member. Hopefully there’ll be caint, ceol agus craic as usual. The space stays open until the last member is left……usually the small hours of Sunday morning.

Our Open Socials are always free to attend for both members and visitors alike. No need to book…. just turn up. You can drop in for 10 minutes, or stay the whole night. We have parking available. Bring beer, food, gadgets! Our doors are open from 7PM. Hope to see you there.

https://www.meetup.com/Tog-Dublin-Hackerspace/
https://lists.tog.ie/mailman/listinfo/tog
https://twitter.com/TOG_Dublin
https://www.facebook.com/togdublin

June 11 2017

Next two weeks at Tog

Hello all, we have another fortnight of our regular events coming up, along with an Open Social on Saturday 17th.

Monday June 12th – Coding Group Night from 19:00
Tuesday June 13th – Lockpicking Group Night from 19:30
Wednesday June 14th – Crafts Night from 19:00
Saturday June 17th – Join us at our Open Social from 19:00
Monday June 19th – Electronics/Microcontroller night is on from 19:00, as is our CAD & 3D Modelling night also from 19:00

June 07 2017

Milwaukee Vice!

This is our Post Vice at the Makerspace. It is a specialized type of vice used by blacksmiths. Designed to handle the abuse of clamping something very hot and heavy and allowing to to beat on it with a hammer. These are not really mass produced anymore, so when you outfit a forge you typically end up buying a used vice. Since these were so well built they basically last forever. Ours is somewhere around 100 to 150 years old.

Most of the dents you see were there when I bought the vice 4 years ago. However, the slices running near the top, left face are new damage, and go considerably deeper than the surrounding dents. I was concerned about the concentration of stress at those points and the potential for cracking the jaws. It was time for refurbishment and repair.

In this photo, 2 recent dents that were caused by someone missing with a heavy hammer. And you can see where the jaws of the vice have cracked away. The vice is constructed using some kind of iron, but due to the age, we’re not exactly sure what. The jaws style of construction indicates that they were forged, not cast. This is a good thing, it suggests they could be either steel (exact carbon content would be unknown) or wrought iron. It means it should be possible to weld new steel onto them! This is what we’re going to try to do.

We don’t know exactly what alloying elements are mixed into the iron, so it could weld smooth, or it could blow bubbles and burn up under the torch! No way to tell until you try, which is nerve wracking when the object in question is an antique. There are things you can do to tilt the odds in your favor. When welding, cleanliness is next to godliness. Much time was spent with a file and dremal to grind out every dent and crack, to clean and shiny metal. 

This is a shot after I finished welding up the slices. This is about the best result we could hope for. The welds are strong and clean, with only a few spots of porosity towards the face of the jaws. The way to fix the jaws is to deliberately weld on excess material and then grind them smooth again. They will be good as new.

After I finished welding almost all the spots, this is a shot of me filling in the missing corner of the front jaw. In this photo you can see that this is not the first time this vice has been repaired. Someone welded a new plate to the face of the back jaw, but the welds around the edges were quite corroded, so I ground them out to be filled in.

The front jaw is finished and ground, now all the welding is done on the back jaw.

And here are both jaws clean and ground. 

Such straight and sharp corners! Just waiting to get a new set of “beauty marks” in the service of our forge!

Reposted byAndiDIYmushu

c-base hack weekend, July 7-9

We’d like to announce a c-base hack weekend with the theme „reprogramming the space station„.

This event is around the c-flo project where we’re connecting all the different c-base systems on c-beam MQTT, editable using the Flowhub visual editor.

Focus areas:

  • MsgFlo IoT and new functionality — connecting internet services with MsgFlo, adding new smarts to the space station IoT setup.
    Skills: Python, Node.js, Rust
  • Hardware hacking — connecting more devices with MsgFlo.
    Skills: microcontroller programming, electronics
  • Information displays — new infoscreen designs, data visualizations.
    Skills: web design, React.js, Django
  • Mobile app — bringing the hackerspace IoT functionality to mobile.
    Skills: Android (maybe iOS too?)
  • Woodworking — new cases, mounts, decorations for various systems.
    Skills: woodworking, painting

You don’t have to an expert to participate. We’ll be there to help you get up to speed!
Add yourself to the Facebook event.

c-flo in action, detecting ABBA and setting disco mode

Read more:

June 06 2017

The Four Fiscal Quarters

June 05 2017

Welder repair

An ancient proverb says that power tool failures occur in threes. During the same week that the lathe motor burned out and the compressor wire meltdown has turned it into a scary noise machine, we also had a mysterious malfunction in one of our welders. It didn’t release any smoke, didn’t make weird noises. It just didn’t weld. 0 volts across the output.

Inside of the welder (top view) before the repair attempt. Note that a large number of similar – but not identical – top boards can be found on Aliexpress.

This machine, a small 180A TIG/MMA inverter welder, has recently been repaired. It was not used heavily, as TOG has only so much use for welding. In fact, it was only used by a couple members to practise their welding skills.

On one hand, this machine was 5 year old, has been used extensively in the past and could have been written off as death of old age. The cost of a commercial repair would probably exceed the value of the welder. On the other hand – why not try to repair it ourselves before scrapping it?

So, first pass – take it apart and look for obvious damage, like blown up MOSFETs or rectifier diodes. Nope, nothing there. All power resistors look intact, too. All cables securely plugged in. No charred FR4. Nothing clearly damaged or even suspiciously-looking.

What next? Seems hopeless. First, we don’t know how an inverter welder works exactly. Second, we don’t want any deaths, either from electrocution or any other reasons.

The safety problem was somewhat addressed by setting up all the probing while device was unpowered – both in terms of mains and its storage capacitors. The device was only powered for a few seconds to read the measurements off the screen, without touching anything. All measurement instruments were battery-operated, to avoid unintended coupling through protective earth in mains-operated instruments. This approach doesn’t really avoid all dangers, but at least addresses the well-known risks.

The lack of knowledge issue is solved by a rabid and desperate search for any possible schematics and repair hints, from Chinese search engines to Polish electronics repair forums and, of course, various chip manufacturer datasheets. Turns out that many of the MMA and TIG machines from the last decade are variants of the same 3-board design (+daughterboards). We haven’t found the schematic for the machine in question but we didn’t really need it – a PDF schematic of a 160A MMA welder found using Chinese image search was very helpful in identifying major blocks and their connections. A TIG machine obviously contains more logic to handle the TIG functionality (gas valve, pulsing, HF/HV arc start etc.) but those were ruled out as a likely root cause.

All of those welders have a common set of functional units. A primary-side rectifier on a lower board rectifies the 230V mains voltage and provides slow-start functionality to prevent power surges. A power MOSFET module on the top board does all the switching work. Then there’s a set of transformers and secondary rectifier diodes on the middle board that provides output DC voltage for welding, with help of some power chokes to filter out the switching AC component. The whole switching process is controlled by a drive module on a vertical daughterboard, which in turn feeds a H-bridge of small MOSFETs that feeds a transformer that provides gate voltages with correct polarity for all the power transistors. Sounds complicated, but at least everything is nicely divided into modules with a well-defined purpose each. The drive circuit is powered by an auxiliary 24V power supply that turned out to be one of the known weak spot in these units, according to some forum posts.

So, the investigation first looked at the output voltage from the power MOSFETs. There was none. Input voltage for that block – present. So, it will likely be something between the primary rectifier and the output from the MOSFET module. Broken MOSFETs? Nope, they measure fine. Is there a gate signal? Nope. Not on the power MOSFETs themselves, and not even in the H-bridge that is driving the gate-drive transformer. So, perhaps it’s the drive module. After a few hours of connecting alligator clips to various points (heavily coated in non-conducting lacquer, by the way) it was apparent that the SG3525A that was supposed to generate signal that (eventually) drives the power MOSFETs was shut down by one of three protection circuits. There are several: overheating protection (based on temperature switch) – certainly not an issue here. Overcurrent protection – also not to blame here. The one that turned out to be responsible was the undervoltage protection for the 24V supply. Why is too low a voltage a problem? because keeping switching MOSFETs in active region instead of saturation makes them overheat and fail very quickly – and insufficient gate drive prevents saturation. And, bingo – the aux supply voltage was definitely below the usual 24V. The actual value depended on which modules have been left connected – with TIG timing board connected it dropped to almost 12V! Without it was still 19V or so – so, it was definitely a wimpy power supply issue, not, say, a short on the TIG timing board. Trying the welder with 24V supplied from outside (via isolated lab supply) confirmed no other faults – the machine seemed fully working with that setup.

The 24V supply is a flyback converter, based – in this case on a UC3843 PWM chip. In other welders, it may be a bunch of discrete components instead. The converter is using a 200:33:33:16 transformer. One secondary winding used to provide the output voltage, which is then rectified and compared to 24V threshold, with feedback given via an optoisolator. Another, smaller secondary is used to power the PWM chip itself, with help of a standard rectifier diode and a filtering capacitor – although the initial voltage is provided with a high-value high-wattage resistor directly from 310V. There’s also a current sense on the primary that limits the current going through it to 1V/2.2ohm =~ 0.45A.  The initial suspicion was that the transformer had burned windings which didn’t provide enough current. However, this turned out to be incorrect. The transformer looked perfectly, and replacing it “just in case” would involve waiting and risking damaging the board, already damaged a bit by a careless replacement of a perfectly well-behaved filtering capacitor.

A careful observation with an oscilloscope revealed that the voltage on the Vcc of the UC3843 was drooping as the chip was outputting PWM pulses, getting below the undervoltage lockout of the UC3843, then recovering slowly after some time. So, that small secondary winding wasn’t providing enough power through the diode. But why? Bad transformer? The diode itself tested perfect.

What ended up being the root cause was a broken track on the PCB, between the diode and the rest of the Vcc-related circuitry. So, the PWM chip was only powered through a high-value resistor and running out of power quickly as it started generating the PWM pulses, without getting any extra power via the aux secondary. This caused a weird “a bunch of pulses, then silence, then the cycle repeats” behaviour seen on the oscilloscope, correlated with variations in chip’s Vcc. A piece of wire to mimic the function of the broken trace fixed the 24V output and allowed the rest of the machine to work correctly.

NERP Tonite: MOSFETs with Ste!

Mosfets – They can be “on” — They can be “off” — They can even be in between! Tonight at NERP, entrepreneur, engineer, and really good teacher Ste Kulov will guide us into the world of mosfets. Metal Oxide Semiconductor Field Effect Transistors if you want to impress your friends.

N-Channel MOSFET [wikipedia]

Since mosfets are a wide and deep subject, I asked Ste where he wanted to focus his talk. With great economy of words, he said it better than I could.

“Most of the generic stuff, fairly quickly.  A few simple examples I was probably going to cover are: making a logic inverter, a logic controlled load-switch, and reverse battery protection. Simulating in LTspice [circuit CAD], since I can draw that stuff in two seconds and put it on the screen.  If you want to do a power MOSFET application, I would need to see the datasheet for it.  Also keep in mind that high current stuff is no good for breadboards. If you need a list:  4 vs 3 terminal, body diode, Rds(on), gate drive, switching speed, N-channel, P-channel, CMOS digital logic, CMOS analog switches.”

NERP is not exclusively Raspberry Pi, the small computer and embedded systems interest group at Pumping Station:One in Chicago. NERP meets every other Monday at 7pm at Pumping Station:One, 3519 N. Elston Ave. in Chicago. Find NERP and Pumping Station:One at

www.meetup.com/NERP-Not-Exclusively-Raspberry-Pi/­
and
http://pumpingstationone.org/­

Tags: electronics, embedded, NERP, Open Source, raspberry pi, hackerspace, Beagle Bone, Pumping Station One

Older posts are this way If this message doesn't go away, click anywhere on the page to continue loading posts.
Could not load more posts
Maybe Soup is currently being updated? I'll try again automatically in a few seconds...
Just a second, loading more posts...
You've reached the end.

Don't be the product, buy the product!

Schweinderl