I created an integrated EAGLE library for the affordable 100V Infineon IRS10752 MOSFET Gate driver (the download from Infineon provides a package and schematic symbol, but not a ‘device’ with the right connections).

Blackmagic Design make some really awesome hardware. With some shitty flaws that means they break on lousy parts like a micro USB-connector.

I recently got a SDI-HDMI converter that only powered on intermittedly, probably because of the micro-USB connector coming loose after a mechanical shock/snag/… accident.

The current version of these is using USB-C, which is a lot more mechanically robust. But still, if you’re using these in a production environment, you’ll probably want to tape down the wires so there is no accidental force on the connector. It’s not really a field-proof power supply connector imho, especially for the types of environment BMD’s customers operate in.

To be fair, there is a plastic block holding the USB-connector down, but mechanically that doesn’t do much. A proper design of the connector’s PCB footprint might have saved this specimen though.

Fault finding 1

Under a microscope, cracks in the soldering of the connector are clearly visible. There’s a nice power LED (D2) next to the connector. Let’s get those cracks fixed up and power it up.

Fault finding 2

Insert power – “pop”! But no light. Hmmm. Maybe the seller lied about it working intermittently? Was it maybe fried altogether? Check all IC’s for visual damage or magic smoke – nothing. No smell either.

Turned out that the “pop”-sound was made by something else completely unrelated. Look closely at the rightmost pin on the connector (next to the white dot – VCC or +5v). Looks good, right?

The real culprit

Turns out that the power for this thing runs through a *tiny* (4mil?) trace which also had been severed clean when the force was applied. Very hard to spot, even under a professional stereo microscope with proper lighting (and experience).

Fix

Mitigation

If you use BMD products (or any devices with micro USB or similarly flimsy connectors) in a rough environment, secure down the cable (e.g. with gaffer tape) near the connector. This way, pulling the cable doesn’t disconnect or break the connector.

One more step could be to put blocks of wood, LEGO or similar material around the connector to create a ‘hole’ for the connector to live in, so any snagging, pulling or falls strains the block in stead of the connector.dwdew

Proper mitigation

If you’re a device designer, PCB layouter, product architect or in any similar position to do something about this during design or scoping of the project – do so. Millions of devices end up in landfill because of simple problems like this – while the world is quite literally on fire.

I’ve fixed tens (maybe a hundred?) devices with this problem. Usually, the time-cost doesn’t match the value of the item (hello Chinese disco-light!). BMD is by no means the only company not doing this right.

Some Tips:

1. Select the most robust connector (type) that’s available and appropriate for the job (I can understand going for micro-USB here because of the versatility, cost, and ubiquitous-ness back when it was designed (2015).
2. Design the enclosure such that the connector receptacle is as recessed as possible. This way, the bulky, ‘hard’ part of the connector plug sticks out as little as possible, lowering the chance of impacts and shortening the ‘arm’ of a moment that is applied to the receptacle in case of a snagging momentum on the cable.
3. Mount the receptacle as sturdy as possible within the constraints of your assembly (soldering) workflow. No tiny traces. No relying (solely) on copper-substrate bonding. Use the board’s thickness and the mechanical resilience of vias to your advantage.
   USB-C receptacles are much better in this respect because of their pins that actually pierce the board in stead of just sitting on top (and relying on the copper bonding).
4. Add extra mechanical anchoring: Add clear copper pads next to a connector to (manually) add bulk solder blobs if necessary. Anchor the receptacle to the enclosure if possible. Add mechanical brackets over a connector (that actually do something – see next point). Hell, break out the hotmelt gun if you’re skimping.
5. Test! TEST! Test the living bejesus out of your prototypes and products. Put yourself in the boots of your users. If it’s meant to be used on a film set, visit those. Products will fall. Cables will be pulled by a stray piece of equipment.
   Put a weight on a typical cable and let it swing while plugged into your device. Let if fall. If the connection breaks, the cable (plug) should break – (and this is important) the receptacle should be fine. Cables are almost free – the device is expensive.

Keep in mind: Circuit board, thin copper traces and solder are all very brittle – meaning there is little displacement required for the mechanical and electrical connections to get separated. Even a tiny tolerance or play (e.g. on a bracket) of say 0,2mm might be enough to reach ‘yield’ conditions in any of those. Make sure they can’t be reached (with a safety margin). That’s why the plastic block holding down the USB-connector inside this product is a non-working afterthought.

But!…

Yes this adds costs. But these are tiny in comparison to the cost of products failing, (film) productions halted or delayed, and the damage to your brand’s reputation. The environment (and your users, co-workers in warranty, stock holders etc. etc.) will thank you.

Sometimes you want your code to be as independent as possible. Arduinos have a hardware PWM-function, but not on every pin. For instance, on the (awesome!) Seeedstudio Grove Beginner Kit, both onboard LEDs (pin 13 and 4) are on pins that don’t have hardware PWM. There is a library to do software PWM, but sometimes you don’t want to use a library (e.g. extra overhead).

Here is an example that simulates PWM using only Arduino native functions:

/*
Analog Input

Demonstrates analog input by reading an analog sensor on analog pin 0 and
turning on and off a light emitting diode(LED) connected to digital pin 13.
The amount of time the LED will be on and off depends on the value obtained
by analogRead().

The circuit:

• potentiometer
  center pin of the potentiometer to the analog input 0
  one side pin (either one) to ground
  the other side pin to +5V
• LED
  anode (long leg) attached to digital output 13
  cathode (short leg) attached to ground
• Note: because most Arduinos have a built-in LED attached to pin 13 on the
  board, the LED is optional. created by David Cuartielles
  modified 30 Aug 2011
  By Tom Igoe This example code is in the public domain. http://www.arduino.cc/en/Tutorial/AnalogInput
  */

int sensorPin = A0; // select the input pin for the potentiometer
int ledPin = 4; // select the pin for the LED
float sensorValue = 0; // variable to store the value coming from the sensor

void setup() {
// declare the ledPin as an OUTPUT:
pinMode(ledPin, OUTPUT);

}

void loop() {
// read the value from the sensor:
sensorValue = analogRead(sensorPin);
// turn the ledPin on
digitalWrite(ledPin, HIGH);
// stop the program for microseconds:
delayMicroseconds(sensorValue);
// turn the ledPin off:
digitalWrite(ledPin, LOW);
// stop the program for for microseconds:
delayMicroseconds(1024-sensorValue);
}

Materialise is an awesome 3D-printing service. In light of the COVID-crisis, they designed a suite of door handles that you could print and attach to doorhandles around your home or office, so people can open doors without touching them with their hands.

However, I found that none of their designs quite fit the most common door handles in use in The Netherlands. Also, one model tended to break easily because of the thin walls. So I adjusted their designs.

Best printed in PETG w/ 100% infill for sturdiest end results.

Quick write-up of steps and info to get the D7000 to output clean 720p HDMI (no menus etc.) for recording or streaming.

You’ll have to download firmware version 1.03 which is available here (same site has a Windows download):

http://www.opendrivers.com/download/driver-166837.html

And put it through an online “patcher” here:

http://simeonpilgrim.com/nikon-patch/nikon-patch.html

Further instructions are available here:

https://nikonhacker.com/wiki/Nikon_Patch

When a tool works, I like to keep it as long as possible. About 8 years ago, I bought a 15″ MacBook Pro, and it has served me well all this time. Apple replaced the logic board a few years ago (under warranty!) because of graphics issues. Since purchasing, it’s had it’s memory upgraded, the hard drive replaced for an SSD and the optical drive swapped out with a hard drive. I replaced the SSD again for a larger 512GB model (Crucial CT512MX100SSD1) when they became cheaper. So far so good.

Recently, a friend gave me a defective MBP, which turned out to be the exact same model (A1286). Great for spares, I thought. Lo and behold, there was a whopping 960GB SSD (A Crucial ct960m500ssd1) inside. “Why not swap that into mine?”

I wish I hadn’t. While the swap was relatively easy, and the computer worked okay, it got a little sluggish over time. I upgraded macOS and it got even slower. Until I got to a point where I couldn’t even send an e-mail without getting the hated macOS “Spinning Pinwheel Of Death” (SPOD). Once every 30 seconds. Sometimes Spotify would even stop playing music, even when I didn’t even touch the machine.

I’ve had this OS install since I bought it. All the development kits, homebrew, libusb-type hacks etc; maybe something is acting up. Let’s try a clean install. Nope, still getting beachballs right after booting a VM. Different SATA cable – nope. Other memory. Oh well.

Must be all the OS upgrades, I thought. Also, I’ve been asking quite a lot from such an old machine, running a lot of programs simultaneously – even some VM’s too. “But it always worked flawlessly!”. I was seriously considering buying a new laptop, even looking at Windows machines because of the price difference. A MacBook Pro with similar storage and memory would’ve cost ±€4000, though.

Then it hit me. The problems started – gradually – about a year ago. About the same time I upgraded to the M500 SSD. Swapped back the MX100, clean install; bingo! VMWare Fusion, running Windows 10 and Windows XP simultaneously, different browsers with 4K video, some file transfers running in the background; no problem.

Since I couldn’t find anything online about this apparent incompatibility, I thought I’d share it here. What makes it even more problematic is that the drive seems to work fine; the OS installs, it boots, etc. You’ll only notice the hangs and freezes after a while. That’s what makes this so insidious.

I don’t know the root cause of the problem, and at this point I don’t care. 512GB ought to be enough for anybody 😉

Inspired by this hack, I decided to make my own version with the intent of making it more stable. Buy 2 cheap Chinese USB soldering irons, add a 3D-print and some tie-raps and you have solder tweezers that normally costs hundreds of Euro’s.

STL File

SolidWorks File

Van de week las ik dit artikel in de Volkskrant:

“Hij heeft het wel ooit geprobeerd”. Hoeveel van die 20.000 stuks zouden op die manier de deur weer uitgedaan zijn?

Dit is geen elektrische auto

I know. You’re hacking and whipping a proof-of-concept up. You’d like to laser cut a baseplate for your favorite dev tool: the BeagleBone Black. Unfortunately, BeagleBone doesn’t provide nice technical drawings like Raspberry Pi does.

I noticed a shift in visitors to my blog; most of them used to come for the cheap chinese laser cutter pages, or the Solidworks Macros. But lately, most of them are coming for the Arduino Uno technical drawing! Since there are no good mechanical drawings for the BBB out there, I decided to whip one up.

Thanks for Logic Supply for the (very detailed!) CAD-model.

If you have a 2011 MacBook Pro and are experiencing WiFi-connection issues, the antenna cables might be broken due to repeated opening and closing of the lid. Mine’s a A1286, but I guess other models have the same issue. Irritatingly, macOS shows “5 bars” of signal when in fact it’s getting none.

After doing some research, I found I wasn’t the only one with this problem.

If you alt-click the WiFi-icon, you get this info screen:

Before replacing, both “RSSI” (signal strength) and “Noise” both were at -80dBm.

Here is the cable, you can clearly see the severed connections.

This could have easily been spotted in a durability-test of the hinge; maybe this only happens when the cables are twisted some way.

Come to think of it, Apple replaced my screen once due to glitches. Could be that the display cable had the same issue?