New video! Repairing traces and vias and then finally testing the Amiga 3000T board for the first time.
YouTube: https://youtu.be/_DvfH7xoIy0
PeerTube: https://makertube.net/w/wppRk4kX71JCihswoL2PED
#Amiga #Amiga3000T #Amiga3000 #A3000T #Commodore #Restoration #BatteryLeakage #Corrosion #Repair #FPU #SolderMask #RetroComputing #VintageComputing #RareCommodore
The Best DIY PCB Method? https://hackaday.com/2024/06/23/the-best-diy-pcb-method/ #soldermask #PCBHacks #etching #diypcb #laser
Pick and Place Hack Chat Reveals Assembly Secrets
These days we've got powerful free tools to do CAD and circuit design, cheap desktop 3D printers that can knock out bespoke enclosures, and convenient services that will spin up a stack of your PCBs and send them hurtling towards your front door for far less than anyone could have imagined. In short, if you want to build your own professional-looking gadgets, the only limit is your time and ambition. Well, assuming you only want to build a few of them, anyway.
Once you start adding some zeros to the number of units you're looking to produce, hand assembling PCBs quickly becomes a non-starter. Enter the pick and place machine. This wonder of modern technology can drop all those microscopic components on your board in a fraction of the time it would take a human, and never needs to take a bathroom break. This week Chris Denney stopped by the Hack Chat to talk about these incredible machines and all the minutiae of turning your circuit board design into a finished product.
Chris is the Chief Technology Officer (CTO) of Worthington Assembly, a quick turn electronics manufacturer in South Deerfield, Massachusetts that has been building and shipping custom circuit boards since 1974. He knows a thing or two about PCB production, and looking to help junior and mid-level engineers create easier to manufacture designs, he started the "Pick, Place, Podcast" when COVID hit and in-person tours of the facility were no longer possible. Now he says he can tell when a board comes from a regular listener by how many of his tips make it into the design.
So what should you be doing to make sure your board assembly goes as smoothly as possible? Chris says a lot of it is pretty common sense stuff, like including clear polarity indicators, having a legible silkscreen, and the use of fiducial markers. But some of the tips might come as something of a surprise, such as his advice to stick with the classic green solder mask. While modern board houses might let you select from a rainbow of colors, the fact is that green is what most equipment has been historically designed to work with.
That black PCB might look slick, but can confuse older pick and place machines or conveyors which were designed with the reflectivity of the classic green PCB in mind. It also makes automated optical inspection (AOI) much more difficult, especially with smaller component packages. That said, other colors such as white and red are less of a problem and often just require some fine tuning of the equipment.
He also pulled back the curtain a bit on how the contract manufacturing (CM) world works. While many might have the impression that the PCB game has moved overseas, Chris says orders of less than 10,000 units are still largely handheld by domestic CMs to minimize turnaround time. He also notes that many assembly houses are supported almost entirely by a few key accounts, so while they may be juggling 50 customers, there's usually just two or three "big fish" that provide 80% of their business. With such a tight-knit group, he cautions CMs can be a bit selective; so if a customer is difficult to work with they can easily find themselves on the short end of the stick.
While the Hack Chat is officially only scheduled for an hour, Chris hung out for closer to three, chatting with community members about everything and anything to do with electronic design and production. His knowledge and passion for the subject was readily apparent, and we're glad he was able to make time in his schedule to join us.
The Hack Chat is a weekly online chat session hosted by leading experts from all corners of the hardware hacking universe. It 's a great way for hackers connect in a fun and informal way, but if you can't make it live, these overview posts as well as the transcripts posted to Hackaday.io make sure you don't miss out.
#hackadaycolumns #designforassembly #designformanufacturing #hackchat #pickandplace #soldermask
Investigating a Defective USB Power Bank Module
Call us old fashioned, but we feel like when you buy a piece of hardware, the thing should actually function. Now don't get us wrong, like most of you, we're willing to put up with the occasional dud so long as the price is right. But when something you just bought is so screwed up internally that there's no chance it ever could have ever worked in the first place, that's a very different story.
Unfortunately, that's exactly what [Majenko] discovered when he tried out one of the USB-C power bank modules he recently ordered. The seemed to charge the battery well enough, but when he plugged a device into the USB output, he got nothing. We don't mean just a low voltage either, probing with his meter, he became increasingly convinced that the 5 V pin on the module's IP5306 chip literally wasn't connected to anything.
Curious to know what had gone wrong, he removed all the components from the board and started sanding off the solder mask. With the copper exposed, his suspicions were confirmed. While they did route a trace from the chip to the via that would take the 5 V output the other side of the board, it wasn't actually connected.
This is a pretty blatant bug to get left in the board, but to be fair, something similar has happened at least once or twice to pretty much everyone who's ever designed their own PCB. Then again, those people didn't leave said flaw in a commercially released module…
At any rate, [Majenko] just needed to solder a jumper on one of the non-sanded boards, and everything should be good to go. Well, not exactly. Adding a jumper got some power heading to the output side of the module, but only 3 V. Clearly, something else was wrong in the circuit. After some additional research and a few [BigClive] videos, he realized that some components were actually missing. Consulting the application schematic from the datasheet, it was clear that a capacitor and resistor had been left off. While the absent capacitor didn't seem like it would be enough to cause the failure he was seeing, the resistor is supposed to be connected to the battery sense pin, and could explain why the chip wasn't boosting the voltage.
IP5306 application schematic, missing components highlighted.
With the resistor in place, the module started working as expected. [Majenko] notes that he still can't get more than 1.5 amps out of these boards that are supposedly good for 2 A, but that's another story entirely. With so many problems, it seems likely this module was a failed prototype and never meant to actually go on sale. But with part shortages ravaging the industry, it seems even the broken modules are getting pushed out the door.
Incidentally this isn't the first time [Majenko] has tracked down some errant components in a supposedly turn-key board. Earlier this year he noticed that tweaking some of the parts used in the Raspberry Pi 4's power supply helped reduce low-voltage warnings.
#parts #repairhacks #applicationschematic #boostconverter #datasheet #ip5306 #soldermask #usbpowerbank
