A friend of ours here at Hackaday has an audacious design in the works that we hope will one day become a prototype that we can feature here. That day may be a little while coming though, because it has somewhere close to a thousand of the smaller SMD components in multiple repeated blocks on a modestly sized board, and his quote from a Chinese board house for assembly is eye-watering. He lacks a pick-and-place machine of his own, and unsurprisingly the idea of doing the job by hand is a little daunting.
We can certainly feel his pain, for in the past we’ve been there. The job described in the linked article had a similar number of components with much more variety and on a much larger board, but still took two experienced engineers all day and into the night to populate. The solder paste had started to spread by the end, morphing from clearly defined blocks to an indistinct mush often covering more than one pad. Our eyes meanwhile were somewhat fatigued by the experience, and it’s not something any sane person would wish to repeat.
Mulling over our friend’s board and comparing it with the experience related above, are we on the edge of what is possible with hand pick-and-place, or should we be working at the next level? Board assembly is a finely judged matter of economics at a commercial level, but when at a one-off personal construction level the option of paying for assembly just isn’t there, is there a practical limit to the scale of the task? Where do you, our readers, draw the line? We’d love to hear your views.
Meanwhile our friend’s audacious project is still shrouded in a bit of secrecy, but we’ll continue to encourage him to show it to the world. It’s not often that you look at a circuit diagram and think “I wish I’d thought of that!”, but from what we’ve seen this fits the category. If he pulls it off then we’ll bring you the result.
We’d be entirely wrong to think that Fichertechnik is just a toy for kids. It’s also perfect for prototyping the control system of robots. [davidatfsg]’s recent entry in the Hackaday Prize, Delta Robot, shows how complex robotics can be implemented without the hardship of having to drill, cut, bolt together or weld components. The added bonus is that the machine can be completely disassembled non-destructively and rebuilt with a new and better design with little or no waste.
The project uses inverse kinematics running on an Arduino Mega to pick coloured objects off a moving conveyor belt and drop them in their respective bins. There’s also also an optical encoder for regulating the speed of the conveyor and a laser light beam for sensing that the object on the conveyor has reached the correct position to be picked.
Not every component is ‘off the shelf’. [davidatfsg] 3D printed a simple nozzle for the actual ‘pick’ and the vacuum required was generated by the clever use of a pair of pneumatic cylinders and solenoid operated air valves.
Thinkpads are great, especially the old ones. You find a T420, and you can have a battery hanging off the back, a battery in the optical drive bay, and for some old Thinkpads, there’s a gigantic ‘slice’ battery that doubles the thickness of your laptop. Here’s the most batteries in a Thinkpad ever, with the requisite reddit post. It’s 27 cells, with an all-up capacity of 212 Watt-hours. There are two interesting takeaways from the discussion here. First, this may, technically, be allowed on a commercial flight. The FAA limit is 100 Watt-hours per battery, and the Ultrabay is a second battery. You’re allowed two additional, removable batteries on a carry on, and this is removable and reconfigurable into some form that the TSA should allow it on a plane. Of course no TSA agent is going to allow this on a plane so that really doesn’t matter. Secondly, the creator of this Frankenpad had an argument if Hatsune Miku is anime or not. Because, yeah, of course the guy with a Thinkpad covered in Monster energy drink stickers and two dozen batteries glued on is going to have an opinion of Miku being anime or not. That’s just the way the world works.
Prices for vintage computers are now absurd. The best example I can call upon is expansion cards for the Macintosh SE/30, and for this computer you have a few choice cards that have historically commanded a few hundred dollars on eBay. The Micron XCEED Color 30 Video Card, particularly, is a special bit of computer paraphernalia that allows for grayscale on the internal monitor. One of these just sold for two grand. That’s not all, either: a CPU accelerator just sold for $1200. These prices are double what they were just a few years ago. We’re getting to the point where a project to reverse engineer and produce clones of these special cards may make financial sense.
The biggest news in consumer electronics this week is the Playdate. It’s a pocket game console that has a crank. Does the crank do anything? No, except that it has a rotary encoder, so this can nominally be used for games. It will cost $150, and there are zero details on the hardware other than the industrial design was done by Teenage Engineering. There’s WiFi, and games will be delivered wireless on a weekly basis. A hundred thousand people are on the wait list to buy this.
If you want a pick and place in your garage workshop, there aren’t many options. There’s a Neoden for about ten grand, but nothing cheaper or smaller. The Boarditto is a two thousand dollar pick and place machine that fits comfortably on your desk. It has automatic tape feeders, a vision system, and for the most part it looks like what you’d expect a small, desktop pick and place machine to be. That’s all the information for now, with the pre-order units shipping in December 2019.
[Paco]’s project involves combining a robotic arm with computer vision tools in order to allow it to pick and place small objects – in this example, toys. The robot arm is of a gantry type, built on an aluminium frame with 3D printed components. The computer vision side of things is handled by a Raspberry Pi, fitted with the standard camera and running OpenCV software for object recognition. This then passes commands to an Arduino which runs the stepper motors controlling the arm.
[Paco] notes that the hardest part of the build was learning how to generate real-world coordinates from a single camera feed in OpenCV. With that mastered, the rest of the dominoes began to fall. With trigonometry and kinematics knowledge in hand, the robot has grown capable of reliably picking and placing small objects across its range of motion. Future work aims to improve the robot’s abilities to rotate and otherwise manipulate its end effector for more versatility.
There’s something mesmerizing about delta robots. Whether they are used at a stately pace for a 3D-printer or going so fast you can barely see them move in a pick and place machine, the way that three rotary actuators can work together to produce motion in three axes is always a treat to watch. Especially with a delta robot as small as this one.
[KarelK16] says this is one of those “just because I can” projects with no real application. And he appears to have been working on it for a while; the video below is from eight years ago. Regardless, the post is new, and it’s pretty interesting stuff. The tiny ball joints used in the arms are made from jewelry parts; small copper crank arms connect the three upper arms to micro-servos. The manipulator [KarelK16] attached is very clever, too – rather than load down the end of the arms with something heavy, a fourth servo opens an closes a flexible plastic grasper through a Bowden cable. It’s surprisingly nimble, and grasps small objects firmly.
There are certainly bigger deltas – much bigger – and more useful ones, too, but we really like this build. And who knows – perhaps model robotics will join model railroading as a hobby someday. If it does, [KarelK16]’s diminutive delta might be the shape of things to come.
Would you believe that you can take a cheap 3D printer and easily convert it into a full function pick and place machine to help assemble your PCBs? No? Well good, because you can’t. A real pick and place needs all kinds of sensors and logic to identify parts, rotate them, make sure everything is aligned, etc, etc. There’s no way you could just bolt all that onto a cheap 3D printer, and let’s not even talk about the lack of closed loop control.
But if you have a very specific use case, namely a PCB that only has a relatively large single part that doesn’t need to be rotated, [Connor Nishijima] might have a solution for you. He bought a $150 USD Monoprice Mini, and with the addition of a few printed parts, was able to build a machine that drastically cuts down the time it takes for him to build his LED boards. Best of all the modification doesn’t involve any permanent changes to the printer, he can just pop off the vacuum attachment when he wants to print something.
Beyond the 3D printed parts (which were made on the printer itself), the only thing you need to make the modification is the vacuum pump. [Connor] is using a hot air station that includes a vacuum pump for picking up SMD components, but he mentions that you’d probably better off just modifying an aquarium pump and using that. A printed holder snaps over the cooling fan of the Monoprice Mini to hold the vacuum pickup tool, and another printed piece holds the strip of LEDs and the PCB. It’s worth noting that the machine has no ability to control the vacuum pump, and doesn’t need to. The pickup tool is so weak that when the LED lands in the solder paste it sticks to the board well enough that the tool can’t lift it back off.
The real genius in this build comes from the manually written G-Code. You load it from the printer’s built in menu system as if it was a normal 3D print, and it instructs the printer to move the vacuum tool over the line of LEDs, pick one up, and drop it in place on the PCB. It then uses a small peg built into the vacuum tool holder to advance the line of LEDs before starting the cycle all over again. Incredibly, it does this whole complex dance 20 times for each PCB without ever having any kind of feedback or alignment check. It only works because [Connor] was willing to go through the trial and error of getting the calibration and G-Code down as close to perfect as can be expected for such a cheap machine.
Surface mount is where the action is in the world of DIY PCBs, and deservedly so. SMDs are so much smaller than through-hole components, and fewer holes to drill make surface-mount PCBs easier to manufacture. Reflow soldering is even a snap now thanks to DIY ovens and solder stencils you can get when you order your boards.
So what’s the point of adding another stencil to the surface-mount process? These component placement stencils are [James Bowman]’s solution for speeding up assembly of boards in production runs too small to justify a pick and place robot. [James] finds that placing small components like discrete resistors and caps easy, but struggles with the placement of the larger components, like QFN packaged microcontrollers. Getting such packages lined up exactly is hard when the leads are underneath, and he found repositioning led to smeared solder paste. His acrylic stencils, which are laser-cut from SVGs derived directly from the Eagle files with a script he provides, sandwich the prepped board and let him just drop the big packages into their holes. The acrylic pops off after placement, leaving the components stuck to the solder paste and ready for their trip to the Easy Bake.
[James] claims it really speeds up hand placement in his biggish runs, and it’s a whole lot cheaper than a dedicated robot. But as slick as we think this idea is, a DIY pick and place is still really sweet.