• A Review of the OSH Park Jelly Beans

    12/09/2016 at 14:18 5 comments

    I received my latest OSH park PCBs yesterday - some TritiLED boards without integrated battery holders. Inside the envelope was also a package of OSH Park jelly beans:

    I've previously compared the board quality and level of service at OSH Park to that of a few China-based services I've tried (OSH Park wins), so this time, I thought I'd focus on the confectionery aspects of the PCB service.

    True to the made-in-USA-quality branding of OSH Park, the beans themselves are genuine Jelly Belly's: well-regarded beans, indeed. The back of the package displays a prominent "Made in the USA" label:

    Even though the beans were sealed in the plastic, I took the precaution of wiping down the package to remove any contaminants that may have transferred from the PCBs. I'd be more concerned if these were lead-solder HASL boards, but still - I'm a stickler for no food or drink in the lab.

    When I first saw the package in the poor lighting of my garage, I thought the beans were black. It didn't take long to realize that they must instead be purple! Like the PCBs themselves, these beans take on such a rich hue that this mistake is easily made. In better light, they appear quite handsome: deep purple in color, with the classic white "Jelly Belly" logo emblazoned on the side, mirroring the PCB's soldermask and silk-screen. Bravo, OSH Park!

    Upon opening the bag, I detected a sweet, fruity boquet of berries and plums: they could be Island Punch or Wild Blackberry - or knowing OSH Park, a custom purple mix signifying that they're a brand apart. A small bite revealed the expected firm outer skin surrounding a crystallized sweet flesh and a supple inner core. I found the bean to be big, bright, and complex: definitely fruit-forward in character. A second taste showed a not unpleasant jammy flavor - this is an opulent candy. I'm not an expert at jelly bean tasting, but I would guess this is a fine example of the Wild Blackberry style - probably from California.

    I'd like to be able to compare these to the candies sent with boards from other fabs - but alas, no other fab has ever sent any. Therefore, I declare these to be the absolute best candy of any batch prototype PCB fabricator (by default).

    Oh, and the PCBs were excellent as usual. but that's not really news.

    FULL DISCLOSURE: I received a package of jelly beans from OSH Park in a recent order. However, they asked for nothing in return.

  • Diode-capacitor RAM

    11/14/2016 at 00:51 3 comments

    This isn't an original idea - I have found similar notions dating back to at least 1952!


    Anyway, the idea is to make a DRAM memory cell with just a capacitor and two diodes. I played with one back in 2009, and am documenting it here briefly since there seems to be some interest in these things. I never made it fully operational with decoded address lines, but was able to verify storage of a few 4-bit words with transistor sense amps and LEDs. Here's the unit (glad I dated it!!):

    Here is the schematic:

    There are two diodes and a capacitor in each cell. The diodes connect to two individual lines per column - a "write 1" line and and a "write 0" line. To write a 1 to the cell, the appropriate row line is grounded, then a positive voltage is applied to the appropriate "write 1" column line. Current flows through the one diode to charge the cap. To write a zero, the row line is grounded, and so is the appropriate "write 0" column line. Current flows through the other diode to discharge the cap. Reading also uses the "write 0" line - any time the row line is grounded, current flows from the cap to the sense amplifiers.

    Like any DRAM, this needs to be periodically refreshed. My original intent was to use this idea for the RAM of a diode-only computer, but I could not get diode-only sense amps sensitive enough, so I shelved the idea. In the intervening years, I figured out the diode amplification problem, but never got back to the memory idea.

    I was always a little worried about somehow reverse-biasing the electrolytic caps in this circuit, but I'm pretty sure I convinced myself that it was OK. It has been a while.

    I was a little surprised I still had this thing. I save way too much crap.

  • Custom LEDs!

    11/09/2016 at 03:06 4 comments

    I had an unusually pleasant interaction with an LED manufacturer lately - I had some custom LEDs made!

    I have been using some Chanzon LEDs in my projects lately (find them on Amazon, Ebay, or AliExpress). They have two different chip sizes (30x30 mil and 45x45 mil) they package into various cases. Here are images of their "1W" and "3W" red LEDs, which consist of the two different chips in the same package:

    I didn't quite figure out how the 3W chip has only 2.25x the area of the 1W chip - anyway, little chip = less power, big chip = more power.

    Now, I'm working on a project where I need some really bright, really short flashes of red, green, and blue light. Like 100A of current for sub-microsecond pulses into a handful of LED chips. Chanzon makes a 10W LED in a larger package with a 3x3 array of the smaller 1W chips, but what I really would like is an array of the larger chips. After seeing how manual the LED assembly process actually is in the Sparkfun article (a really good read):


    I figured they might be able to make some custom LEDs - just plop the larger chips where you usually put the small ones. A week of back-and-forth messaging on AliExpress, and we had it all sorted out. The custom parts arrived today:

    The custom part is on the right with the larger chips; the normal part is shown on the left. Of course, if you actually ran these LEDs at 30W continuously, they'd easily overheat and fail. Driving them in pulsed mode with very low duty cycle will allow some extremely bright flashes.

    I'm still amazed I was able to have these made. The only other custom parts I've used have been inductors or transformers I've wound myself. Having a manufacturer accommodate a small (10x of three colors) custom order like this is unbelievable. And just plain awesome.

    EDIT 20161109

    If you look very carefully at the top left LED in the array above, you can see it's missing a bond wire. Here's a close-up (same orientation):

    I just put all 30 of them under the microscope...and this is the only one with the defect (I was just lucky to pull it out for pictures yesterday). Electrically, it still tests out OK, because one bond wire is present. I just wouldn't trust this particular unit with high-current pulses. Still a heck of a deal.