Computer memory has taken on many forms over the years, from mercury-based delay-line tubes to handwoven magnetic core. These days, volatile storage using semiconductors has become ubiquitous with computing, but what if there was a better way? [Michael Kohn] has been working on a new standard for computer memory that uses glow in the dark stickers.
Clearly we jest, however we’re still mighty impressed by the demonstration. Eight delightful star-shaped phosphorescent stickers represent eight bits of memory, totaling one byte. The glow in the dark material is stuck to the inside of short cylinders, each of which contains a white LED and a phototransistor. The memory array is wired up to an iceFUN FPGA board, which is then connected via level shifters to a Western Design Center MENSCH single board computer.
To write a ‘1’ to memory is as simple as writing to the corresponding memory address using 6502/65C816 assembly language. Using the STA command will illuminate the white LED at that memory address, in turn irradiating the glow in the dark sticker and ‘saving’ the state. Conversely, LDA at the same address will read from the phototransistor, which picks up the glow (or lack thereof) emitted from the sticker.
A refresh cycle is needed to sustain the 0s and 1s across the memory array as the phosphorescence fades, not dissimilar to modern DRAM requiring frequent recharging to maintain memory contents. The entire setup is a tangible demonstration of the fundamentals of volatile computer memory, and would make for a fun beginner project. [Michael] has more details on his website and GitHub page.
While the FPGA board has its own little set of blinkenlights, an 8-bit RGB LED array would make this project even brighter.
Not as cool as this.
Sadly, some of the links are broken.
https://hackaday.com/2014/08/07/dots-and-dashes-on-a-roll/
There was another example of memory implemented using a moving strip of luminescent tape, here on HaD, but I haven’t been able to find it.
Courtesy of The Internet Archive, you’re in luck: https://janderogee.com/projects/morse/morse.htm
HaD Editors: Please add “glow-in-the-dark” tag to this article.
Why not use a UV LED?
That was fun. I didn’t hear mention of clearing bits in the video though. While writing a one seems pretty straightforward, won’t erasing it be a time intensive operation that can’t exceed the natural decay rate of the material? Use of a minimum glow threshold to represent a zero could speed things up but that would increase the refresh rate needed to maintain the enabled bits and in turn reduce the IO rate. Still pretty fun looking.
I think one could use two glow in the dark stars, two leds, two photodiodes and. Differential amp. Writing a 1 would energize the one star, writing a zero would energize the other. The decay still slows it down but this method would be faster.
This. The diff amp means that the retention time is limited mainly by the noise level in the system. But there is still a hazard, where if you write a number of times in either state, a single write to the other state may not be enough to flip the read value. This is because the writes get integrated by the phosphor compound.
maybe we can use some heat to speed up the discharging? Not that heating plastics is the fastest thing to do either…
Can a lower wavelength of light be used to encourage the pigment to emit and discharge without having enough energy to reenergize it? I swear I dimly remember something like that being possible.
“dimly” remember? I see what you did there =P
I remember it much less dimly. There was an article on HaD that was similar to this one, where a glow-in-the-dark drum or tape loop was used as a storage device, and someone in the comments section mentioned that phosphorescent compounds can be “discharged” by specific wavelengths of light. Unfortunately I have not been able to find that article again.
Hm. Found the article: https://hackaday.com/2021/05/24/persistent-displays-with-uv-light/
However, the comment wasn’t all that helpful:
“sjm4306 says:
May 25, 2021 at 7:54 am
From what I’ve read it should be possible to “quench” the written surface on the backside of the conveyor belt using IR leds to effectively erase what was written and leave a blank surface”
So HEY, SJM4306: where did you read this?
You can stimulate forbidden transition by using red LED. so UV/blue for write and red for erase.
Have you done this, or seen it done? References? Seems like it would be useful.
Sure, done it on the kitchen counter years ago. They made IR test cards that you’d have to charge up before using. Sold them at RadioShack to verify that your remote was working.
Wow! I remember those, but had no idea how they worked. This suggests a number of interesting projects!
If you build this on a whole wall – and so that, the glow can be seen would be impressing to view while the computer runs.
The Williams and Selectron tubes use similar approaches – would like to see them in action.
Practical Williams tubes were not visible, since ambient light would scramble them. They were sometimes “mirrored” by visible CRT’s though so the memory contents could be read by a human operator.
You know, bistable storage CRTs used a flood gun to refresh the charge on what is essentially an array of capacitors on their faces, but these were of course limited in size. BUT, if you can read it and write it, for example by having a row of photodiodes as well as the row of UV LEDs, you could re-energize any bits that are above a certain threshold, which would create a persistent memory array. Not unlike dynamic RAM, but with the added benefit of being directly viewable.
You could also use a red light to write a zero quicker. A red laser will discharge most photo luminescent materials.
How about using a laser and drawing a pattern on the sticker….
I liked it the most when you shared that we should choose a material that picks up the glow emitted from the sticker. My friend wants photoluminescent exit signs for their building. I should advise him to go for it to ensure it’s visibility.