New Part Day: Battery-Less NFC E-Paper Display

Waveshare, known for e-ink components aimed at hobbyists among other cool parts, has recently released a very interesting addition to their product line. This is an enclosed e-ink display which gets updated over a wireless NFC connection. By that description, nothing head-turning, but the kicker is that there is no battery inside the device at all, as it harvests the energy needed from the wireless communication itself.

Just like wireless induction charging in certain smartphones, the communication waves involved in NFC can generate a small current when passing through a coil, located on this device’s PCB. Since microcontrollers and e-ink displays consume a very small amount of current compared to other components such as a backlit LCD or OLED display, this harvested passive energy is enough to allow the display to update. And because e-paper requires no power at all to retain its image, once the connection is ended, no further battery backup is needed.

The innovation here doesn’t come from Waveshare however, as in 2013 Intel had already demoed a very similar device to promising results. There’s some more details about the project, but it never left the proof of concept stage despite being awarded two best paper awards. We wonder why it hadn’t been made into a commercial product for 5 years, but we’re glad it’s finally here for us to tinker with it.

E-paper is notorious for having very low refresh rates when compared to more conventional screens, much more so when driven in this method, but there are ways to speed them up a bit. Nevertheless, even when used as designed, they’re perfectly suited for being used in clocks which are easy on the eyes without a glaring backlight.

[Thanks Steveww for the tip!]

30 thoughts on “New Part Day: Battery-Less NFC E-Paper Display

      1. How long are technology patents like this, 15 or 16 years? Must’ve expired by now surely. I’m pretty certain of it, but if I’m wrong I bet there’s a year or two left tops.

        I think the real cause of the price being fairly high is lack of competition, caused by lack of demand. There just isn’t enough demand for the displays. They didn’t get significantly faster, as was promised. Or higher in res. Colour depth is fine with 16 greys which will do for still images. Actual colour never worked out either, nor did better refresh rates.

        The technology didn’t advance, which has limited it’s use to E-books and that’s about it. Those bistable LCDs are better for almost every other purpose. Ebooks died out because tablets took over their uses and did enough extra things to make up for their few shortfalls, and medium-size colour LCD tablets are ever more practical now there’s USB and Wifi all over the joint.

        The only think E-Ink is best at, is a small market, and it isn’t worth investing in a factory to compete over that. Wouldn’t be worth starting the first factory if we’d known, but at the time people saw advancements that never came.

        China doesn’t believe in patents, they still haven’t bothered bootlegging E-ink, preferring the razor-thin margins of LCD. There lies a lesson.

    1. Yeah I’m also waiting for cheap e-ink screens. Dirt cheap – cheap enough to buy them by the dozens on a “might come in handy” whim. The biggest mistake with e-ink devices has been making them too smart, too feature rich, too processing powerful. Most people already have a quite capable computer in the pocket and do not long for yet another expensive device to charge frequently, more app stores and accounts to set up, etcetera. No, produce e-ink items that are stupid and cheap like paper, only instead of laserprinting static text/image on it we cast to from a smartphone.

      1. “The biggest mistake with e-ink devices has been making them too smart, too feature rich, too processing powerful.”

        I agree. The technology lends itself best to low-feature devices where the display of infrequently-changing information is really useful.

        The part in the story doesn’t seem to add much compared to USB, or a cradle with UART pins, because you still have to get the device close to the mothership, by hand, at the time you want the update to happen, and then initiate the transfer. It is just like wireless charging for a cell phone; you still have to have the charger plugged in on the desk. You save 1 gesture, but one that only happens at the same time as other gestures you still have to do, like interacting with the phone to cause it to display a charge level.

        I think people are mostly considering the use cases wrong; other than things like ebook readers, it doesn’t seem like a very good fit for a monitor-type screen. A better fit might simply be a label; single line text type screens for devices that don’t normally have a display. Then you can display a status or error message easily; you could have a “low battery” indicator that doesn’t itself use up the battery! Or to display an error message even after the device is turned off, until it is cleared! “Oh, that’s why it didn’t work, the feed was jammed and I didn’t notice! I’ll try again.” These types of uses add a small amount of value, but for people complaining they’re too expensive I’d also point out it is premium-type value, not necessary features. I can easily imagine a shelf of parts bins with labels that can be edited by an app. Something a rich person with OCD would really enjoy using, but is going to generate more ridicule than interest from the average person.

  1. Huh. Clever. It’s neat how the parts for this have existed for years and nobody has put them together yet. I’ve worked with all of these components and I have been very interested in NFC harvesting applications and never thought of putting them together in this way.

    1. This is actually used in stores to update the price tags along the shelves. I saw it in use first time around 10 years ago, but then it was powered by a tiny solar cell. Nowadays they removed the solar cell and the tags are updated by NFC.

      1. It’s certainly possible, but updating 100 of such tags daily will get tedious real fast.
        As far as I know it’s much more common to update these tags via IR transmitters scattered around the ceiling.
        I’m not sure about the rest of the system. The IR beamers could be integrated in smoke detectors or other stuff and communicate further over WiFi or other.

        1. Somebody going around checking and stocking the shelves anyways. All they have to do is wave the hand terminal up to the tag.

          It’s main benefit is to save the effort and paper for printing at the back office. Updating the tags over the air isn’t really all that useful. In one store they displayed the number of items on the shelf, which would update when they sold one – but what for? Anyone can count should they wish to.

          1. I think you have vastly overestimated how often individual facings get touched. If it’s not a hot item that sells out constantly I think it’s very likely nobody pays attention for a month or two. If there’s no printed out and mailed-in deal tag nobody’s going to bother to check the thing for a quiet price change.

          2. For most stores, the prices are only updated when the shelves are re-stocked, because the existing items cost the store X amount so they want to sell them all at X*n price.

            Of course you could keep tally of your existing inventory and change the price dynamically, but your customers would catch the price going up and down during the day, and they’d start to avoid you because that looks like cheating.

      2. There have even been an open design or two, however they look like they never got finished.

        https://twitter.com/FauthNiklas/status/1045267775219027969

        https://github.com/NiklasFauth/nfc-epaper/blob/master/hw/rf430_epaper.pdf

        Shame, I’d love to see nametag sized versions made nice and cheap to play with.
        I found out about the waveshare one a few months back when researching for how to do this exact thing.
        Turns out lots of people had the same idea. :)

    1. More important then wireless: It’s battery less.

      It’s almost trivial to put a thing like this on a flat sheet of glass an pour some epoxy, silicone or other suitable sealant over the back.

      A power cord can be easily sealed also, but if access for batteries is needed it becomes more difficult. (Although cutting though some silicone every 5 years if rechargeable batteries have worn out is also not such a biggie.

      1. Have to agree – the biggest advantage of these things is not needing any sort of power supply, batteries or wired.

        Great for weatherproofing, but probably not so good for re-usability – dang, need to update the firmware. Oh well, better reorder that batch then!

        1. Why would you need to update the firmware? It’s a screen. It takes stuff off the NFC and puts it on the screen. Surely they can get that right first time! If they really needed to, they could have an NFC bootloader and that could update it’s firmware over NFC, but what would that firmware even be for?

          If you keep it simple enough it’s not going to need anything more. It’s kept simple by it’s form.

          1. Supporting new formats? (Webp is going to take off one day or an other)
            Opening the firmware? (one can dream)
            Adding functions? (Updating only part of the screen? Making an infinite “daily check” with each people having it’s own line, maybe stats? Changing the protocol if a new version of hardware compatible NFC comes out? )

          2. TheDatkTiger: Those e-ink labels don’t support any format at all, you send them raw pixels. Supporting new graphics formats is the job of the programming device. As for new NFC hardware or protocol, things are usually kept backwards-compatible in the hardware world. If it works today, just make sure your new version of programmer can still talk to the older versions of screens.

  2. This isn’t really a part, is it? Looks more like a product. Basically another smart card but with a useful semi-permanent display. Possibly with useful security features through only letting the onboard CPU access the display. Maybe 2-factor stuff or something.

  3. Seems the design of the Intel one is actually open, the pcb design files and firmware can still be found on github.com/wisp/nfc-wisp-hw and /nfc-wisp-fw but their wiki was hosted on wikispaces and is very much dead

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.