Comment by rtkwe | Hacker News Reader

Comment by rtkwe | Hacker News Reader

Parent Full thread

Not by your standard NFC, those cap out at ~50mW [0] while eInk screens consume 750-1800mW during an active update. [1] That number is probably out of date but it's tough to find good numbers quickly, maybe if you accept EXTREMELY slow updates you could do that but the display drivers probably consume a fair amount themselves. [2] It's low enough you could pair a Qi charger with the updater to power the screen temporarily. Most solutions that use eInk like adjustable price labels just put a low self discharge battery in them to allow standard devices to do the updating and those last several years before they need to be replaced depending on update frequency.

[0] https://www.nucurrent.com/nfc-power-harvesting-explained/#:~...

[1] https://wiki.mobileread.com/wiki/E_Ink_display#:~:text=Activ...

[2] People's/site's answer defaults to the "eink uses no power to maintain an image" even when the question asked is specifically about power use during updates.

I'm confused (and an electricity layman, so please excuse), but why would that power requirement of NFC matter in conjunction with the eInk power usage? Why can't the NFC reader be isolated from the display's power usage?

Because the person I was replying to specifically asked if you could pair the two and power/update a stand alone display with just NFC delivered power?

The devices for updatable price labels based on eInk already exists and I mention it in my post, it's powered by a low self discharge battery. Those IIRC use a special radio fob that connects to your phone instead of using NFC but you could probably do a largely drop in replacement for the radio circuit with NFC but that still doesn't solve the issue of updating the eInk screen. That's an order of magnitude more power required.

YEs, but their description of what they were imagining made clear they were using the word 'powered' to mean 'controlled by', not in the literal sense of delivering electrical power. I blame advertising for this, 'powered by [software product]' has become ubiquitous and exploits the relatively low literacy level of the public.

I really think they actually meant literally powered by here. Being on HN I make the presumption that commenters are a bit more savvy than the average Joe Public.

https://news.ycombinator.com/item?id=38002621

"Hang frame on the wall, choose the image in the app, bring phone close to frame, boom, the frame has new image."

It's extremely obvious the person you replied to was talking about controlling the tablet, not supplying electricity to it. Their subsequent remarks are beside the point, this is what led to the confusion in this thread.

Beside the point? It's pretty on point if you're trying to say they weren't talking about powering the device too... Also there are only two people here confused about GP's meaning. Can it be controlled by NFC is a trivial question.

The original poster in the thread was suggesting that if power usage were low enough, it could be powered by the power provided by the NFC in the phone; where the phone provides power for an NFC circuit that is entirely unpowered. If you could power an eink display that way, you could have a completely unpowered frame, which you could set via NFC.

So yes, of course NFC can be isolated from the displays power usage, the originator of this thread was just suggesting it would be neat if you could power it via NFC.

But NFC doesn't provide nearly enough power, nor likely enough bandwidth to do that in a reasonable amount of time.

In theory you could maybe charge a supercapacitor over time with the power from NFC and then perform the update but you're looking at at least 15x (50mW vs 750mW average update draw) the update time standing there holding the phone to the display while you trickle power over that connection. In reality with losses and extra draws to manage the screen you might not break net positive on the energy from the phone or you'd be looking at a 30-50x multiple instead of /just/ 15x.

GP said: "I wonder if these thingies can be powered by NFC sensor."

Thanks for the informative answer. It just occured to me that modern phones like iPhone or Motorola/Samsung/etc phones often have Qi charging capabilities for powering accessories, so powering a display wirelessly shouldn't be a problem.

Qi wireless charging isn't by default bidirectional but yes more things are moving that way so you could, on supporting devices, in theory have them powered by a Qi pad only for the duration of the refresh.

The only issues I can think of there are I'm not sure if the Qi process can start up without power on the receiving end, the source device isn't always broadcasting power afaik so I don't know what the trigger is and if it requires some communication with the receiver device. The last is the NFC has a pitiful data rate of ~424 kbits/s but if we're powering it through Qi we can use more power hungry transfer methods to speed it up.

Yeah the "use your phone to charge your headphones" type Qi charging is what you want to use here. Top up your sign and do an nfc tap to set it.

Or if it's outdoors, add a small solar panel.

Can the Qi process bootstrap if the receiving device has no power to begin with? My limited understanding of the process is that there's a communication step between the sending device (a phone in this case) and the recipient before the source begins transmitting.

edit: A quick skim of the massive Qi specs makes it seem like maybe it can boot strap but I'll be honest I don't have the time to really dig in super deep into it. Bing Chat also seems to think it can but its sourcing is a bit spotty and the link it provided for that isn't actually saying anything about unpowered bootstrapping on the Power Receiver side.

Reply on news.ycombinator.com